爱吱声

标题: 世界杯预热，墨西哥对南非预测 [打印本页]

作者: xiejin77 时间: 7 天前
标题: 世界杯预热，墨西哥对南非预测
# 2026 世界杯预测报告：Mexico vs South Africa

- **预测时间**: 2026-06-10 13:48:48
- **Run ID**: 20260610_134848
- **比赛**: 第 1 场 · 小组赛 · A组
- **开球时间**: 2026-06-12 03:00 北京时间（原始: 2026-06-11 15:00 ET）
- **场馆**: Estadio Azteca · Mexico City
- **贝叶斯数据更新时间**: 2026-06-10T13:48:33

## 关键判断

- 模型当前最倾向于“主队胜”，概率为 55.0%，领先第二选项 32.0 个百分点。
- 综合置信度为低；模型分歧为 27.7%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 Polymarket，“主队胜”的模型概率低于市场 12.5 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-10T13:48:33。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## 综合概率

- Mexico 胜: 55.0%
- 平局: 21.9%
- South Africa 胜: 23.0%
- 综合置信度: 低
- 模型分歧: 27.7%

## 贝叶斯分析

- 先验/贝叶斯概率: Mexico 68.9% / 平局 15.5% / South Africa 15.6%
- Elo: 1682 vs 1440
- audited recent form: Weighted recent form gap: home=0.75 vs away=0.17（权重 6.4%）
- lineup: Mexico 4-3-3 lineup confirmed from expert; key starters: Edson Alvarez, Orbelin Pineda, Alvaro Fidalgo（权重 3.8%）
- lineup: South Africa 4-2-3-1 lineup confirmed from expert; key starters: Sphephelo Sithole, Teboho Mokoena, Oswin Appollis（权重 3.8%）
- simulation factors: Weather condition: drizzle（权重 2.5%）
- simulation factors: Pitch quality: heavy（权重 4.0%）

## 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模证据[form]: Weighted recent form gap: home=0.75 vs away=0.17；置信度 0.55；最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。
- 入模证据[lineup]: Mexico 4-3-3 lineup confirmed from expert; key starters: Edson Alvarez, Orbelin Pineda, Alvaro Fidalgo；置信度 0.62；首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。
- 入模证据[lineup]: South Africa 4-2-3-1 lineup confirmed from expert; key starters: Sphephelo Sithole, Teboho Mokoena, Oswin Appollis；置信度 0.62；首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。
- 入模证据[weather]: Weather condition: drizzle；置信度 0.42；小雨/毛毛雨只作为很弱的低比分信号，不再自动推断为沉重场地。
- 入模证据[pitch]: Pitch quality: heavy；置信度 0.5；沉重或较差场地会压低进攻流畅度，因此只轻微提高平局倾向。
- 原始伤停线索 10 条仅保存审计，未匹配/低可信内容不入模。

## 蒙特卡洛模拟

- 模拟次数: 1000
- 胜平负: Mexico 41.2% / 平局 28.4% / South Africa 30.4%
- 平均进球: 1.22 - 1.02
- 90 分钟结果: Mexico 41.2% / 平局 28.4% / South Africa 30.4%

## 模拟参数与外部因素

- 随机种子: 806584068
- 模拟阶段: 小组赛，90 分钟允许平局
- 天气: drizzle; 温度 23.2°C; 风速 5.5 km/h; 湿度 57%; 来源 open-meteo
- 场地: 草皮 unknown，质量偏重
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 场地偏重或质量较差：传球和盘带成功率下降。

## 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## 阵容来源与匹配状态

- 主队 Mexico: 4-3-3 / expert / predicted / confidence 0.65
  - 首发: Guillermo Ochoa, Jorge Sanchez, César Montes, Johan Vásquez, Jesus Gallardo, Orbelin Pineda, Edson Alvarez, Alvaro Fidalgo, Alexis Vega, Raul Jimenez, Julian Quiñones
- 客队 South Africa: 4-2-3-1 / expert / predicted / confidence 0.74
  - 首发: Ronwen Williams, Khuliso Mudau, Nkosinathi Sibisi, Khulumani Ndamane, Aubrey Modiba, Sphephelo Sithole, Teboho Mokoena, Oswin Appollis, Themba Zwane, Relebohile Mofokeng, Lyle Foster

## 综合诊断

- Bayesian is higher on home win by 27.7pp. Likely evidence-layer effect (5 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 12.9pp. Likely generative simulation effect from squad duels, tempo, or factors (1 factor effects).
- Monte Carlo is higher on away win by 14.8pp. Likely generative simulation effect from squad duels, tempo, or factors (1 factor effects).

## Polymarket 概率对照

- 市场问题: Bookmaker h2h odds: Mexico vs South Africa
- 市场更新时间: 2026-06-10T13:48:20
- 主队胜: 模型 55.0% / Polymarket 67.5% / 差值 -12.5pp
- 平局: 模型 21.9% / Polymarket 21.0% / 差值 +1.0pp
- 客队胜: 模型 23.0% / Polymarket 11.5% / 差值 +11.5pp

## 引擎完整输出

# Match Prediction Report
## World Cup 2026 Group Stage: Mexico vs South Africa

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Mexico Win | 68.9% | 41.2% | 55.0% |
| Draw | 15.5% | 28.4% | 21.9% |
| South Africa Win | 15.6% | 30.4% | 23.0% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Mexico Elo: 1682
- South Africa Elo: 1440
- P0: Mexico 68.5% / Draw 15.5% / South Africa 16.0%

### Likelihood Updates
- [home] Weighted recent form gap: home=0.75 vs away=0.17 (±6%) — audited recent form
- [home] Mexico 4-3-3 lineup confirmed from expert; key starters: Edson Alvarez, Orbelin Pineda, Alvaro Fidalgo (±4%) — lineup
- [away] South Africa 4-2-3-1 lineup confirmed from expert; key starters: Sphephelo Sithole, Teboho Mokoena, Oswin Appollis (±4%) — lineup
- [draw] Weather condition: drizzle (±2%) — simulation factors
- [draw] Pitch quality: heavy (±4%) — simulation factors

### P1 (Final Bayesian)
- Mexico: 68.9%  |  Draw: 15.5%  |  South Africa: 15.6%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 1,000
- Avg Goals: Mexico 1.22 / South Africa 1.02 / Total 2.25
- Avg Shots: Mexico 17.2 / South Africa 15.2
- Conversion: Mexico 7.1% / South Africa 6.8%
- Avg Cards: 3.7

### Top Scorelines
- Mexico 1-1 South Africa: 13.7%
- Mexico 1-0 South Africa: 13.5%
- Mexico 0-1 South Africa: 10.7%
- Mexico 0-0 South Africa: 10.2%
- Mexico 2-1 South Africa: 8.2%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 27.7%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Bayesian is higher on home win by 27.7pp. Likely evidence-layer effect (5 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 12.9pp. Likely generative simulation effect from squad duels, tempo, or factors (1 factor effects).
- Monte Carlo is higher on away win by 14.8pp. Likely generative simulation effect from squad duels, tempo, or factors (1 factor effects).

### Consensus Estimate
- Mexico: 55.0%  |  Draw: 21.9%  |  South Africa: 23.0%

## 免责声明

本报告仅供数据分析参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: 老票 时间: 7 天前

请理性看待体育赛事结果，远离非法赌博。

在本坛热烈欢迎下爱元赌注，越大越好！

作者: 方恨少 时间: 7 天前
墨西哥大概率获胜，唯一要考虑的是由于作为东道主首战心理压力过大导致发挥失常。

墨西哥届届16郎，好不容易当回东道主，可以借主场之利打破魔咒，结果扩军了，小组出线就算再赢一场。。。还是16郎

作者: 方恨少 时间: 7 天前

老票发表于 2026-6-10 13:54
在本坛热烈欢迎下爱元赌注，越大越好！

需要一个勇于自我牺牲的庄家

作者: 老票 时间: 7 天前

方恨少发表于 2026-6-10 14:57
需要一个勇于自我牺牲的庄家

赔率设计好，庄家都是两面通吃的

作者: xiejin77 时间: 5 天前
# 2026 世界杯预测报告：Korea Republic vs Czechia

- **预测时间**: 2026-06-12 07:58:32
- **Run ID**: 20260612_075832
- **比赛信息**: 第 2 场 · A组
- **开球时间**: 2026-06-13 10:00 北京时间（原始：2026-06-12 22:00 ET）
- **场馆**: Estadio Guadalajara · Guadalajara
- **贝叶斯数据更新时间**: 2026-06-11T08:18:36

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 47.2%，领先第二选项 19.3 个百分点。
- 综合置信度为高；模型分歧为 4.2%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 The Odds API，“主队胜”的模型概率高于市场 11.5 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-11T08:18:36。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Korea Republic 胜: 47.2%
- 平局: 24.9%
- Czechia 胜: 27.9%
- 综合置信度: 高
- 方法分歧: 4.2%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Korea Republic 48.5% / 平局 22.8% / Czechia 28.6%
- Elo: 1589 vs 1501
- lineup: South Korea 4-2-3-1 lineup confirmed from manual; key starters: Hwang Inbeom, Bae Junho, Eom Jisung（幅度 6.5%）
- lineup: Czech Republic 3-5-2 lineup confirmed from manual; key starters: Alexandr Sojka, Denis Visinsky, Hugo Sochurek（幅度 6.5%）
- lineup: Czech Republic uses 3-5-2; conservative/extra-defender shape（幅度 3.5%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[formation]: 三中卫/五后卫阵型通常更保守，贝叶斯侧仅轻微提高平局倾向。；置信度 0.78
- 原始伤停线索 9 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 1000
- 90分钟胜平负: Korea Republic 45.8% / 平局 27.0% / Czechia 27.2%
- 平均进球: 1.34 - 0.94
- 模拟引擎: fm-style-chain-simulator (simulate_match -> simulate_attack_chain)
- 攻击链审计: 总计 40962 条，场均 41.0 条
- 射门/转化率: 总射门 32105 次，进球转化率 7.1%
- 稳定性审计: 中；单项 95% 误差上限约 ±3.1 个百分点
- 稳定性提醒: Monte Carlo sample is usable, but close probabilities should be interpreted with the confidence interval.
- 90分钟结果: Korea Republic 45.8% / 平局 27.0% / Czechia 27.2%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 920110718
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 19.3°C; 风速 2.5 km/h; 湿度 94%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 South Korea: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Jo Hyeonwoo, Cho Yumin (Sharjah, Jens Castrop, Kim Minjae, Kim Moonhwan, Bae Junho, Eom Jisung, Cho Guesung, Hwang Heechan, Hwang Inbeom, Oh Hyeongyu
- 客队 Czech Republic: 3-5-2 / manual / confirmed / confidence 0.78
  - 首发: Jindrich Stanek, David Doudera, David Jurasek, David Zima, Alexandr Sojka, Denis Visinsky, Hugo Sochurek, Adam Hlozek, Jan Kuchta, Mojmir Chytil, Patrik Schick

## Diagnosis / 综合诊断

- Bayesian and Monte Carlo agree closely under the shared baseline.

## Market Comparison / 外部市场概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Korea Republic vs Czechia
- 市场更新时间: 2026-06-11T08:18:36
- 主队胜: 模型 47.2% / The Odds API 35.7% / 差值 +11.5pp
- 平局: 模型 24.9% / The Odds API 30.7% / 差值 -5.8pp
- 客队胜: 模型 27.9% / The Odds API 33.6% / 差值 -5.7pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: South Korea vs Czech Republic

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: HIGH
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| South Korea Win | 48.5% | 45.8% | 47.2% |
| Draw | 22.8% | 27.0% | 24.9% |
| Czech Republic Win | 28.6% | 27.2% | 27.9% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- South Korea Elo: 1589
- Czech Republic Elo: 1501
- P0: South Korea 48.6% / Draw 22.8% / Czech Republic 28.6%

### Likelihood Updates
- [home] South Korea 4-2-3-1 lineup confirmed from manual; key starters: Hwang Inbeom, Bae Junho, Eom Jisung (±6%) — lineup
- [away] Czech Republic 3-5-2 lineup confirmed from manual; key starters: Alexandr Sojka, Denis Visinsky, Hugo Sochurek (±6%) — lineup
- [draw] Czech Republic uses 3-5-2; conservative/extra-defender shape (±4%) — lineup

### P1 (Final Bayesian)
- South Korea: 48.5%  |  Draw: 22.8%  |  Czech Republic: 28.6%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 1,000
- Avg Goals: South Korea 1.34 / Czech Republic 0.94 / Total 2.28
- Avg Shots: South Korea 18.1 / Czech Republic 14.0
- Conversion: South Korea 7.4% / Czech Republic 6.7%
- Avg Cards: 3.7

### Top Scorelines
- South Korea 1-0 Czech Republic: 13.1%
- South Korea 1-1 Czech Republic: 12.2%
- South Korea 2-0 Czech Republic: 10.8%
- South Korea 0-0 Czech Republic: 10.3%
- South Korea 0-1 Czech Republic: 10.1%

## Layer 3: Synthesis

**Confidence: HIGH** (max divergence: 4.2%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Bayesian evidence posterior and Monte Carlo simulation share the same baseline; equal weights are used by default.

### Divergence Analysis
- Bayesian and Monte Carlo agree closely under the shared baseline.

### Consensus Estimate
- South Korea: 47.2%  |  Draw: 24.9%  |  Czech Republic: 27.9%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 5 天前

xiejin77 发表于 2026-6-12 08:01
# 2026 世界杯预测报告：Korea Republic vs Czechia

- **预测时间**: 2026-06-12 07:58:32

上午十点这场，感觉算法倾向韩国

作者: mezhan 时间: 5 天前
"### Top Scorelines
- Mexico 1-1 South Africa: 13.7%
- Mexico 1-0 South Africa: 13.5%
- Mexico 0-1 South Africa: 10.7%
- Mexico 0-0 South Africa: 10.2%
- Mexico 2-1 South Africa: 8.2%"

Real Score Mexico 2-0 South Africa

作者: xiejin77 时间: 5 天前

mezhan 发表于 2026-6-12 09:54
"### Top Scorelines
- Mexico 1-1 South Africa: 13.7%
- Mexico 1-0 South Africa: 13.5%

这个是蒙特卡洛的算法跑的，仿足球经理的设计，但感觉不是很准

作者: 老票 时间: 5 天前
本帖最后由老票于 2026-6-12 11:56 编辑

xiejin77 发表于 2026-6-12 08:01
# 2026 世界杯预测报告：Korea Republic vs Czechia

- **预测时间**: 2026-06-12 07:58:32

开赛15分钟，韩国正在压着捷克打，比分还是0:0

捷克队衣服质量太差，韩国队一把扯破，漏出了捷克15号舒尔茨的腰间赘肉

下半场60分钟，捷克居然借着界外球手榴弹，禁区内头球攻门，进了！ 1:0！

66分钟，韩国队反复在前场传切跑终于抓住了机会，黄仁范插入禁区轻松一扣，然后搓球后角，1:1！

孙兴慜被换下，他还是亚洲最顶级的前锋，技术意识身体力量都上佳

76分钟，捷克前场任意球，头球破门！可惜越位了... 捷克平均身高1.86米以上，角球任意球是最有力的手段

79分钟，韩国再次利用快速穿插右路低传到禁区，无线亏左脚弓一推，2：1！捷克中后卫跟不上了，门将瞬间反应也差点意思

镜头扫过观众席，内德维德焦急的面容，他曾是我最喜欢的捷克球星，也是捷克足球历史上最辉煌的印证

进球后韩国主帅洪明甫迅速换上在中超浙江队效力的防守后腰朴振燮，站在前中超北京国安中后卫金玟哉身前，准备守住这个领先优势

93分钟，捷克队回敲大禁区前，插上队员一脚劲射，韩国门将迅速下地扑出，这是他第二次世界级扑救了！

洪明甫还是蛮帅的，面部沧桑感如同刀刻一般...埃及主裁判的分寸感很好，比赛很流畅

最后，韩国胜出！

作者: xiejin77 时间: 5 天前
# 2026 世界杯预测报告：Canada vs Bosnia and Herzegovina

- **预测时间**: 2026-06-12 12:35:59
- **Run ID**: 20260612_123559
- **比赛信息**: 第 3 场 · B组
- **开球时间**: 2026-06-13 06:00 北京时间（原始：2026-06-12 18:00 ET）
- **场馆**: BMO Field · Toronto
- **贝叶斯数据更新时间**: 2026-06-11T08:19:09

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 49.1%，领先第二选项 21.9 个百分点。
- 综合置信度为低；模型分歧为 23.4%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 The Odds API，“主队胜”的模型概率低于市场 3.6 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-11T08:19:09。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Canada 胜: 49.1%
- 平局: 23.7%
- Bosnia and Herzegovina 胜: 27.2%
- 综合置信度: 低
- 方法分歧: 23.4%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Canada 60.8% / 平局 18.1% / Bosnia and Herzegovina 21.1%
- Elo: 1556 vs 1380
- lineup: Canada 4-4-2 lineup confirmed from expert; key starters: Tajon Buchanan, Ismael Kone, Stephen Eustáquio（幅度 3.8%）
- lineup: Bosnia 4-4-2 lineup confirmed from expert; key starters: Ivan Sunjic, Benjamin Tahirovic, Amar Memic（幅度 3.8%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.62
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.62
- 原始伤停线索 5 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 1000
- 90分钟胜平负: Canada 37.4% / 平局 29.3% / Bosnia and Herzegovina 33.3%
- 平均进球: 1.19 - 1.13
- 模拟引擎: fm-style-chain-simulator (simulate_match -> simulate_attack_chain)
- 是否直接出结果: 否，逐场逐攻击链模拟
- 攻击链审计: 总计 40937 条，场均 40.9 条
- 射门/转化率: 总射门 34518 次，进球转化率 6.7%
- 稳定性审计: 中；单项 95% 误差上限约 ±3.0 个百分点
- 执行链路: sample starting XI players -> simulate possession-weighted attack windows -> resolve pass/dribble/shot duels with player attributes -> add set pieces, cards and penalties -> aggregate regulation probabilities across all simulations

### Monte Carlo Samples / 模拟样本轨迹

- 样本 1: 90分钟 2-3 / 最终 2-3；射门 17-17，攻击链 41，定位球链 6
- 样本 2: 90分钟 1-2 / 最终 1-2；射门 18-17，攻击链 44，定位球链 8
- 样本 3: 90分钟 3-0 / 最终 3-0；射门 16-21，攻击链 44，定位球链 9
- 样本 4: 90分钟 0-2 / 最终 0-2；射门 12-15，攻击链 35，定位球链 5
- 样本 5: 90分钟 4-0 / 最终 4-0；射门 14-24，攻击链 46，定位球链 11
- 稳定性提醒: Monte Carlo sample is usable, but close probabilities should be interpreted with the confidence interval.
- 90分钟结果: Canada 37.4% / 平局 29.3% / Bosnia and Herzegovina 33.3%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 920110718
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 23.9°C; 风速 21.2 km/h; 湿度 50%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 未知球队: {'headline': 'Haris Tabaković expected to miss World Cup with Bosnia after suffering injury for Gladbach', 'team': 'Bosnia', 'sourceType': 'google-news-raw', 'status': 'candidate', 'applied': False, 'warnings': [], 'matchedPlayers': ['haris tabakovic'], 'confidence': 0.72} -> 未映射到具体球员，仅记录 warning。

## Lineups / 阵容来源与匹配状态

- 主队 Canada: 4-4-2 / expert / predicted / confidence 0.74
  - 首发: Maxime Crépeau, Niko Sigur, Moïse Bombito, Alfie Jones, Richie Laryea, Tajon Buchanan, Ismael Kone, Stephen Eustáquio, Alphonso Davies, Tani Oluwaseyi, Jonathan David
- 客队 Bosnia: 4-4-2 / expert / predicted / confidence 0.74
  - 首发: Nikola Vasilj, Amar Dedic, Nikola Katic, Tarik Muharemovic, Sead Kolasinac, Esmir Bajraktarevic, Ivan Sunjic, Benjamin Tahirovic, Amar Memic, Ermedin Demirović, Edin Dzeko

## Diagnosis / 综合诊断

- Bayesian is higher on home win by 23.4pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 11.2pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on away win by 12.2pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

## Market Comparison / 外部市场概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Canada vs Bosnia and Herzegovina
- 市场更新时间: 2026-06-11T08:19:09
- 主队胜: 模型 49.1% / The Odds API 52.7% / 差值 -3.6pp
- 平局: 模型 23.7% / The Odds API 26.7% / 差值 -3.0pp
- 客队胜: 模型 27.2% / The Odds API 20.6% / 差值 +6.6pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Canada vs Bosnia

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.83

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Canada Win | 60.8% | 37.4% | 49.1% |
| Draw | 18.1% | 29.3% | 23.7% |
| Bosnia Win | 21.1% | 33.3% | 27.2% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Canada Elo: 1556
- Bosnia Elo: 1380
- P0: Canada 60.7% / Draw 18.2% / Bosnia 21.1%

### Likelihood Updates
- [home] Canada 4-4-2 lineup confirmed from expert; key starters: Tajon Buchanan, Ismael Kone, Stephen Eustáquio (±4%) — lineup
- [away] Bosnia 4-4-2 lineup confirmed from expert; key starters: Ivan Sunjic, Benjamin Tahirovic, Amar Memic (±4%) — lineup

### P1 (Final Bayesian)
- Canada: 60.8%  |  Draw: 18.1%  |  Bosnia: 21.1%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 1,000
- Avg Goals: Canada 1.19 / Bosnia 1.13 / Total 2.32
- Avg Shots: Canada 17.5 / Bosnia 17.0
- Conversion: Canada 6.8% / Bosnia 6.7%
- Avg Cards: 3.6

### Top Scorelines
- Canada 1-1 Bosnia: 15.1%
- Canada 1-0 Bosnia: 12.3%
- Canada 0-1 Bosnia: 10.0%
- Canada 2-1 Bosnia: 9.1%
- Canada 0-0 Bosnia: 8.7%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 23.4%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Bayesian is higher on home win by 23.4pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 11.2pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on away win by 12.2pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

### Consensus Estimate
- Canada: 49.1%  |  Draw: 23.7%  |  Bosnia: 27.2%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 5 天前
# 2026 世界杯预测报告：USA vs Paraguay

- **预测时间**: 2026-06-12 19:58:21
- **Run ID**: 20260612_195821
- **比赛信息**: 第 4 场 · D组
- **开球时间**: 2026-06-13 09:00 北京时间（原始：2026-06-12 21:00 ET）
- **场馆**: SoFi Stadium · Los Angeles
- **贝叶斯数据更新时间**: 2026-06-11T08:20:04

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 48.9%，领先第二选项 19.9 个百分点。
- 综合置信度为低；模型分歧为 22.4%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 当前没有可标准化的外部市场胜平负对照，市场概率不进入模型，只作为外部参考。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-11T08:20:04。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- USA 胜: 48.9%
- 平局: 22.2%
- Paraguay 胜: 28.9%
- 综合置信度: 低
- 方法分歧: 22.4%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: USA 60.1% / 平局 18.1% / Paraguay 21.8%
- Elo: 1673 vs 1504
- lineup: USA 4-3-3 lineup confirmed from manual; key starters: Weston McKennie, Sebastian Berhalter, Tyler Adams（幅度 6.5%）
- lineup: Paraguay 4-4-2 lineup confirmed from manual; key starters: Andres Cubas, Braian Ojeda, Alex Arce（幅度 6.5%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 原始伤停线索 8 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 1000
- 90分钟胜平负: USA 37.7% / 平局 26.2% / Paraguay 36.1%
- 平均进球: 1.23 - 1.20
- 模拟引擎: fm-style-chain-simulator (simulate_match -> simulate_attack_chain)
- 是否直接出结果: 否，逐场逐攻击链模拟
- 攻击链审计: 总计 41040 条，场均 41.0 条
- 射门/转化率: 总射门 35378 次，进球转化率 6.9%
- 稳定性审计: 中；单项 95% 误差上限约 ±3.0 个百分点
- 执行链路: sample starting XI players -> simulate possession-weighted attack windows -> resolve pass/dribble/shot duels with player attributes -> add set pieces, cards and penalties -> aggregate regulation probabilities across all simulations

### Monte Carlo Samples / 模拟样本轨迹

- 样本 1: 90分钟 1-1 / 最终 1-1；射门 17-18，攻击链 42，定位球链 7
- 样本 2: 90分钟 4-2 / 最终 4-2；射门 16-19，攻击链 41，定位球链 9
- 样本 3: 90分钟 3-2 / 最终 3-2；射门 21-11，攻击链 39，定位球链 5
- 样本 4: 90分钟 0-0 / 最终 0-0；射门 24-15，攻击链 47，定位球链 9
- 样本 5: 90分钟 1-3 / 最终 1-3；射门 19-18，攻击链 42，定位球链 8
- 稳定性提醒: Monte Carlo sample is usable, but close probabilities should be interpreted with the confidence interval.
- 90分钟结果: USA 37.7% / 平局 26.2% / Paraguay 36.1%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 920110718
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 23°C; 风速 11.3 km/h; 湿度 58%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 USA: 4-3-3 / manual / confirmed / confidence 0.78
  - 首发: Matt Freese, Alex Freeman, Chris Richards, Auston Trusty, Antonee Robinson, Weston McKennie, Sebastian Berhalter, Tyler Adams, Tim Weah, Folarin Balogun, Christian Pulisic
- 客队 Paraguay: 4-4-2 / manual / confirmed / confidence 0.78
  - 首发: Gaston Olveira, Alexandro Maidana, Fabian Balbuena, Gustavo Gomez, Gustavo Velazquez, Andres Cubas, Braian Ojeda, Alex Arce, Antonio Sanabria, Gabriel Avalos, Gustavo Caballero

## Diagnosis / 综合诊断

- Bayesian is higher on home win by 22.4pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 8.1pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on away win by 14.3pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

## Market Comparison / 外部市场概率对照

暂无外部市场对照。

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: USA vs Paraguay

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| USA Win | 60.1% | 37.7% | 48.9% |
| Draw | 18.1% | 26.2% | 22.2% |
| Paraguay Win | 21.8% | 36.1% | 28.9% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- USA Elo: 1673
- Paraguay Elo: 1504
- P0: USA 59.8% / Draw 18.6% / Paraguay 21.7%

### Likelihood Updates
- [home] USA 4-3-3 lineup confirmed from manual; key starters: Weston McKennie, Sebastian Berhalter, Tyler Adams (±6%) — lineup
- [away] Paraguay 4-4-2 lineup confirmed from manual; key starters: Andres Cubas, Braian Ojeda, Alex Arce (±6%) — lineup

### P1 (Final Bayesian)
- USA: 60.1%  |  Draw: 18.1%  |  Paraguay: 21.8%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 1,000
- Avg Goals: USA 1.23 / Paraguay 1.20 / Total 2.43
- Avg Shots: USA 18.2 / Paraguay 17.2
- Conversion: USA 6.7% / Paraguay 7.0%
- Avg Cards: 3.6

### Top Scorelines
- USA 1-1 Paraguay: 12.5%
- USA 0-1 Paraguay: 11.4%
- USA 1-0 Paraguay: 11.0%
- USA 0-0 Paraguay: 8.4%
- USA 1-2 Paraguay: 7.8%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 22.4%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Bayesian is higher on home win by 22.4pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 8.1pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on away win by 14.3pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

### Consensus Estimate
- USA: 48.9%  |  Draw: 22.2%  |  Paraguay: 28.9%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 4 天前
# 2026 世界杯预测报告：Australia vs Türkiye

- **预测时间**: 2026-06-13 13:32:05
- **Run ID**: 20260613_133205
- **比赛信息**: 第 6 场 · D组
- **开球时间**: 2026-06-14 12:00 北京时间（原始：2026-06-13 21:00 local）
- **场馆**: BC Place Vancouver · Vancouver
- **贝叶斯数据更新时间**: 2026-06-13T11:44:34

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 37.7%，领先第二选项 2.2 个百分点。
- 综合置信度为中；模型分歧为 7.8%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 The Odds API，“主队胜”的模型概率高于市场 18.6 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-13T11:44:34。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Australia 胜: 37.7%
- 平局: 26.9%
- Türkiye 胜: 35.5%
- 综合置信度: 中
- 方法分歧: 7.8%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Australia 34.4% / 平局 26.2% / Türkiye 39.4%
- Elo: 1581 vs 1599
- audited recent form: Weighted recent form gap: home=0.40 vs away=0.86（幅度 5.1%）
- lineup: Australia 4-2-3-1 lineup confirmed from manual; key starters: Irvine, Devlin, Hrustic（幅度 6.5%）
- lineup: Turkey 4-3-3 lineup confirmed from manual; key starters: Hakan Calhanoglu, Ismail Yuksek, Kaan Ayhan（幅度 6.5%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[form]: 最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。；置信度 0.55
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 原始伤停线索 5 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 1000
- 90分钟胜平负: Australia 40.9% / 平局 27.5% / Türkiye 31.6%
- 平均进球: 1.21 - 1.07
- 模拟引擎: fm-style-chain-simulator (simulate_match -> simulate_attack_chain)
- 是否直接出结果: 否，逐场逐攻击链模拟
- 攻击链审计: 总计 41023 条，场均 41.0 条
- 射门/转化率: 总射门 34990 次，进球转化率 6.5%
- 稳定性审计: 中；单项 95% 误差上限约 ±3.0 个百分点
- 执行链路: sample starting XI players -> simulate possession-weighted attack windows -> resolve pass/dribble/shot duels with player attributes -> add set pieces, cards and penalties -> aggregate regulation probabilities across all simulations

### Monte Carlo Samples / 模拟样本轨迹

- 样本 1: 90分钟 1-2 / 最终 1-2；射门 22-19，攻击链 47，定位球链 12
- 样本 2: 90分钟 0-1 / 最终 0-1；射门 16-23，攻击链 47，定位球链 11
- 样本 3: 90分钟 2-1 / 最终 2-1；射门 20-20，攻击链 46，定位球链 8
- 样本 4: 90分钟 1-1 / 最终 1-1；射门 15-20，攻击链 42，定位球链 6
- 样本 5: 90分钟 1-1 / 最终 1-1；射门 10-18，攻击链 36，定位球链 8
- 稳定性提醒: Monte Carlo sample is usable, but close probabilities should be interpreted with the confidence interval.
- 90分钟结果: Australia 40.9% / 平局 27.5% / Türkiye 31.6%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 920110718
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 24.4°C; 风速 7.1 km/h; 湿度 38%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Australia: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Beach, Behich, Bos, Burgess, Circati, Devlin, Hrustic, Irankunda, Irvine, Leckie, Mabil
- 客队 Turkey: 4-3-3 / manual / confirmed / confidence 0.78
  - 首发: Altay Bayindir, Abdulkerim Bardakcı, Caglar Soyuncu (Fenerbahce, Eren Elmalı, Ferdi Kadioglu, Hakan Calhanoglu, Ismail Yuksek, Kaan Ayhan, Arda Guler, Barış Alper Yilmaz, Can Uzun

## Diagnosis / 综合诊断

- 两种方法分歧较低或暂无额外诊断。

## Market Comparison / 外部市场概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Australia vs Türkiye
- 市场更新时间: 2026-06-13T11:44:34
- 主队胜: 模型 37.7% / The Odds API 19.1% / 差值 +18.6pp
- 平局: 模型 26.9% / The Odds API 25.6% / 差值 +1.3pp
- 客队胜: 模型 35.5% / The Odds API 55.3% / 差值 -19.8pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Australia vs Turkey

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: MEDIUM
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Australia Win | 34.4% | 40.9% | 37.7% |
| Draw | 26.2% | 27.5% | 26.9% |
| Turkey Win | 39.4% | 31.6% | 35.5% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Australia Elo: 1581
- Turkey Elo: 1599
- P0: Australia 34.6% / Draw 26.9% / Turkey 38.5%

### Likelihood Updates
- [away] Weighted recent form gap: home=0.40 vs away=0.86 (±5%) — audited recent form
- [home] Australia 4-2-3-1 lineup confirmed from manual; key starters: Irvine, Devlin, Hrustic (±6%) — lineup
- [away] Turkey 4-3-3 lineup confirmed from manual; key starters: Hakan Calhanoglu, Ismail Yuksek, Kaan Ayhan (±6%) — lineup

### P1 (Final Bayesian)
- Australia: 34.4%  |  Draw: 26.2%  |  Turkey: 39.4%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 1,000
- Avg Goals: Australia 1.21 / Turkey 1.07 / Total 2.28
- Avg Shots: Australia 17.5 / Turkey 17.5
- Conversion: Australia 6.9% / Turkey 6.1%
- Avg Cards: 3.7

### Top Scorelines
- Australia 1-1 Turkey: 14.1%
- Australia 1-0 Turkey: 13.4%
- Australia 0-1 Turkey: 10.9%
- Australia 2-1 Turkey: 10.4%
- Australia 0-0 Turkey: 8.7%

## Layer 3: Synthesis

**Confidence: MEDIUM** (max divergence: 7.8%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Bayesian evidence posterior and Monte Carlo simulation share the same baseline; equal weights are used by default.

### Consensus Estimate
- Australia: 37.7%  |  Draw: 26.9%  |  Turkey: 35.5%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 4 天前
# 2026 世界杯预测报告：Qatar vs Switzerland

- **预测时间**: 2026-06-13 13:55:47
- **Run ID**: 20260613_135547
- **比赛信息**: 第 8 场 · B组
- **开球时间**: 2026-06-14 03:00 北京时间（原始：2026-06-13 12:00 local）
- **场馆**: San Francisco Bay Area Stadium · San Francisco Bay Area
- **贝叶斯数据更新时间**: 2026-06-13T13:55:32

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“客队胜”，概率为 51.5%，领先第二选项 24.2 个百分点。
- 综合置信度为低；模型分歧为 22.6%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 The Odds API，“客队胜”的模型概率低于市场 26.7 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-13T13:55:32。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Qatar 胜: 27.3%
- 平局: 21.2%
- Switzerland 胜: 51.5%
- 综合置信度: 低
- 方法分歧: 22.6%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Qatar 19.7% / 平局 17.5% / Switzerland 62.8%
- Elo: 1460 vs 1649
- audited recent form: Weighted recent form gap: home=0.07 vs away=0.55（幅度 5.2%）
- lineup: Qatar 4-3-3 lineup confirmed from manual; key starters: Abdulaziz Hatem, Ahmed Fathi, Assim Madibo（幅度 6.5%）
- lineup: Switzerland 3-5-2 lineup confirmed from manual; key starters: Ardon Jashari, Denis Zakaria, Djibril Sow（幅度 6.5%）
- lineup: Switzerland uses 3-5-2; conservative/extra-defender shape（幅度 3.5%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[form]: 最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。；置信度 0.55
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[formation]: 三中卫/五后卫阵型通常更保守，贝叶斯侧仅轻微提高平局倾向。；置信度 0.78
- 原始伤停线索 9 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 1000
- 90分钟胜平负: Qatar 34.9% / 平局 24.9% / Switzerland 40.2%
- 平均进球: 1.23 - 1.34
- 模拟引擎: fm-style-chain-simulator (simulate_match -> simulate_attack_chain)
- 是否直接出结果: 否，逐场逐攻击链模拟
- 攻击链审计: 总计 40925 条，场均 40.9 条
- 射门/转化率: 总射门 35285 次，进球转化率 7.3%
- 稳定性审计: 中；单项 95% 误差上限约 ±3.0 个百分点
- 执行链路: sample starting XI players -> simulate possession-weighted attack windows -> resolve pass/dribble/shot duels with player attributes -> add set pieces, cards and penalties -> aggregate regulation probabilities across all simulations

### Monte Carlo Samples / 模拟样本轨迹

- 样本 1: 90分钟 1-1 / 最终 1-1；射门 16-20，攻击链 42，定位球链 7
- 样本 2: 90分钟 0-4 / 最终 0-4；射门 10-23，攻击链 40，定位球链 8
- 样本 3: 90分钟 2-1 / 最终 2-1；射门 16-15，攻击链 37，定位球链 7
- 样本 4: 90分钟 1-1 / 最终 1-1；射门 14-15，攻击链 39，定位球链 7
- 样本 5: 90分钟 2-3 / 最终 2-3；射门 18-19，攻击链 42，定位球链 8
- 稳定性提醒: Monte Carlo sample is usable, but close probabilities should be interpreted with the confidence interval.
- 90分钟结果: Qatar 34.9% / 平局 24.9% / Switzerland 40.2%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 920110718
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 25.1°C; 风速 10.1 km/h; 湿度 38%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Qatar: 4-3-3 / manual / confirmed / confidence 0.78
  - 首发: Mahmoud Abunada, Ayoub Alawi, Boualem Khoukhi, Hashmi Hussein, Homam Al-Amin, Abdulaziz Hatem, Ahmed Fathi, Assim Madibo, Ahmed Al-Janahi, Ahmed Alaa, Akram Afif
- 客队 Switzerland: 3-5-2 / manual / confirmed / confidence 0.78
  - 首发: Gregor Kobel, Aurele Amenda, Eray Comert, Luca Jaquez, Ardon Jashari, Denis Zakaria, Djibril Sow, Breel Embolo, Cedric Itten, Christian Fassnacht, Dan Ndoye

## Diagnosis / 综合诊断

- Monte Carlo is higher on home win by 15.2pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Bayesian is higher on away win by 22.6pp. Likely evidence-layer effect (4 evidence items) rather than player-chain simulation.

## Market Comparison / 外部市场概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Qatar vs Switzerland
- 市场更新时间: 2026-06-13T13:55:32
- 主队胜: 模型 27.3% / The Odds API 7.0% / 差值 +20.2pp
- 平局: 模型 21.2% / The Odds API 14.8% / 差值 +6.4pp
- 客队胜: 模型 51.5% / The Odds API 78.2% / 差值 -26.7pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Qatar vs Switzerland

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Qatar Win | 19.7% | 34.9% | 27.3% |
| Draw | 17.5% | 24.9% | 21.2% |
| Switzerland Win | 62.8% | 40.2% | 51.5% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Qatar Elo: 1460
- Switzerland Elo: 1649
- P0: Qatar 20.0% / Draw 17.7% / Switzerland 62.3%

### Likelihood Updates
- [away] Weighted recent form gap: home=0.07 vs away=0.55 (±5%) — audited recent form
- [home] Qatar 4-3-3 lineup confirmed from manual; key starters: Abdulaziz Hatem, Ahmed Fathi, Assim Madibo (±6%) — lineup
- [away] Switzerland 3-5-2 lineup confirmed from manual; key starters: Ardon Jashari, Denis Zakaria, Djibril Sow (±6%) — lineup
- [draw] Switzerland uses 3-5-2; conservative/extra-defender shape (±4%) — lineup

### P1 (Final Bayesian)
- Qatar: 19.7%  |  Draw: 17.5%  |  Switzerland: 62.8%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 1,000
- Avg Goals: Qatar 1.23 / Switzerland 1.34 / Total 2.57
- Avg Shots: Qatar 17.0 / Switzerland 18.3
- Conversion: Qatar 7.2% / Switzerland 7.3%
- Avg Cards: 3.6

### Top Scorelines
- Qatar 1-1 Switzerland: 11.1%
- Qatar 0-1 Switzerland: 10.3%
- Qatar 1-0 Switzerland: 9.1%
- Qatar 1-2 Switzerland: 9.1%
- Qatar 0-2 Switzerland: 8.6%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 22.6%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Monte Carlo is higher on home win by 15.2pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Bayesian is higher on away win by 22.6pp. Likely evidence-layer effect (4 evidence items) rather than player-chain simulation.

### Consensus Estimate
- Qatar: 27.3%  |  Draw: 21.2%  |  Switzerland: 51.5%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 4 天前

xiejin77 发表于 2026-6-13 13:56
# 2026 世界杯预测报告：Qatar vs Switzerland

- **预测时间**: 2026-06-13 13:55:47

# 2026 世界杯预测报告：Haiti vs Scotland

- **预测时间**: 2026-06-13 13:56:24
- **Run ID**: 20260613_135624
- **比赛信息**: 第 5 场 · C组
- **开球时间**: 2026-06-14 09:00 北京时间（原始：2026-06-13 21:00 local）
- **场馆**: Boston Stadium · Boston
- **贝叶斯数据更新时间**: 2026-06-13T13:56:19

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“客队胜”，概率为 50.8%，领先第二选项 24.5 个百分点。
- 综合置信度为中；模型分歧为 11.8%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 The Odds API，“客队胜”的模型概率低于市场 10.6 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-13T13:56:19。
- 阵容数据存在未确认或未匹配项，建议先在阵容弹窗中手动修正首发名单。

## Synthesis / 综合概率

- Haiti 胜: 26.3%
- 平局: 22.9%
- Scotland 胜: 50.8%
- 综合置信度: 中
- 方法分歧: 11.8%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Haiti 23.7% / 平局 19.6% / Scotland 56.7%
- Elo: 1350 vs 1498
- audited recent form: Weighted recent form gap: home=0.73 vs away=0.44（幅度 3.2%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[form]: 最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。；置信度 0.55
- 原始伤停线索 4 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 1000
- 90分钟胜平负: Haiti 28.9% / 平局 26.2% / Scotland 44.9%
- 平均进球: 1.02 - 1.34
- 模拟引擎: fm-style-chain-simulator (simulate_match -> simulate_attack_chain)
- 是否直接出结果: 否，逐场逐攻击链模拟
- 攻击链审计: 总计 40875 条，场均 40.9 条
- 射门/转化率: 总射门 33617 次，进球转化率 7.0%
- 稳定性审计: 中；单项 95% 误差上限约 ±3.1 个百分点
- 执行链路: sample starting XI players -> simulate possession-weighted attack windows -> resolve pass/dribble/shot duels with player attributes -> add set pieces, cards and penalties -> aggregate regulation probabilities across all simulations

### Monte Carlo Samples / 模拟样本轨迹

- 样本 1: 90分钟 0-1 / 最终 0-1；射门 15-21，攻击链 43，定位球链 8
- 样本 2: 90分钟 0-4 / 最终 0-4；射门 7-21，攻击链 38，定位球链 5
- 样本 3: 90分钟 1-1 / 最终 1-1；射门 15-19，攻击链 39，定位球链 9
- 样本 4: 90分钟 1-1 / 最终 1-1；射门 14-15，攻击链 40，定位球链 7
- 样本 5: 90分钟 1-2 / 最终 1-2；射门 20-21，攻击链 44，定位球链 8
- 稳定性提醒: Monte Carlo sample is usable, but close probabilities should be interpreted with the confidence interval.
- 90分钟结果: Haiti 28.9% / 平局 26.2% / Scotland 44.9%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 920110718
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 18.6°C; 风速 11.5 km/h; 湿度 84%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Haiti: 4-2-3-1 / roster-default / needs-confirmation / confidence 0.28
  - 首发: Alexandre Pierre, Carlens Arcus, Duke Lacroix, Hannes Delcroix, Jean-Kevin Duverne, Carl-Fred Sainthe, Dominique Simon, Derrick Etienne, Jean-Jacques Danley, Duckens Nazon, Frantzdy Pierrot
- 客队 Scotland: 3-5-2 / roster-default / needs-confirmation / confidence 0.28
  - 首发: Angus Gunn, Aaron Hickey, Andy Robertson, Anthony Ralston, Ben Gannon-Doak, Findlay Curtis, Jack Fletcher, Che Adams, George Hirst, Lawrence Shankland, Lyndon Dykes
- 阵容提醒: Haiti: no reliable predicted/recent XI available; using roster-default starters by formation slots. Please confirm manually.；External formation source was not parsed; using local tactical profile fallback.；Haiti lineup needs manual review.；Scotland: no reliable predicted/recent XI available; using roster-default starters by formation slots. Please confirm manually.；External formation source was not parsed; using local tactical profile fallback.；Scotland lineup needs manual review.

## Diagnosis / 综合诊断

- Bayesian is higher on away win by 11.8pp. Likely evidence-layer effect (1 evidence items) rather than player-chain simulation.

## Market Comparison / 外部市场概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Haiti vs Scotland
- 市场更新时间: 2026-06-13T13:56:19
- 主队胜: 模型 26.3% / The Odds API 16.6% / 差值 +9.7pp
- 平局: 模型 22.9% / The Odds API 21.9% / 差值 +1.0pp
- 客队胜: 模型 50.8% / The Odds API 61.4% / 差值 -10.6pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Haiti vs Scotland

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: MEDIUM
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Haiti Win | 23.7% | 28.9% | 26.3% |
| Draw | 19.6% | 26.2% | 22.9% |
| Scotland Win | 56.7% | 44.9% | 50.8% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Haiti Elo: 1350
- Scotland Elo: 1498
- P0: Haiti 23.4% / Draw 19.6% / Scotland 57.0%

### Likelihood Updates
- [home] Weighted recent form gap: home=0.73 vs away=0.44 (±3%) — audited recent form

### P1 (Final Bayesian)
- Haiti: 23.7%  |  Draw: 19.6%  |  Scotland: 56.7%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 1,000
- Avg Goals: Haiti 1.02 / Scotland 1.34 / Total 2.36
- Avg Shots: Haiti 15.5 / Scotland 18.1
- Conversion: Haiti 6.6% / Scotland 7.4%
- Avg Cards: 3.7

### Top Scorelines
- Haiti 1-1 Scotland: 12.8%
- Haiti 0-1 Scotland: 11.5%
- Haiti 1-0 Scotland: 10.5%
- Haiti 1-2 Scotland: 9.8%
- Haiti 0-2 Scotland: 9.7%

## Layer 3: Synthesis

**Confidence: MEDIUM** (max divergence: 11.8%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Bayesian evidence posterior and Monte Carlo simulation share the same baseline; equal weights are used by default.

### Divergence Analysis
- Bayesian is higher on away win by 11.8pp. Likely evidence-layer effect (1 evidence items) rather than player-chain simulation.

### Consensus Estimate
- Haiti: 26.3%  |  Draw: 22.9%  |  Scotland: 50.8%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 4 天前

xiejin77 发表于 2026-6-13 13:56
# 2026 世界杯预测报告：Haiti vs Scotland

- **预测时间**: 2026-06-13 13:56:24

# 2026 世界杯预测报告：Brazil vs Morocco

- **预测时间**: 2026-06-13 13:57:09
- **Run ID**: 20260613_135709
- **比赛信息**: 第 7 场 · C组
- **开球时间**: 2026-06-14 06:00 北京时间（原始：2026-06-13 18:00 local）
- **场馆**: New York/New Jersey Stadium · New Jersey
- **贝叶斯数据更新时间**: 2026-06-13T13:56:59

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 46.4%，领先第二选项 17.9 个百分点。
- 综合置信度为低；模型分歧为 23.6%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 The Odds API，“主队胜”的模型概率低于市场 10.7 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-13T13:56:59。
- 阵容数据存在未确认或未匹配项，建议先在阵容弹窗中手动修正首发名单。

## Synthesis / 综合概率

- Brazil 胜: 46.4%
- 平局: 25.1%
- Morocco 胜: 28.5%
- 综合置信度: 低
- 方法分歧: 23.6%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Brazil 34.6% / 平局 26.9% / Morocco 38.6%
- Elo: 1761 vs 1756
- mapped team news: Wesley injury mapped (major)（幅度 13.0%）
- lineup: Brazil 4-4-2 lineup confirmed from expert; key starters: Bruno Guimaraes, Raphinha, Vinicius Junior（幅度 3.8%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[injury]: 伤停已映射到球队球员，且严重度明确；因此以保守权重影响对手方向。；置信度 0.85
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.62
- 原始伤停线索 9 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 1000
- 90分钟胜平负: Brazil 58.2% / 平局 23.4% / Morocco 18.4%
- 平均进球: 1.68 - 0.80
- 模拟引擎: fm-style-chain-simulator (simulate_match -> simulate_attack_chain)
- 是否直接出结果: 否，逐场逐攻击链模拟
- 攻击链审计: 总计 41222 条，场均 41.2 条
- 射门/转化率: 总射门 32365 次，进球转化率 7.7%
- 稳定性审计: 中；单项 95% 误差上限约 ±3.1 个百分点
- 执行链路: sample starting XI players -> simulate possession-weighted attack windows -> resolve pass/dribble/shot duels with player attributes -> add set pieces, cards and penalties -> aggregate regulation probabilities across all simulations

### Monte Carlo Samples / 模拟样本轨迹

- 样本 1: 90分钟 2-2 / 最终 2-2；射门 20-18，攻击链 45，定位球链 10
- 样本 2: 90分钟 3-0 / 最终 3-0；射门 19-13，攻击链 46，定位球链 10
- 样本 3: 90分钟 2-2 / 最终 2-2；射门 15-18，攻击链 40，定位球链 9
- 样本 4: 90分钟 2-0 / 最终 2-0；射门 18-8，攻击链 33，定位球链 5
- 样本 5: 90分钟 1-2 / 最终 1-2；射门 15-17，攻击链 43，定位球链 6
- 稳定性提醒: Monte Carlo sample is usable, but close probabilities should be interpreted with the confidence interval.
- 90分钟结果: Brazil 58.2% / 平局 23.4% / Morocco 18.4%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 920110718
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 29.8°C; 风速 19.4 km/h; 湿度 60%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- Brazil: {'headline': 'Brazil Right Back Wesley To Miss World Cup To Injury, Éderson Named Replacement', 'team': 'Brazil', 'sourceType': 'google-news-raw', 'status': 'candidate', 'applied': False, 'warnings': [], 'matchedPlayers': ['wesley', 'ederson'], 'confidence': 0.72} -> 命中 Wesley，明确缺阵/重罚。

## Lineups / 阵容来源与匹配状态

- 主队 Brazil: 4-4-2 / expert / predicted / confidence 0.74
  - 首发: Alisson, Wesley, Marquinhos, Gabriel Magalhaes, Douglas Santos, Luiz Henrique, Casemiro, Bruno Guimaraes, Raphinha, Vinicius Junior, Matheus Cunha
- 客队 Morocco: 4-2-3-1 / expert / predicted / confidence 0.74
  - 首发: Yassine Bounou, Achraf Hakimi, Issa Diop, Nayef Aguerd, Anass Salah-Eddine, Azzedine Ounahi, Neil El Aynaoui, Brahim Díaz, Ismael Saibari, Chemsdine Talbi, Ayoub El Kaabi
- 阵容提醒: Morocco: Bono fuzzy matched to Yassine Bounou; please review.；Morocco: Salah-Eddine fuzzy matched to Anass Salah-Eddine; please review.；Morocco: El Aynaoui fuzzy matched to Neil El Aynaoui; please review.；Morocco: El Kaabi fuzzy matched to Ayoub El Kaabi; please review.；Morocco lineup needs manual review.

## Diagnosis / 综合诊断

- Monte Carlo is higher on home win by 23.6pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Bayesian is higher on away win by 20.2pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.

## Market Comparison / 外部市场概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Brazil vs Morocco
- 市场更新时间: 2026-06-13T13:56:59
- 主队胜: 模型 46.4% / The Odds API 57.1% / 差值 -10.7pp
- 平局: 模型 25.1% / The Odds API 25.2% / 差值 -0.1pp
- 客队胜: 模型 28.5% / The Odds API 17.7% / 差值 +10.8pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Brazil vs Morocco

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.83

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Brazil Win | 34.6% | 58.2% | 46.4% |
| Draw | 26.9% | 23.4% | 25.1% |
| Morocco Win | 38.6% | 18.4% | 28.5% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Brazil Elo: 1761
- Morocco Elo: 1756
- P0: Brazil 36.7% / Draw 27.7% / Morocco 35.6%

### Likelihood Updates
- [away] Wesley injury mapped (major) (±13%) — mapped team news
- [home] Brazil 4-4-2 lineup confirmed from expert; key starters: Bruno Guimaraes, Raphinha, Vinicius Junior (±4%) — lineup

### P1 (Final Bayesian)
- Brazil: 34.6%  |  Draw: 26.9%  |  Morocco: 38.6%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 1,000
- Avg Goals: Brazil 1.68 / Morocco 0.80 / Total 2.48
- Avg Shots: Brazil 17.9 / Morocco 14.5
- Conversion: Brazil 9.4% / Morocco 5.5%
- Avg Cards: 3.6

### Top Scorelines
- Brazil 1-0 Morocco: 13.1%
- Brazil 2-0 Morocco: 11.5%
- Brazil 1-1 Morocco: 11.2%
- Brazil 2-1 Morocco: 9.4%
- Brazil 0-0 Morocco: 8.2%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 23.6%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Monte Carlo is higher on home win by 23.6pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Bayesian is higher on away win by 20.2pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.

### Consensus Estimate
- Brazil: 46.4%  |  Draw: 25.1%  |  Morocco: 28.5%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 3 天前
# 2026 世界杯预测报告：Côte d'Ivoire vs Ecuador

- **预测时间**: 2026-06-14 13:51:39
- **Run ID**: 20260614_135139
- **比赛信息**: 第 9 场 · E组
- **开球时间**: 2026-06-15 07:00 北京时间（原始：2026-06-14 19:00 local）
- **场馆**: Philadelphia Stadium · Philadelphia
- **贝叶斯数据更新时间**: 2026-06-14T13:51:13

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“客队胜”，概率为 38.7%，领先第二选项 2.4 个百分点。
- 综合置信度为中；模型分歧为 5.2%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 The Odds API，“客队胜”的模型概率高于市场 1.0 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-14T13:51:13。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Côte d'Ivoire 胜: 36.3%
- 平局: 25.0%
- Ecuador 胜: 38.7%
- 综合置信度: 中
- 方法分歧: 5.2%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Côte d'Ivoire 35.1% / 平局 23.5% / Ecuador 41.3%
- Elo: 1533 vs 1595
- mapped team news: Moises Caicedo injury mapped (major)（幅度 13.0%）
- lineup: Ivory Coast 4-3-3 lineup confirmed from manual; key starters: Franck Kessie, Ibrahim Sangaré, Christ Inao Oulaï（幅度 6.5%）
- lineup: Ecuador 4-3-3 lineup confirmed from expert; key starters: Moises Caicedo, Pedro Vite, Denil Castillo（幅度 3.8%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[injury]: 伤停已映射到球队球员，且严重度明确；因此以保守权重影响对手方向。；置信度 0.85
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.62
- 原始伤停线索 3 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 1000
- 90分钟胜平负: Côte d'Ivoire 37.4% / 平局 26.5% / Ecuador 36.1%
- 平均进球: 1.22 - 1.18
- 模拟引擎: fm-style-chain-simulator (simulate_match -> simulate_attack_chain)
- 是否直接出结果: 否，逐场逐攻击链模拟
- 攻击链审计: 总计 40943 条，场均 40.9 条
- 射门/转化率: 总射门 34689 次，进球转化率 6.9%
- 稳定性审计: 中；单项 95% 误差上限约 ±3.0 个百分点
- 执行链路: sample starting XI players -> simulate possession-weighted attack windows -> resolve pass/dribble/shot duels with player attributes -> add set pieces, cards and penalties -> aggregate regulation probabilities across all simulations

### Monte Carlo Samples / 模拟样本轨迹

- 样本 1: 90分钟 1-3 / 最终 1-3；射门 22-14，攻击链 43，定位球链 8
- 样本 2: 90分钟 1-1 / 最终 1-1；射门 17-18，攻击链 45，定位球链 10
- 样本 3: 90分钟 1-0 / 最终 1-0；射门 20-15，攻击链 39，定位球链 6
- 样本 4: 90分钟 3-3 / 最终 3-3；射门 17-16，攻击链 44，定位球链 7
- 样本 5: 90分钟 4-0 / 最终 4-0；射门 15-20，攻击链 41，定位球链 9
- 稳定性提醒: Monte Carlo sample is usable, but close probabilities should be interpreted with the confidence interval.
- 90分钟结果: Côte d'Ivoire 37.4% / 平局 26.5% / Ecuador 36.1%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 920110718
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 29.9°C; 风速 21.3 km/h; 湿度 46%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- Ecuador: {'headline': 'Moises Caicedo out of starting XI: Confirmed lineups for Ecuador vs Guatemala in pre-World Cup 2026 interna...', 'team': 'Ecuador', 'sourceType': 'google-news-raw', 'status': 'candidate', 'applied': False, 'warnings': [], 'matchedPlayers': ['moises caicedo'], 'confidence': 0.72} -> 命中 Moises Caicedo，明确缺阵/重罚。

## Lineups / 阵容来源与匹配状态

- 主队 Ivory Coast: 4-3-3 / manual / confirmed / confidence 0.78
  - 首发: Yahia Fofana, Guela Doue, Odilon Kossounou, Evan Ndicka, Ghislain Konan, Franck Kessie, Ibrahim Sangaré, Christ Inao Oulaï, Nicolas Pepe, Evann Guessand, Ousmane Diomande
- 客队 Ecuador: 4-3-3 / expert / predicted / confidence 0.74
  - 首发: Hernan Galíndez, Angelo Preciado, Joel Ordonez, Willian Pacho, Piero Hincapie, Pedro Vite, Moises Caicedo, Denil Castillo, John Yeboah, Anthony Valencia, Nilson Angulo

## Diagnosis / 综合诊断

- 两种方法分歧较低或暂无额外诊断。

## Market Comparison / 外部市场概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Côte d'Ivoire vs Ecuador
- 市场更新时间: 2026-06-14T13:49:49
- 主队胜: 模型 36.3% / The Odds API 29.2% / 差值 +7.1pp
- 平局: 模型 25.0% / The Odds API 33.1% / 差值 -8.1pp
- 客队胜: 模型 38.7% / The Odds API 37.7% / 差值 +1.0pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Ivory Coast vs Ecuador

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: MEDIUM
- **Shared Baseline**: group / context quality 0.83

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Ivory Coast Win | 35.1% | 37.4% | 36.3% |
| Draw | 23.5% | 26.5% | 25.0% |
| Ecuador Win | 41.3% | 36.1% | 38.7% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Ivory Coast Elo: 1533
- Ecuador Elo: 1595
- P0: Ivory Coast 30.9% / Draw 24.3% / Ecuador 44.8%

### Likelihood Updates
- [home] Moises Caicedo injury mapped (major) (±13%) — mapped team news
- [home] Ivory Coast 4-3-3 lineup confirmed from manual; key starters: Franck Kessie, Ibrahim Sangaré, Christ Inao Oulaï (±6%) — lineup
- [away] Ecuador 4-3-3 lineup confirmed from expert; key starters: Moises Caicedo, Pedro Vite, Denil Castillo (±4%) — lineup

### P1 (Final Bayesian)
- Ivory Coast: 35.1%  |  Draw: 23.5%  |  Ecuador: 41.3%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 1,000
- Avg Goals: Ivory Coast 1.22 / Ecuador 1.18 / Total 2.39
- Avg Shots: Ivory Coast 17.3 / Ecuador 17.4
- Conversion: Ivory Coast 7.0% / Ecuador 6.8%
- Avg Cards: 3.6

### Top Scorelines
- Ivory Coast 1-1 Ecuador: 11.9%
- Ivory Coast 0-1 Ecuador: 11.5%
- Ivory Coast 1-0 Ecuador: 10.5%
- Ivory Coast 2-1 Ecuador: 8.3%
- Ivory Coast 0-2 Ecuador: 8.2%

## Layer 3: Synthesis

**Confidence: MEDIUM** (max divergence: 5.2%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Bayesian evidence posterior and Monte Carlo simulation share the same baseline; equal weights are used by default.

### Consensus Estimate
- Ivory Coast: 36.3%  |  Draw: 25.0%  |  Ecuador: 38.7%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 3 天前

xiejin77 发表于 2026-6-14 13:52
# 2026 世界杯预测报告：Côte d'Ivoire vs Ecuador

- **预测时间**: 2026-06-14 13:51:39

# 2026 世界杯预测报告：Germany vs Curaçao

- **预测时间**: 2026-06-14 13:57:49
- **Run ID**: 20260614_135749
- **比赛信息**: 第 10 场 · E组
- **开球时间**: 2026-06-15 01:00 北京时间（原始：2026-06-14 12:00 local）
- **场馆**: Houston Stadium · Houston
- **贝叶斯数据更新时间**: 2026-06-14T13:56:40

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 71.8%，领先第二选项 55.7 个百分点。
- 综合置信度为低；模型分歧为 19.6%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 The Odds API，“主队胜”的模型概率低于市场 19.6 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-14T13:56:40。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Germany 胜: 71.8%
- 平局: 16.1%
- Curaçao 胜: 12.1%
- 综合置信度: 低
- 方法分歧: 19.6%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Germany 81.6% / 平局 10.9% / Curaçao 7.6%
- Elo: 1730 vs 1340
- audited recent form: Weighted recent form gap: home=1.00 vs away=0.31（幅度 7.5%）
- lineup: Germany 4-2-3-1 lineup confirmed from manual; key starters: Leon Goretzka, Kai Havertz, Nico Schlotterbeck（幅度 6.5%）
- lineup: Curacao 4-3-3 lineup confirmed from manual; key starters: Ar’Jany Martha, Godfried Roemeratoe, Juninho Bacuna（幅度 6.5%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[form]: 最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。；置信度 0.55
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 原始伤停线索 6 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 1000
- 90分钟胜平负: Germany 62.0% / 平局 21.4% / Curaçao 16.6%
- 平均进球: 1.90 - 0.92
- 模拟引擎: fm-style-chain-simulator (simulate_match -> simulate_attack_chain)
- 是否直接出结果: 否，逐场逐攻击链模拟
- 攻击链审计: 总计 40933 条，场均 40.9 条
- 射门/转化率: 总射门 33374 次，进球转化率 8.5%
- 稳定性审计: 中；单项 95% 误差上限约 ±3.0 个百分点
- 执行链路: sample starting XI players -> simulate possession-weighted attack windows -> resolve pass/dribble/shot duels with player attributes -> add set pieces, cards and penalties -> aggregate regulation probabilities across all simulations

### Monte Carlo Samples / 模拟样本轨迹

- 样本 1: 90分钟 4-1 / 最终 4-1；射门 21-16，攻击链 42，定位球链 7
- 样本 2: 90分钟 2-2 / 最终 2-2；射门 16-18，攻击链 40，定位球链 8
- 样本 3: 90分钟 3-1 / 最终 3-1；射门 17-15，攻击链 41，定位球链 8
- 样本 4: 90分钟 2-1 / 最终 2-1；射门 22-13，攻击链 45，定位球链 8
- 样本 5: 90分钟 0-0 / 最终 0-0；射门 14-14，攻击链 37，定位球链 7
- 稳定性提醒: Monte Carlo sample is usable, but close probabilities should be interpreted with the confidence interval.
- 90分钟结果: Germany 62.0% / 平局 21.4% / Curaçao 16.6%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 920110718
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 29.4°C; 风速 21.6 km/h; 湿度 81%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Germany: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Manuel Neuer, Joshua Kimmich (Bayern Munich, Jonathan Tah, Nico Schlotterbeck, David Raum, Alexander Pavlovic, Leon Goretzka, Leroy Sane, Jamal Musiala, Florian Wirtz, Kai Havertz
- 客队 Curacao: 4-3-3 / manual / confirmed / confidence 0.78
  - 首发: Eloy Room, Armando Obispo, Deveron Fonville, Joshua Brenet, Jurien Gaari, Ar’Jany Martha, Godfried Roemeratoe, Juninho Bacuna, Brandley Kuwas, Gervane Kastaneer, Jearl Margaritha

## Diagnosis / 综合诊断

- Bayesian is higher on home win by 19.6pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 10.5pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

## Market Comparison / 外部市场概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Germany vs Curaçao
- 市场更新时间: 2026-06-14T13:56:40
- 主队胜: 模型 71.8% / The Odds API 91.4% / 差值 -19.6pp
- 平局: 模型 16.1% / The Odds API 5.8% / 差值 +10.3pp
- 客队胜: 模型 12.1% / The Odds API 2.8% / 差值 +9.3pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Germany vs Curacao

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Germany Win | 81.6% | 62.0% | 71.8% |
| Draw | 10.9% | 21.4% | 16.1% |
| Curacao Win | 7.6% | 16.6% | 12.1% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Germany Elo: 1730
- Curacao Elo: 1340
- P0: Germany 80.8% / Draw 11.4% / Curacao 7.8%

### Likelihood Updates
- [home] Weighted recent form gap: home=1.00 vs away=0.31 (±8%) — audited recent form
- [home] Germany 4-2-3-1 lineup confirmed from manual; key starters: Leon Goretzka, Kai Havertz, Nico Schlotterbeck (±6%) — lineup
- [away] Curacao 4-3-3 lineup confirmed from manual; key starters: Ar’Jany Martha, Godfried Roemeratoe, Juninho Bacuna (±6%) — lineup

### P1 (Final Bayesian)
- Germany: 81.6%  |  Draw: 10.9%  |  Curacao: 7.6%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 1,000
- Avg Goals: Germany 1.90 / Curacao 0.92 / Total 2.82
- Avg Shots: Germany 19.2 / Curacao 14.2
- Conversion: Germany 9.9% / Curacao 6.5%
- Avg Cards: 3.7

### Top Scorelines
- Germany 2-0 Curacao: 12.2%
- Germany 1-0 Curacao: 11.6%
- Germany 2-1 Curacao: 10.9%
- Germany 1-1 Curacao: 9.5%
- Germany 3-0 Curacao: 7.7%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 19.6%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Bayesian is higher on home win by 19.6pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 10.5pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

### Consensus Estimate
- Germany: 71.8%  |  Draw: 16.1%  |  Curacao: 12.1%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 3 天前

xiejin77 发表于 2026-6-14 13:59
# 2026 世界杯预测报告：Germany vs Curaçao

- **预测时间**: 2026-06-14 13:57:49

# 2026 世界杯预测报告：Netherlands vs Japan

- **预测时间**: 2026-06-14 14:06:54
- **Run ID**: 20260614_140654
- **比赛信息**: 第 11 场 · F组
- **开球时间**: 2026-06-15 04:00 北京时间（原始：2026-06-14 15:00 local）
- **场馆**: Dallas Stadium · Dallas
- **贝叶斯数据更新时间**: 2026-06-14T14:02:05

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 43.9%，领先第二选项 14.6 个百分点。
- 综合置信度为中；模型分歧为 12.0%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 The Odds API，“主队胜”的模型概率低于市场 3.3 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-14T14:02:05。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Netherlands 胜: 43.9%
- 平局: 26.9%
- Japan 胜: 29.3%
- 综合置信度: 中
- 方法分歧: 12.0%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Netherlands 49.9% / 平局 21.7% / Japan 28.4%
- Elo: 1758 vs 1660
- audited recent form: Weighted recent form gap: home=0.54 vs away=1.00（幅度 5.1%）
- lineup: Netherlands 4-2-3-1 lineup confirmed from manual; key starters: Frenkie de Jong, Guus Til, Marten de Roon（幅度 6.5%）
- lineup: Japan 4-2-3-1 lineup confirmed from manual; key starters: Doan Ritsu, Endo Wataru, Kamada Daichi（幅度 6.5%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[form]: 最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。；置信度 0.55
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 原始伤停线索 6 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 1000
- 90分钟胜平负: Netherlands 37.9% / 平局 32.0% / Japan 30.1%
- 平均进球: 1.14 - 0.96
- 模拟引擎: fm-style-chain-simulator (simulate_match -> simulate_attack_chain)
- 是否直接出结果: 否，逐场逐攻击链模拟
- 攻击链审计: 总计 41016 条，场均 41.0 条
- 射门/转化率: 总射门 34184 次，进球转化率 6.1%
- 稳定性审计: 中；单项 95% 误差上限约 ±3.0 个百分点
- 执行链路: sample starting XI players -> simulate possession-weighted attack windows -> resolve pass/dribble/shot duels with player attributes -> add set pieces, cards and penalties -> aggregate regulation probabilities across all simulations

### Monte Carlo Samples / 模拟样本轨迹

- 样本 1: 90分钟 1-2 / 最终 1-2；射门 15-21，攻击链 44，定位球链 9
- 样本 2: 90分钟 0-0 / 最终 0-0；射门 10-17，攻击链 42，定位球链 9
- 样本 3: 90分钟 1-1 / 最终 1-1；射门 13-20，攻击链 37，定位球链 7
- 样本 4: 90分钟 2-1 / 最终 2-1；射门 19-16，攻击链 43，定位球链 10
- 样本 5: 90分钟 0-0 / 最终 0-0；射门 18-19，攻击链 43，定位球链 9
- 稳定性提醒: Monte Carlo sample is usable, but close probabilities should be interpreted with the confidence interval.
- 90分钟结果: Netherlands 37.9% / 平局 32.0% / Japan 30.1%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 920110718
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 28.3°C; 风速 12.4 km/h; 湿度 76%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Netherlands: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Bart Verbruggen, Denzel Dumfries, John Paul van Hecke, Jorrel Hato, Lutsharel Geertruida, Frenkie de Jong, Guus Til, Brian Brobbey, Marten de Roon, Cody Gakpo, Crysencio Summerville
- 客队 Japan: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Hayakawa Tomoki, Itakura Ko, Ito Hiroki, Nagatomo Yuto, Seko Ayumu, Doan Ritsu, Endo Wataru, Goto Keisuke, Kamada Daichi, Ito Junya, Kubo Takefusa

## Diagnosis / 综合诊断

- Bayesian is higher on home win by 12.0pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 10.3pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

## Market Comparison / 外部市场概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Netherlands vs Japan
- 市场更新时间: 2026-06-14T13:59:42
- 主队胜: 模型 43.9% / The Odds API 47.1% / 差值 -3.3pp
- 平局: 模型 26.9% / The Odds API 26.9% / 差值 -0.0pp
- 客队胜: 模型 29.3% / The Odds API 26.0% / 差值 +3.3pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Netherlands vs Japan

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: MEDIUM
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Netherlands Win | 49.9% | 37.9% | 43.9% |
| Draw | 21.7% | 32.0% | 26.9% |
| Japan Win | 28.4% | 30.1% | 29.3% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Netherlands Elo: 1758
- Japan Elo: 1660
- P0: Netherlands 50.0% / Draw 22.2% / Japan 27.8%

### Likelihood Updates
- [away] Weighted recent form gap: home=0.54 vs away=1.00 (±5%) — audited recent form
- [home] Netherlands 4-2-3-1 lineup confirmed from manual; key starters: Frenkie de Jong, Guus Til, Marten de Roon (±6%) — lineup
- [away] Japan 4-2-3-1 lineup confirmed from manual; key starters: Doan Ritsu, Endo Wataru, Kamada Daichi (±6%) — lineup

### P1 (Final Bayesian)
- Netherlands: 49.9%  |  Draw: 21.7%  |  Japan: 28.4%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 1,000
- Avg Goals: Netherlands 1.14 / Japan 0.96 / Total 2.09
- Avg Shots: Netherlands 16.9 / Japan 17.3
- Conversion: Netherlands 6.7% / Japan 5.5%
- Avg Cards: 3.7

### Top Scorelines
- Netherlands 1-1 Japan: 15.0%
- Netherlands 0-0 Japan: 12.7%
- Netherlands 1-0 Japan: 12.5%
- Netherlands 0-1 Japan: 11.8%
- Netherlands 2-1 Japan: 8.1%

## Layer 3: Synthesis

**Confidence: MEDIUM** (max divergence: 12.0%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Bayesian evidence posterior and Monte Carlo simulation share the same baseline; equal weights are used by default.

### Divergence Analysis
- Bayesian is higher on home win by 12.0pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 10.3pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

### Consensus Estimate
- Netherlands: 43.9%  |  Draw: 26.9%  |  Japan: 29.3%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 3 天前

xiejin77 发表于 2026-6-14 14:07
# 2026 世界杯预测报告：Netherlands vs Japan

- **预测时间**: 2026-06-14 14:06:54

# 2026 世界杯预测报告：Sweden vs Tunisia

- **预测时间**: 2026-06-14 14:13:38
- **Run ID**: 20260614_141338
- **比赛信息**: 第 12 场 · F组
- **开球时间**: 2026-06-15 10:00 北京时间（原始：2026-06-14 20:00 local）
- **场馆**: Monterrey Stadium · Monterrey
- **贝叶斯数据更新时间**: 2026-06-14T14:13:15

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 39.6%，领先第二选项 4.7 个百分点。
- 综合置信度为高；模型分歧为 3.1%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 The Odds API，“主队胜”的模型概率低于市场 10.1 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-14T14:13:15。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Sweden 胜: 39.6%
- 平局: 25.6%
- Tunisia 胜: 34.9%
- 综合置信度: 高
- 方法分歧: 3.1%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Sweden 41.1% / 平局 25.4% / Tunisia 33.4%
- Elo: 1515 vs 1483
- audited recent form: Weighted recent form gap: home=0.66 vs away=0.39（幅度 3.0%）
- lineup: Sweden 4-3-3 lineup confirmed from manual; key starters: Besfort Zeneli, Jesper Kalstrom, Lucas Bergvall（幅度 6.5%）
- lineup: Tunisia 4-3-3 lineup confirmed from manual; key starters: Ismael Gharbi, Ellyes Skhiri, Hannibal Mejbri（幅度 6.5%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[form]: 最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。；置信度 0.55
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 原始伤停线索 6 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 1000
- 90分钟胜平负: Sweden 38.0% / 平局 25.7% / Tunisia 36.3%
- 平均进球: 1.22 - 1.15
- 模拟引擎: fm-style-chain-simulator (simulate_match -> simulate_attack_chain)
- 是否直接出结果: 否，逐场逐攻击链模拟
- 攻击链审计: 总计 41117 条，场均 41.1 条
- 射门/转化率: 总射门 35334 次，进球转化率 6.7%
- 稳定性审计: 中；单项 95% 误差上限约 ±3.0 个百分点
- 执行链路: sample starting XI players -> simulate possession-weighted attack windows -> resolve pass/dribble/shot duels with player attributes -> add set pieces, cards and penalties -> aggregate regulation probabilities across all simulations

### Monte Carlo Samples / 模拟样本轨迹

- 样本 1: 90分钟 1-2 / 最终 1-2；射门 22-19，攻击链 47，定位球链 12
- 样本 2: 90分钟 0-1 / 最终 0-1；射门 17-22，攻击链 47，定位球链 11
- 样本 3: 90分钟 2-1 / 最终 2-1；射门 20-23，攻击链 47，定位球链 9
- 样本 4: 90分钟 1-1 / 最终 1-1；射门 13-16，攻击链 36，定位球链 6
- 样本 5: 90分钟 2-0 / 最终 2-0；射门 12-21，攻击链 39，定位球链 9
- 稳定性提醒: Monte Carlo sample is usable, but close probabilities should be interpreted with the confidence interval.
- 90分钟结果: Sweden 38.0% / 平局 25.7% / Tunisia 36.3%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 920110718
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 27.2°C; 风速 8.9 km/h; 湿度 72%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Sweden: 4-3-3 / manual / confirmed / confidence 0.78
  - 首发: Jacob Widell Zetterstrom, Carl Starfelt, Daniel Svensson, Elliot Stroud Mjallby), Emil Holm, Besfort Zeneli, Jesper Kalstrom, Lucas Bergvall, Alexander Bernhardsson, Alexander Isak, Anthony Elanga
- 客队 Tunisia: 4-3-3 / manual / confirmed / confidence 0.78
  - 首发: Aymen Dahman, Yan Valery, Dylan Bronn, Montassar Talbi, Ali Abdi, Ismael Gharbi, Ellyes Skhiri, Hannibal Mejbri, Elias Achouri, Hazem Mastouri, Sebastian Tounekti

## Diagnosis / 综合诊断

- Bayesian and Monte Carlo agree closely under the shared baseline.

## Market Comparison / 外部市场概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Sweden vs Tunisia
- 市场更新时间: 2026-06-14T14:13:15
- 主队胜: 模型 39.6% / The Odds API 49.7% / 差值 -10.1pp
- 平局: 模型 25.6% / The Odds API 27.8% / 差值 -2.2pp
- 客队胜: 模型 34.9% / The Odds API 22.6% / 差值 +12.3pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Sweden vs Tunisia

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: HIGH
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Sweden Win | 41.1% | 38.0% | 39.6% |
| Draw | 25.4% | 25.7% | 25.6% |
| Tunisia Win | 33.4% | 36.3% | 34.9% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Sweden Elo: 1515
- Tunisia Elo: 1483
- P0: Sweden 40.5% / Draw 26.1% / Tunisia 33.4%

### Likelihood Updates
- [home] Weighted recent form gap: home=0.66 vs away=0.39 (±3%) — audited recent form
- [home] Sweden 4-3-3 lineup confirmed from manual; key starters: Besfort Zeneli, Jesper Kalstrom, Lucas Bergvall (±6%) — lineup
- [away] Tunisia 4-3-3 lineup confirmed from manual; key starters: Ismael Gharbi, Ellyes Skhiri, Hannibal Mejbri (±6%) — lineup

### P1 (Final Bayesian)
- Sweden: 41.1%  |  Draw: 25.4%  |  Tunisia: 33.4%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 1,000
- Avg Goals: Sweden 1.22 / Tunisia 1.15 / Total 2.37
- Avg Shots: Sweden 17.7 / Tunisia 17.6
- Conversion: Sweden 6.9% / Tunisia 6.5%
- Avg Cards: 3.7

### Top Scorelines
- Sweden 0-1 Tunisia: 11.6%
- Sweden 1-1 Tunisia: 11.5%
- Sweden 1-0 Tunisia: 11.5%
- Sweden 0-0 Tunisia: 9.7%
- Sweden 1-2 Tunisia: 8.2%

## Layer 3: Synthesis

**Confidence: HIGH** (max divergence: 3.1%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Bayesian evidence posterior and Monte Carlo simulation share the same baseline; equal weights are used by default.

### Divergence Analysis
- Bayesian and Monte Carlo agree closely under the shared baseline.

### Consensus Estimate
- Sweden: 39.6%  |  Draw: 25.6%  |  Tunisia: 34.9%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 3 天前

xiejin77 发表于 2026-6-13 13:54
# 2026 世界杯预测报告：Australia vs Türkiye

- **预测时间**: 2026-06-13 13:32:05

这个算法跑的，蒙特卡洛的土澳赢球标准很高啊。没想到结果居然对的上

作者: xiejin77 时间: 前天 13:01
# 2026 世界杯预测报告：Saudi Arabia vs Uruguay

- **预测时间**: 2026-06-15 13:00:45
- **Run ID**: 20260615_130045
- **比赛信息**: 第 13 场 · H组
- **开球时间**: 2026-06-16 06:00 北京时间（原始：2026-06-15 18:00 local）
- **场馆**: Miami Stadium · Miami
- **贝叶斯数据更新时间**: 2026-06-15T12:59:50

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“客队胜”，概率为 52.5%，领先第二选项 26.3 个百分点。
- 综合置信度为低；模型分歧为 28.4%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 Polymarket，“客队胜”的模型概率低于市场 12.0 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-15T12:59:50。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Saudi Arabia 胜: 26.2%
- 平局: 21.4%
- Uruguay 胜: 52.5%
- 综合置信度: 低
- 方法分歧: 28.4%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Saudi Arabia 17.5% / 平局 15.8% / Uruguay 66.7%
- Elo: 1450 vs 1673
- lineup: Saudi Arabia 4-3-3 lineup confirmed from manual; key starters: Abdullah Al Khaibari, Alaa ⁠Al Hejji, Ayman Yahya（幅度 6.5%）
- lineup: Uruguay 4-3-3 lineup confirmed from manual; key starters: Agustin Canobbio, Brian Rodriguez, Emiliano Martinez（幅度 6.5%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 原始伤停线索 7 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 10000
- 90分钟胜平负: Saudi Arabia 34.8% / 平局 26.9% / Uruguay 38.2%
- 平均进球: 1.17 - 1.24
- 90分钟结果: Saudi Arabia 34.8% / 平局 26.9% / Uruguay 38.2%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 1235697829
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 27.3°C; 风速 16 km/h; 湿度 78%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Saudi Arabia: 4-3-3 / manual / confirmed / confidence 0.78
  - 首发: Ahmed Alkassar, Abdulelah Al Amri, Ali Majrashi, Hassan Kadesh, Hassan Tambakti, Abdullah Al Khaibari, Alaa ⁠Al Hejji, Ayman Yahya, Abdullah Al Hamdan, Feras Al Brikan, Saleh Al Shehri
- 客队 Uruguay: 4-3-3 / manual / confirmed / confidence 0.78
  - 首发: Fernando Muslera, Guillermo Varela, Joaquin Piquerez, Jose Maria Gimenez, Mathias Olivera, Agustin Canobbio, Brian Rodriguez, Emiliano Martinez, Darwin Nunez, Facundo Pellistri, Federico Vinas

## Diagnosis / 综合诊断

- Monte Carlo is higher on home win by 17.4pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on draw by 11.1pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Bayesian is higher on away win by 28.4pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.

## Market Comparison / Polymarket 概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Saudi Arabia vs Uruguay
- 市场更新时间: 2026-06-15T12:59:50
- 主队胜: 模型 26.2% / Polymarket 13.2% / 差值 +13.0pp
- 平局: 模型 21.4% / Polymarket 22.3% / 差值 -1.0pp
- 客队胜: 模型 52.5% / Polymarket 64.5% / 差值 -12.0pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Saudi Arabia vs Uruguay

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Saudi Arabia Win | 17.5% | 34.8% | 26.2% |
| Draw | 15.8% | 26.9% | 21.4% |
| Uruguay Win | 66.7% | 38.2% | 52.5% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Saudi Arabia Elo: 1450
- Uruguay Elo: 1673
- P0: Saudi Arabia 17.4% / Draw 16.2% / Uruguay 66.4%

### Likelihood Updates
- [home] Saudi Arabia 4-3-3 lineup confirmed from manual; key starters: Abdullah Al Khaibari, Alaa ⁠Al Hejji, Ayman Yahya (±6%) — lineup
- [away] Uruguay 4-3-3 lineup confirmed from manual; key starters: Agustin Canobbio, Brian Rodriguez, Emiliano Martinez (±6%) — lineup

### P1 (Final Bayesian)
- Saudi Arabia: 17.5%  |  Draw: 15.8%  |  Uruguay: 66.7%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 10,000
- Avg Goals: Saudi Arabia 1.17 / Uruguay 1.24 / Total 2.41
- Avg Shots: Saudi Arabia 16.9 / Uruguay 18.4
- Conversion: Saudi Arabia 6.9% / Uruguay 6.8%
- Avg Cards: 3.6

### Top Scorelines
- Saudi Arabia 1-1 Uruguay: 12.8%
- Saudi Arabia 0-1 Uruguay: 10.7%
- Saudi Arabia 1-0 Uruguay: 10.5%
- Saudi Arabia 0-0 Uruguay: 8.6%
- Saudi Arabia 1-2 Uruguay: 8.5%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 28.4%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Monte Carlo is higher on home win by 17.4pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on draw by 11.1pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Bayesian is higher on away win by 28.4pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.

### Consensus Estimate
- Saudi Arabia: 26.2%  |  Draw: 21.4%  |  Uruguay: 52.5%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 前天 13:04
# 2026 世界杯预测报告：Spain vs Cabo Verde

- **预测时间**: 2026-06-15 13:03:00
- **Run ID**: 20260615_130300
- **比赛信息**: 第 14 场 · H组
- **开球时间**: 2026-06-16 00:00 北京时间（原始：2026-06-15 12:00 local）
- **场馆**: Atlanta Stadium · Atlanta
- **贝叶斯数据更新时间**: 2026-06-15T13:02:23

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 64.7%，领先第二选项 46.3 个百分点。
- 综合置信度为低；模型分歧为 42.1%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 Polymarket，“主队胜”的模型概率低于市场 24.5 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-15T13:02:23。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Spain 胜: 64.7%
- 平局: 18.5%
- Cabo Verde 胜: 16.8%
- 综合置信度: 低
- 方法分歧: 42.1%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Spain 85.8% / 平局 9.6% / Cabo Verde 4.6%
- Elo: 1876 vs 1390
- audited recent form: Weighted recent form gap: home=0.60 vs away=0.42（幅度 2.0%）
- lineup: Spain 4-3-3 lineup confirmed from manual; key starters: Gavi, Alex Baena, Fabian Ruiz（幅度 6.5%）
- lineup: Cape Verde 4-2-3-1 lineup confirmed from manual; key starters: Jamiro Monteiro, Kevin Pina, Dailon Livramento（幅度 6.5%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[form]: 最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。；置信度 0.55
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 原始伤停线索 8 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 10000
- 90分钟胜平负: Spain 43.7% / 平局 27.3% / Cabo Verde 29.0%
- 平均进球: 1.31 - 1.00
- 90分钟结果: Spain 43.7% / 平局 27.3% / Cabo Verde 29.0%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 1235697829
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 25.9°C; 风速 16 km/h; 湿度 57%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Spain: 4-3-3 / manual / confirmed / confidence 0.78
  - 首发: David Raya, Alejandro Grimaldo, Aymeric Laporte, Eric Garcia, Marc Cucurella, Alex Baena, Fabian Ruiz, Gavi, Borja Iglesias, Dani Olmo, Ferran Torres
- 客队 Cape Verde: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Carlos Santos, Edilson Borges “Diney”, Ianique Tavares “Stopira”, Joao Paulo Fernandes, Kelvin Pires, Deroy Duarte, Jamiro Monteiro, Dailon Livramento, Kevin Pina, Garry Rodrigues, Gilson Benchimol

## Diagnosis / 综合诊断

- Bayesian is higher on home win by 42.1pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 17.7pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on away win by 24.4pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

## Market Comparison / Polymarket 概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Spain vs Cabo Verde
- 市场更新时间: 2026-06-15T13:02:23
- 主队胜: 模型 64.7% / Polymarket 89.2% / 差值 -24.5pp
- 平局: 模型 18.5% / Polymarket 7.6% / 差值 +10.8pp
- 客队胜: 模型 16.8% / Polymarket 3.1% / 差值 +13.7pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Spain vs Cape Verde

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Spain Win | 85.8% | 43.7% | 64.7% |
| Draw | 9.6% | 27.3% | 18.5% |
| Cape Verde Win | 4.6% | 29.0% | 16.8% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Spain Elo: 1876
- Cape Verde Elo: 1390
- P0: Spain 85.5% / Draw 9.9% / Cape Verde 4.6%

### Likelihood Updates
- [home] Weighted recent form gap: home=0.60 vs away=0.42 (±2%) — audited recent form
- [home] Spain 4-3-3 lineup confirmed from manual; key starters: Gavi, Alex Baena, Fabian Ruiz (±6%) — lineup
- [away] Cape Verde 4-2-3-1 lineup confirmed from manual; key starters: Jamiro Monteiro, Kevin Pina, Dailon Livramento (±6%) — lineup

### P1 (Final Bayesian)
- Spain: 85.8%  |  Draw: 9.6%  |  Cape Verde: 4.6%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 10,000
- Avg Goals: Spain 1.31 / Cape Verde 1.00 / Total 2.31
- Avg Shots: Spain 19.3 / Cape Verde 14.7
- Conversion: Spain 6.8% / Cape Verde 6.8%
- Avg Cards: 3.7

### Top Scorelines
- Spain 1-1 Cape Verde: 13.0%
- Spain 1-0 Cape Verde: 12.6%
- Spain 0-1 Cape Verde: 10.0%
- Spain 0-0 Cape Verde: 9.6%
- Spain 2-0 Cape Verde: 9.0%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 42.1%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Bayesian is higher on home win by 42.1pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 17.7pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on away win by 24.4pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

### Consensus Estimate
- Spain: 64.7%  |  Draw: 18.5%  |  Cape Verde: 16.8%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 前天 13:06
# 2026 世界杯预测报告：Belgium vs Egypt

- **预测时间**: 2026-06-15 13:06:34
- **Run ID**: 20260615_130634
- **比赛信息**: 第 16 场 · G组
- **开球时间**: 2026-06-16 03:00 北京时间（原始：2026-06-15 12:00 local）
- **场馆**: Seattle Stadium · Seattle
- **贝叶斯数据更新时间**: 2026-06-15T13:06:09

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 51.2%，领先第二选项 24.8 个百分点。
- 综合置信度为低；模型分歧为 18.1%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 Polymarket，“主队胜”的模型概率低于市场 8.0 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-15T13:06:09。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Belgium 胜: 51.2%
- 平局: 22.3%
- Egypt 胜: 26.4%
- 综合置信度: 低
- 方法分歧: 18.1%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Belgium 60.3% / 平局 18.1% / Egypt 21.6%
- Elo: 1735 vs 1566
- audited recent form: Weighted recent form gap: home=0.73 vs away=0.53（幅度 2.1%）
- lineup: Belgium 4-2-3-1 lineup confirmed from manual; key starters: Kevin De Bruyne, Youri Tielemans, Maxim De Cuyper（幅度 6.5%）
- lineup: Egypt 4-2-3-1 lineup confirmed from manual; key starters: Ahmed Zizo, Emam Ashour, Haissem Hassan（幅度 6.5%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[form]: 最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。；置信度 0.55
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 原始伤停线索 6 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 10000
- 90分钟胜平负: Belgium 42.2% / 平局 26.6% / Egypt 31.2%
- 平均进球: 1.33 - 1.10
- 90分钟结果: Belgium 42.2% / 平局 26.6% / Egypt 31.2%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 1235697829
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 26.3°C; 风速 6.7 km/h; 湿度 38%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Belgium: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Thibaut Courtois, Timothy Castagne, Zeno Debast, Arthur Theate, Maxim De Cuyper, Youri Tielemans, Amadou Onana, Jeremy Doku, Kevin De Bruyne, Leandro Trossard, Charles De Ketelaere
- 客队 Egypt: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Mohamed Alaa, Ahmed Fotouh, Hamdy Fathy, Hossam Abdelmaguid, Karim Hafez, Ahmed Zizo, Emam Ashour, Aqtay Abdallah, Haissem Hassan, Hamza Abdelkarim, Mohamed Salah

## Diagnosis / 综合诊断

- Bayesian is higher on home win by 18.1pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 8.5pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

## Market Comparison / Polymarket 概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Belgium vs Egypt
- 市场更新时间: 2026-06-15T13:06:09
- 主队胜: 模型 51.2% / Polymarket 59.2% / 差值 -8.0pp
- 平局: 模型 22.3% / Polymarket 24.2% / 差值 -1.9pp
- 客队胜: 模型 26.4% / Polymarket 16.5% / 差值 +9.9pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Belgium vs Egypt

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Belgium Win | 60.3% | 42.2% | 51.2% |
| Draw | 18.1% | 26.6% | 22.3% |
| Egypt Win | 21.6% | 31.2% | 26.4% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Belgium Elo: 1735
- Egypt Elo: 1566
- P0: Belgium 59.8% / Draw 18.6% / Egypt 21.7%

### Likelihood Updates
- [home] Weighted recent form gap: home=0.73 vs away=0.53 (±2%) — audited recent form
- [home] Belgium 4-2-3-1 lineup confirmed from manual; key starters: Kevin De Bruyne, Youri Tielemans, Maxim De Cuyper (±6%) — lineup
- [away] Egypt 4-2-3-1 lineup confirmed from manual; key starters: Ahmed Zizo, Emam Ashour, Haissem Hassan (±6%) — lineup

### P1 (Final Bayesian)
- Belgium: 60.3%  |  Draw: 18.1%  |  Egypt: 21.6%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 10,000
- Avg Goals: Belgium 1.33 / Egypt 1.10 / Total 2.44
- Avg Shots: Belgium 18.3 / Egypt 16.6
- Conversion: Belgium 7.3% / Egypt 6.7%
- Avg Cards: 3.6

### Top Scorelines
- Belgium 1-1 Egypt: 12.5%
- Belgium 1-0 Egypt: 11.5%
- Belgium 0-1 Egypt: 9.9%
- Belgium 2-1 Egypt: 8.9%
- Belgium 0-0 Egypt: 8.5%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 18.1%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Bayesian is higher on home win by 18.1pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 8.5pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

### Consensus Estimate
- Belgium: 51.2%  |  Draw: 22.3%  |  Egypt: 26.4%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 昨天 08:26
伊朗对新西兰的这一场，不知道为什么很多数据源的数据都不齐，我也懒得手工检索填入。只能作罢。

作者: xiejin77 时间: 昨天 09:30
# 2026 世界杯预测报告：France vs Senegal

- **预测时间**: 2026-06-16 09:29:36
- **Run ID**: 20260616_092936
- **比赛信息**: 第 17 场 · I组
- **开球时间**: 2026-06-16 15:00 local（原始：2026-06-16 15:00 local）
- **场馆**: New York/New Jersey Stadium · New Jersey
- **贝叶斯数据更新时间**: 2026-06-16T09:26:22

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 50.2%，领先第二选项 22.8 个百分点。
- 综合置信度为低；模型分歧为 23.9%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 Polymarket，“主队胜”的模型概率低于市场 13.8 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-16T09:26:22。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- France 胜: 50.2%
- 平局: 27.4%
- Senegal 胜: 22.5%
- 综合置信度: 低
- 方法分歧: 23.9%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: France 62.1% / 平局 17.8% / Senegal 20.1%
- Elo: 1877 vs 1689
- lineup: France 4-2-3-1 lineup confirmed from manual; key starters: Adrien Rabiot, Aurelien Tchouameni, Manu Kone（幅度 6.5%）
- lineup: Senegal 4-3-3 lineup confirmed from manual; key starters: Ilay Camara, Bara Sapoko Ndiaye, Habib Diarra（幅度 6.5%）
- simulation factors: Weather condition: rain（幅度 6.0%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[weather]: 雨雪会降低传控和推进稳定性，贝叶斯侧只轻微提高平局/低比分倾向。；置信度 0.55
- 原始伤停线索 2 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 10000
- 90分钟胜平负: France 38.2% / 平局 37.0% / Senegal 24.8%
- 平均进球: 0.84 - 0.60
- 90分钟结果: France 38.2% / 平局 37.0% / Senegal 24.8%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 1309429543
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 雨天; 温度 26.2°C; 风速 15.9 km/h; 湿度 94%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 雨雪天气：传球、盘带和射门稳定性下降，比赛节奏降低。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 France: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Brice Samba, Dayot Upamecano, Ibrahima Konate, Jules Kounde, Lucas Digne, Adrien Rabiot, Aurelien Tchouameni, Bradley Barcola, Manu Kone, Desire Doue, Jean-Philippe Mateta
- 客队 Senegal: 4-3-3 / manual / confirmed / confidence 0.78
  - 首发: Edouard Mendy, Abdoulaye Seck, Antoine Mendy, El Hadji Malick Diouf, Ilay Camara, Bara Sapoko Ndiaye, Habib Diarra, Idrissa Gana Gueye, Assane Diao, Bamba Dieng, Cherif Ndiaye

## Diagnosis / 综合诊断

- Bayesian is higher on home win by 23.9pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 19.2pp. Likely generative simulation effect from squad duels, tempo, or factors (1 factor effects).

## Market Comparison / Polymarket 概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: France vs Senegal
- 市场更新时间: 2026-06-16T09:26:22
- 主队胜: 模型 50.2% / Polymarket 64.0% / 差值 -13.8pp
- 平局: 模型 27.4% / Polymarket 21.6% / 差值 +5.7pp
- 客队胜: 模型 22.5% / Polymarket 14.4% / 差值 +8.1pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: France vs Senegal

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| France Win | 62.1% | 38.2% | 50.2% |
| Draw | 17.8% | 37.0% | 27.4% |
| Senegal Win | 20.1% | 24.8% | 22.5% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- France Elo: 1877
- Senegal Elo: 1689
- P0: France 62.2% / Draw 17.7% / Senegal 20.1%

### Likelihood Updates
- [home] France 4-2-3-1 lineup confirmed from manual; key starters: Adrien Rabiot, Aurelien Tchouameni, Manu Kone (±6%) — lineup
- [away] Senegal 4-3-3 lineup confirmed from manual; key starters: Ilay Camara, Bara Sapoko Ndiaye, Habib Diarra (±6%) — lineup
- [draw] Weather condition: rain (±6%) — simulation factors

### P1 (Final Bayesian)
- France: 62.1%  |  Draw: 17.8%  |  Senegal: 20.1%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 10,000
- Avg Goals: France 0.84 / Senegal 0.60 / Total 1.44
- Avg Shots: France 16.1 / Senegal 12.2
- Conversion: France 5.2% / Senegal 4.9%
- Avg Cards: 3.6

### Top Scorelines
- France 0-0 Senegal: 23.6%
- France 1-0 Senegal: 19.4%
- France 0-1 Senegal: 14.4%
- France 1-1 Senegal: 12.0%
- France 2-0 Senegal: 8.6%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 23.9%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Bayesian is higher on home win by 23.9pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 19.2pp. Likely generative simulation effect from squad duels, tempo, or factors (1 factor effects).

### Consensus Estimate
- France: 50.2%  |  Draw: 27.4%  |  Senegal: 22.5%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 昨天 09:44
# 2026 世界杯预测报告：Iraq vs Norway

- **预测时间**: 2026-06-16 09:44:18
- **Run ID**: 20260616_094418
- **比赛信息**: 第 18 场 · I组
- **开球时间**: 2026-06-17 06:00 北京时间（原始：2026-06-16 18:00 local）
- **场馆**: Boston Stadium · Boston
- **贝叶斯数据更新时间**: 2026-06-16T09:43:35

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“客队胜”，概率为 44.5%，领先第二选项 14.2 个百分点。
- 综合置信度为中；模型分歧为 12.1%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 Polymarket，“客队胜”的模型概率低于市场 35.2 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-16T09:43:35。
- 阵容数据存在未确认或未匹配项，建议先在阵容弹窗中手动修正首发名单。

## Synthesis / 综合概率

- Iraq 胜: 30.3%
- 平局: 25.2%
- Norway 胜: 44.5%
- 综合置信度: 中
- 方法分歧: 12.1%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Iraq 27.4% / 平局 22.1% / Norway 50.5%
- Elo: 1455 vs 1551
- audited recent form: Weighted recent form gap: home=0.51 vs away=0.72（幅度 2.4%）
- lineup: Norway 4-3-3 lineup confirmed from expert; key starters: Kristian Thorstvedt, Patrick Berg, Sander Berge（幅度 3.8%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[form]: 最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。；置信度 0.55
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.62
- 原始伤停线索 4 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 10000
- 90分钟胜平负: Iraq 33.2% / 平局 28.3% / Norway 38.5%
- 平均进球: 1.12 - 1.22
- 90分钟结果: Iraq 33.2% / 平局 28.3% / Norway 38.5%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 1309429543
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 23.8°C; 风速 13 km/h; 湿度 40%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Iraq: 4-2-3-1 / expert / predicted / confidence 0.74
  - 首发: Jalal Hassan, Hussein Ali, Rebin Sulaka, Zaid Tahseen, Merchas Doski, Amir Al-Ammari, Ibrahim Bayesh, Ali Jassim, Zidane Iqbal, Youssef Amyn, Aymen Hussein
- 客队 Norway: 4-3-3 / expert / predicted / confidence 0.74
  - 首发: Orjan Nyland, Julian Ryerson, Torbjorn Heggem, Leo Skiri Ostigard, David Møller Wolfe, Kristian Thorstvedt, Patrick Berg, Sander Berge, Alexander Sorloth, Erling Haaland, Antonio Nusa
- 阵容提醒: Iraq: Al-Ammari fuzzy matched to Amir Al-Ammari; please review.；Iraq lineup needs manual review.

## Diagnosis / 综合诊断

- Bayesian is higher on away win by 12.1pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.

## Market Comparison / Polymarket 概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Iraq vs Norway
- 市场更新时间: 2026-06-16T09:43:35
- 主队胜: 模型 30.3% / Polymarket 6.6% / 差值 +23.7pp
- 平局: 模型 25.2% / Polymarket 13.7% / 差值 +11.5pp
- 客队胜: 模型 44.5% / Polymarket 79.7% / 差值 -35.2pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Iraq vs Norway

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: MEDIUM
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Iraq Win | 27.4% | 33.2% | 30.3% |
| Draw | 22.1% | 28.3% | 25.2% |
| Norway Win | 50.5% | 38.5% | 44.5% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Iraq Elo: 1455
- Norway Elo: 1551
- P0: Iraq 27.9% / Draw 22.3% / Norway 49.7%

### Likelihood Updates
- [away] Weighted recent form gap: home=0.51 vs away=0.72 (±2%) — audited recent form
- [away] Norway 4-3-3 lineup confirmed from expert; key starters: Kristian Thorstvedt, Patrick Berg, Sander Berge (±4%) — lineup

### P1 (Final Bayesian)
- Iraq: 27.4%  |  Draw: 22.1%  |  Norway: 50.5%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 10,000
- Avg Goals: Iraq 1.12 / Norway 1.22 / Total 2.34
- Avg Shots: Iraq 16.8 / Norway 17.9
- Conversion: Iraq 6.7% / Norway 6.8%
- Avg Cards: 3.6

### Top Scorelines
- Iraq 1-1 Norway: 13.3%
- Iraq 0-1 Norway: 11.6%
- Iraq 1-0 Norway: 11.1%
- Iraq 0-0 Norway: 9.3%
- Iraq 1-2 Norway: 8.3%

## Layer 3: Synthesis

**Confidence: MEDIUM** (max divergence: 12.1%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Bayesian evidence posterior and Monte Carlo simulation share the same baseline; equal weights are used by default.

### Divergence Analysis
- Bayesian is higher on away win by 12.1pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.

### Consensus Estimate
- Iraq: 30.3%  |  Draw: 25.2%  |  Norway: 44.5%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 昨天 09:47
# 2026 世界杯预测报告：Argentina vs Algeria

- **预测时间**: 2026-06-16 09:46:46
- **Run ID**: 20260616_094646
- **比赛信息**: 第 19 场 · J组
- **开球时间**: 2026-06-17 09:00 北京时间（原始：2026-06-16 20:00 local）
- **场馆**: Kansas City Stadium · Kansas City
- **贝叶斯数据更新时间**: 2026-06-16T09:46:05

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 57.6%，领先第二选项 36.0 个百分点。
- 综合置信度为低；模型分歧为 36.4%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 当前没有可标准化的 Polymarket 胜平负对照，市场概率不进入模型，只作为外部参考。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-16T09:46:05。
- 阵容数据存在未确认或未匹配项，建议先在阵容弹窗中手动修正首发名单。

## Synthesis / 综合概率

- Argentina 胜: 57.6%
- 平局: 20.8%
- Algeria 胜: 21.6%
- 综合置信度: 低
- 方法分歧: 36.4%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Argentina 75.8% / 平局 12.9% / Algeria 11.2%
- Elo: 1875 vs 1564
- audited recent form: Weighted recent form gap: home=1.00 vs away=0.66（幅度 3.7%）
- lineup: Argentina 4-3-3 lineup confirmed from expert; key starters: Alexis Mac Allister, Enzo Fernandez, Leandro Paredes（幅度 3.8%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[form]: 最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。；置信度 0.55
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.62
- 原始伤停线索 7 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 10000
- 90分钟胜平负: Argentina 39.4% / 平局 28.6% / Algeria 32.0%
- 平均进球: 1.20 - 1.04
- 90分钟结果: Argentina 39.4% / 平局 28.6% / Algeria 32.0%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 1309429543
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 26.2°C; 风速 3.5 km/h; 湿度 29%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Argentina: 4-3-3 / expert / predicted / confidence 0.74
  - 首发: Emiliano Martinez, Nahuel Molina, Nicolas Otamendi, Cristian Romero, Nicolas Tagliafico, Alexis Mac Allister, Leandro Paredes, Enzo Fernandez, Lionel Messi, Julian Alvarez, Thiago Almada
- 客队 Algeria: 4-2-3-1 / expert / predicted / confidence 0.74
  - 首发: Luca Zidane, Rafik Belghali, Zinedine Belaid, Ramy Bensebaini, Rayan Ait-Nouri, Ramiz Zerrouki, Hicham Boudaoui, Riyad Mahrez, Houssem Aouar, Fares Chaibi, Amine Gouiri
- 阵容提醒: Algeria: Ait-Nouri fuzzy matched to Rayan Ait-Nouri; please review.；Algeria lineup needs manual review.

## Diagnosis / 综合诊断

- Bayesian is higher on home win by 36.4pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 15.7pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on away win by 20.8pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

## Market Comparison / Polymarket 概率对照

暂无 Polymarket 对照。

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Argentina vs Algeria

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Argentina Win | 75.8% | 39.4% | 57.6% |
| Draw | 12.9% | 28.6% | 20.8% |
| Algeria Win | 11.2% | 32.0% | 21.6% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Argentina Elo: 1875
- Algeria Elo: 1564
- P0: Argentina 75.1% / Draw 13.2% / Algeria 11.6%

### Likelihood Updates
- [home] Weighted recent form gap: home=1.00 vs away=0.66 (±4%) — audited recent form
- [home] Argentina 4-3-3 lineup confirmed from expert; key starters: Alexis Mac Allister, Enzo Fernandez, Leandro Paredes (±4%) — lineup

### P1 (Final Bayesian)
- Argentina: 75.8%  |  Draw: 12.9%  |  Algeria: 11.2%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 10,000
- Avg Goals: Argentina 1.20 / Algeria 1.04 / Total 2.24
- Avg Shots: Argentina 18.3 / Algeria 16.0
- Conversion: Argentina 6.6% / Algeria 6.5%
- Avg Cards: 3.6

### Top Scorelines
- Argentina 1-1 Algeria: 13.4%
- Argentina 1-0 Algeria: 12.4%
- Argentina 0-1 Algeria: 11.5%
- Argentina 0-0 Algeria: 10.2%
- Argentina 2-1 Algeria: 8.2%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 36.4%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Bayesian is higher on home win by 36.4pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 15.7pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on away win by 20.8pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

### Consensus Estimate
- Argentina: 57.6%  |  Draw: 20.8%  |  Algeria: 21.6%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 昨天 09:53
# 2026 世界杯预测报告：Austria vs Jordan

- **预测时间**: 2026-06-16 09:52:47
- **Run ID**: 20260616_095247
- **比赛信息**: 第 20 场 · J组
- **开球时间**: 2026-06-17 12:00 北京时间（原始：2026-06-16 21:00 local）
- **场馆**: San Francisco Bay Area Stadium · San Francisco Bay Area
- **贝叶斯数据更新时间**: 2026-06-16T09:52:26

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 49.9%，领先第二选项 22.5 个百分点。
- 综合置信度为低；模型分歧为 21.9%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 当前没有可标准化的 Polymarket 胜平负对照，市场概率不进入模型，只作为外部参考。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-16T09:52:26。
- 阵容数据存在未确认或未匹配项，建议先在阵容弹窗中手动修正首发名单。

## Synthesis / 综合概率

- Austria 胜: 49.9%
- 平局: 22.7%
- Jordan 胜: 27.4%
- 综合置信度: 低
- 方法分歧: 21.9%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Austria 60.9% / 平局 18.2% / Jordan 20.9%
- Elo: 1593 vs 1420
- audited recent form: Weighted recent form gap: home=0.83 vs away=0.32（幅度 5.7%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[form]: 最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。；置信度 0.55
- 原始伤停线索 5 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 10000
- 90分钟胜平负: Austria 38.9% / 平局 27.3% / Jordan 33.8%
- 平均进球: 1.24 - 1.14
- 90分钟结果: Austria 38.9% / 平局 27.3% / Jordan 33.8%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 1309429543
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 16°C; 风速 10.4 km/h; 湿度 82%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Austria: 4-2-3-1 / roster-default / needs-confirmation / confidence 0.28
  - 首发: Alexander Schlager, Alexander Prass, David Affengruber, David Alaba, Kevin Danso, Alessandro Schöpf, Carney Chukwuemeka, Christoph Baumgartner, Florian Grillitsch, Konrad Laimer, Marko Arnautovic
- 客队 Jordan: 5-3-2 / roster-default / needs-confirmation / confidence 0.28
  - 首发: Abdallah Al Fakhouri, Abdallah Nasib, Anas Banawi, Ehsan Haddad, Husam Abu Dahab, Mohammad Abu Hashish, Amer Jamous, Ibrahim Sadeh, Mahmoud Al-Mardi ., Ali Azaizeh, Ali Olwan
- 阵容提醒: Austria: no reliable predicted/recent XI available; using roster-default starters by formation slots. Please confirm manually.；External formation source was not parsed; using local tactical profile fallback.；Austria lineup needs manual review.；Jordan: no reliable predicted/recent XI available; using roster-default starters by formation slots. Please confirm manually.；External formation source was not parsed; using local tactical profile fallback.；Jordan lineup needs manual review.

## Diagnosis / 综合诊断

- Bayesian is higher on home win by 21.9pp. Likely evidence-layer effect (1 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 9.1pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on away win by 12.9pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

## Market Comparison / Polymarket 概率对照

暂无 Polymarket 对照。

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Austria vs Jordan

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Austria Win | 60.9% | 38.9% | 49.9% |
| Draw | 18.2% | 27.3% | 22.7% |
| Jordan Win | 20.9% | 33.8% | 27.4% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Austria Elo: 1593
- Jordan Elo: 1420
- P0: Austria 60.3% / Draw 18.4% / Jordan 21.3%

### Likelihood Updates
- [home] Weighted recent form gap: home=0.83 vs away=0.32 (±6%) — audited recent form

### P1 (Final Bayesian)
- Austria: 60.9%  |  Draw: 18.2%  |  Jordan: 20.9%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 10,000
- Avg Goals: Austria 1.24 / Jordan 1.14 / Total 2.38
- Avg Shots: Austria 18.2 / Jordan 16.9
- Conversion: Austria 6.8% / Jordan 6.7%
- Avg Cards: 3.6

### Top Scorelines
- Austria 1-1 Jordan: 13.5%
- Austria 1-0 Jordan: 11.8%
- Austria 0-1 Jordan: 10.6%
- Austria 0-0 Jordan: 8.5%
- Austria 2-1 Jordan: 8.4%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 21.9%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Bayesian is higher on home win by 21.9pp. Likely evidence-layer effect (1 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 9.1pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on away win by 12.9pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).

### Consensus Estimate
- Austria: 49.9%  |  Draw: 22.7%  |  Jordan: 27.4%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 2 小时前
# 2026 世界杯预测报告：Ghana vs Panama

- **预测时间**: 2026-06-17 16:03:50
- **Run ID**: 20260617_160350
- **比赛信息**: 第 21 场 · L组
- **开球时间**: 2026-06-18 07:00 北京时间（原始：2026-06-17 19:00 local）
- **场馆**: Toronto Stadium · Toronto
- **贝叶斯数据更新时间**: 2026-06-17T16:03:39

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“客队胜”，概率为 45.5%，领先第二选项 14.1 个百分点。
- 综合置信度为低；模型分歧为 25.0%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 Polymarket，“客队胜”的模型概率高于市场 16.0 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-17T16:03:39。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Ghana 胜: 31.4%
- 平局: 23.0%
- Panama 胜: 45.5%
- 综合置信度: 低
- 方法分歧: 25.0%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Ghana 24.0% / 平局 17.9% / Panama 58.1%
- Elo: 1370 vs 1541
- mapped team news: Adalberto Carrasquilla injury mapped (major)（幅度 13.0%）
- audited recent form: Weighted recent form gap: home=0.10 vs away=0.50（幅度 4.4%）
- lineup: Ghana 4-3-3 lineup confirmed from manual; key starters: Augustine Boakye, Caleb Yirenkyi, Abdul Fatawu Issahaku（幅度 6.5%）
- lineup: Panama 4-4-2 lineup confirmed from manual; key starters: Adalberto Carrasquilla, Alberto Quintero, Anibal Godoy（幅度 6.5%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[injury]: 伤停已映射到球队球员，且严重度明确；因此以保守权重影响对手方向。；置信度 0.85
- 入模[form]: 最近五场加权状态差距达到阈值；已考虑赛事类型、时间衰减、对手强度和主客/中立场，作为低权重贝叶斯证据入模。；置信度 0.55
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 原始伤停线索 8 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 10000
- 90分钟胜平负: Ghana 38.9% / 平局 28.1% / Panama 33.0%
- 平均进球: 1.15 - 1.05
- 90分钟结果: Ghana 38.9% / 平局 28.1% / Panama 33.0%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 1309429543
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 21°C; 风速 15.2 km/h; 湿度 53%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- Panama: {'headline': '2026 World Cup: Panama star Adalberto Carrasquilla could miss opener against Ghana due to injury', 'team': 'Panama', 'sourceType': 'google-news-raw', 'status': 'candidate', 'applied': False, 'warnings': [], 'matchedPlayers': ['adalberto carrasquilla'], 'confidence': 0.72} -> 命中 Adalberto Carrasquilla，明确缺阵/重罚。

## Lineups / 阵容来源与匹配状态

- 主队 Ghana: 4-3-3 / manual / confirmed / confidence 0.78
  - 首发: Benjamin Asare, Abdul Mumin, Alidu Seidu, Baba Abdul Rahman, Derrick Luckassen, Abdul Fatawu Issahaku, Augustine Boakye, Caleb Yirenkyi, Antoine Semenyo, Brandon Thomas-Asante, Christopher Bonsu Baah
- 客队 Panama: 4-4-2 / manual / confirmed / confidence 0.78
  - 首发: Cesar Samudio, Amir Murillo, Andres Andrade, Cesar Blackman, Edgardo Farina, Adalberto Carrasquilla, Alberto Quintero, Anibal Godoy, Azarias Londono, Cecilio Waterman, Ismael Diaz

## Diagnosis / 综合诊断

- Monte Carlo is higher on home win by 14.9pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on draw by 10.2pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Bayesian is higher on away win by 25.0pp. Likely evidence-layer effect (4 evidence items) rather than player-chain simulation.

## Market Comparison / Polymarket 概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: Ghana vs Panama
- 市场更新时间: 2026-06-17T16:03:39
- 主队胜: 模型 31.4% / Polymarket 40.6% / 差值 -9.1pp
- 平局: 模型 23.0% / Polymarket 29.9% / 差值 -6.9pp
- 客队胜: 模型 45.5% / Polymarket 29.6% / 差值 +16.0pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Ghana vs Panama

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.83

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Ghana Win | 24.0% | 38.9% | 31.4% |
| Draw | 17.9% | 28.1% | 23.0% |
| Panama Win | 58.1% | 33.0% | 45.5% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Ghana Elo: 1370
- Panama Elo: 1541
- P0: Ghana 21.5% / Draw 18.5% / Panama 60.0%

### Likelihood Updates
- [home] Adalberto Carrasquilla injury mapped (major) (±13%) — mapped team news
- [away] Weighted recent form gap: home=0.10 vs away=0.50 (±4%) — audited recent form
- [home] Ghana 4-3-3 lineup confirmed from manual; key starters: Augustine Boakye, Caleb Yirenkyi, Abdul Fatawu Issahaku (±6%) — lineup
- [away] Panama 4-4-2 lineup confirmed from manual; key starters: Adalberto Carrasquilla, Alberto Quintero, Anibal Godoy (±6%) — lineup

### P1 (Final Bayesian)
- Ghana: 24.0%  |  Draw: 17.9%  |  Panama: 58.1%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 10,000
- Avg Goals: Ghana 1.15 / Panama 1.05 / Total 2.20
- Avg Shots: Ghana 16.8 / Panama 15.6
- Conversion: Ghana 6.8% / Panama 6.7%
- Avg Cards: 3.7

### Top Scorelines
- Ghana 1-1 Panama: 13.1%
- Ghana 1-0 Panama: 13.1%
- Ghana 0-1 Panama: 11.8%
- Ghana 0-0 Panama: 10.2%
- Ghana 2-1 Panama: 8.6%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 25.0%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Monte Carlo is higher on home win by 14.9pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on draw by 10.2pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Bayesian is higher on away win by 25.0pp. Likely evidence-layer effect (4 evidence items) rather than player-chain simulation.

### Consensus Estimate
- Ghana: 31.4%  |  Draw: 23.0%  |  Panama: 45.5%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

这场算法跑下来非常看好巴拿马，不知道为啥。

作者: xiejin77 时间: 2 小时前
# 2026 世界杯预测报告：England vs Croatia

- **预测时间**: 2026-06-17 16:18:24
- **Run ID**: 20260617_161824
- **比赛信息**: 第 22 场 · L组
- **开球时间**: 2026-06-18 04:00 北京时间（原始：2026-06-17 15:00 local）
- **场馆**: Dallas Stadium · Dallas
- **贝叶斯数据更新时间**: 2026-06-17T16:17:36

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 42.8%，领先第二选项 13.5 个百分点。
- 综合置信度为低；模型分歧为 17.4%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 对照 Polymarket，“主队胜”的模型概率低于市场 12.6 个百分点，可用于判断市场是否偏热或偏冷。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-17T16:17:36。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- England 胜: 42.8%
- 平局: 27.8%
- Croatia 胜: 29.3%
- 综合置信度: 低
- 方法分歧: 17.4%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: England 51.5% / 平局 21.7% / Croatia 26.8%
- Elo: 1826 vs 1717
- lineup: England 4-2-3-1 lineup confirmed from manual; key starters: Declan Rice, Eberechi Eze, Elliot Anderson（幅度 6.5%）
- lineup: Croatia 4-2-3-1 lineup confirmed from manual; key starters: Luka Sucic, Luka Modric, Mario Pasalic（幅度 6.5%）
- simulation factors: Weather condition: rain（幅度 6.0%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[weather]: 雨雪会降低传控和推进稳定性，贝叶斯侧只轻微提高平局/低比分倾向。；置信度 0.55
- 原始伤停线索 9 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 10000
- 90分钟胜平负: England 34.1% / 平局 33.9% / Croatia 31.9%
- 平均进球: 0.87 - 0.82
- 90分钟结果: England 34.1% / 平局 33.9% / Croatia 31.9%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 1309429543
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 雨天; 温度 32.1°C; 风速 22.1 km/h; 湿度 61%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 雨雪天气：传球、盘带和射门稳定性下降，比赛节奏降低。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 England: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Dean Henderson, Dan Burn, Djed Spence, Ezri Konsa, Jarell Quansah, Declan Rice, Eberechi Eze, Anthony Gordon, Elliot Anderson, Bukayo Saka, Harry Kane
- 客队 Croatia: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Dominik Livakovic, Josip Stanisic, Josip Sutalo, Duje Caleta-Car, Josko Gvardiol, Luka Sucic, Luka Modric, Mario Pasalic, Andrej Kramaric, Ivan Perisic, Ante Budimir

## Diagnosis / 综合诊断

- Bayesian is higher on home win by 17.4pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 12.2pp. Likely generative simulation effect from squad duels, tempo, or factors (1 factor effects).

## Market Comparison / Polymarket 概率对照

- 市场来源: The Odds API
- 市场问题: Bookmaker h2h odds: England vs Croatia
- 市场更新时间: 2026-06-17T16:17:36
- 主队胜: 模型 42.8% / Polymarket 55.4% / 差值 -12.6pp
- 平局: 模型 27.8% / Polymarket 25.6% / 差值 +2.2pp
- 客队胜: 模型 29.3% / Polymarket 19.0% / 差值 +10.4pp

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: England vs Croatia

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| England Win | 51.5% | 34.1% | 42.8% |
| Draw | 21.7% | 33.9% | 27.8% |
| Croatia Win | 26.8% | 31.9% | 29.3% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- England Elo: 1826
- Croatia Elo: 1717
- P0: England 51.6% / Draw 21.6% / Croatia 26.8%

### Likelihood Updates
- [home] England 4-2-3-1 lineup confirmed from manual; key starters: Declan Rice, Eberechi Eze, Elliot Anderson (±6%) — lineup
- [away] Croatia 4-2-3-1 lineup confirmed from manual; key starters: Luka Sucic, Luka Modric, Mario Pasalic (±6%) — lineup
- [draw] Weather condition: rain (±6%) — simulation factors

### P1 (Final Bayesian)
- England: 51.5%  |  Draw: 21.7%  |  Croatia: 26.8%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 10,000
- Avg Goals: England 0.87 / Croatia 0.82 / Total 1.69
- Avg Shots: England 16.2 / Croatia 15.5
- Conversion: England 5.3% / Croatia 5.3%
- Avg Cards: 3.6

### Top Scorelines
- England 0-0 Croatia: 18.2%
- England 1-0 Croatia: 15.5%
- England 0-1 Croatia: 15.3%
- England 1-1 Croatia: 13.1%
- England 2-0 Croatia: 7.1%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 17.4%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Bayesian is higher on home win by 17.4pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.
- Monte Carlo is higher on draw by 12.2pp. Likely generative simulation effect from squad duels, tempo, or factors (1 factor effects).

### Consensus Estimate
- England: 42.8%  |  Draw: 27.8%  |  Croatia: 29.3%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 1 小时前
# 2026 世界杯预测报告：Portugal vs Congo DR

- **预测时间**: 2026-06-17 16:24:35
- **Run ID**: 20260617_162435
- **比赛信息**: 第 23 场 · K组
- **开球时间**: 2026-06-18 01:00 北京时间（原始：2026-06-17 12:00 local）
- **场馆**: Houston Stadium · Houston
- **贝叶斯数据更新时间**: 2026-06-17T16:24:18

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“主队胜”，概率为 38.1%，领先第二选项 3.5 个百分点。
- 综合置信度为低；模型分歧为 40.0%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 当前没有可标准化的 Polymarket 胜平负对照，市场概率不进入模型，只作为外部参考。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-17T16:24:18。
- 阵容数据存在未确认或未匹配项，建议先在阵容弹窗中手动修正首发名单。

## Synthesis / 综合概率

- Portugal 胜: 38.1%
- 平局: 27.2%
- Congo DR 胜: 34.7%
- 综合置信度: 低
- 方法分歧: 40.0%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Portugal 24.8% / 平局 20.5% / Congo DR 54.7%
- Elo: 1764 vs 1900.0
- lineup: Portugal 4-3-3 lineup confirmed from expert; key starters: Bruno Fernandes, Vitinha, Ruben Dias（幅度 3.8%）
- simulation factors: Pitch quality: heavy（幅度 4.0%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 不可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.62
- 入模[pitch]: 沉重或较差场地会压低进攻流畅度，因此只轻微提高平局倾向。；置信度 0.5
- 原始伤停线索 7 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 10000
- 90分钟胜平负: Portugal 51.5% / 平局 33.9% / Congo DR 14.6%
- 平均进球: 1.07 - 0.42
- 90分钟结果: Portugal 51.5% / 平局 33.9% / Congo DR 14.6%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 1309429543
- 模拟规则: 小组赛，90分钟允许平局
- 天气: 晴朗; 温度 24.9°C; 风速 27.2 km/h; 湿度 88%; 来源 open-meteo
- 场地: 草皮 unknown，质量偏重
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 场地偏重或质量较差：传球和盘带成功率下降。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Portugal: 4-3-3 / expert / predicted / confidence 0.74
  - 首发: Diogo Costa, Joao Cancelo, Ruben Dias, Goncalo Inacio, Nuno Mendes, Joao Neves, Vitinha, Bruno Fernandes, Bernardo Silva, Cristiano Ronaldo, Joao Felix
- 客队 Congo DR: 4-3-3 / roster-default / needs-confirmation / confidence 0.28
  - 首发: Mpasi, Kayembe, Inonga, Mbemba, Masuaku, Pickel, Moutoussamy, Elia, Mayele, Bakambu, Kakuta
- 阵容提醒: Congo DR: no reliable predicted/recent XI available; using roster-default starters by formation slots. Please confirm manually.；External formation source was not parsed; using local tactical profile fallback.；Congo DR lineup needs manual review.

## Diagnosis / 综合诊断

- Monte Carlo is higher on home win by 26.7pp. Likely generative simulation effect from squad duels, tempo, or factors (1 factor effects).
- Monte Carlo is higher on draw by 13.4pp. Likely generative simulation effect from squad duels, tempo, or factors (1 factor effects).
- Bayesian is higher on away win by 40.0pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.

## Market Comparison / Polymarket 概率对照

暂无 Polymarket 对照。

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Portugal vs Congo DR

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.5

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Portugal Win | 24.8% | 51.5% | 38.1% |
| Draw | 20.5% | 33.9% | 27.2% |
| Congo DR Win | 54.7% | 14.6% | 34.7% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Portugal Elo: 1764
- Congo DR Elo: 1900.0
- P0: Portugal 24.5% / Draw 20.2% / Congo DR 55.3%

### Likelihood Updates
- [home] Portugal 4-3-3 lineup confirmed from expert; key starters: Bruno Fernandes, Vitinha, Ruben Dias (±4%) — lineup
- [draw] Pitch quality: heavy (±4%) — simulation factors

### P1 (Final Bayesian)
- Portugal: 24.8%  |  Draw: 20.5%  |  Congo DR: 54.7%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 10,000
- Avg Goals: Portugal 1.07 / Congo DR 0.42 / Total 1.49
- Avg Shots: Portugal 15.4 / Congo DR 7.8
- Conversion: Portugal 6.9% / Congo DR 5.4%
- Avg Cards: 3.6

### Top Scorelines
- Portugal 1-0 Congo DR: 23.5%
- Portugal 0-0 Congo DR: 22.5%
- Portugal 2-0 Congo DR: 13.4%
- Portugal 1-1 Congo DR: 10.2%
- Portugal 0-1 Congo DR: 9.5%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 40.0%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Monte Carlo is higher on home win by 26.7pp. Likely generative simulation effect from squad duels, tempo, or factors (1 factor effects).
- Monte Carlo is higher on draw by 13.4pp. Likely generative simulation effect from squad duels, tempo, or factors (1 factor effects).
- Bayesian is higher on away win by 40.0pp. Likely evidence-layer effect (2 evidence items) rather than player-chain simulation.

### Consensus Estimate
- Portugal: 38.1%  |  Draw: 27.2%  |  Congo DR: 34.7%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

作者: xiejin77 时间: 1 小时前
# 2026 世界杯预测报告：Uzbekistan vs Colombia

- **预测时间**: 2026-06-17 16:28:11
- **Run ID**: 20260617_162811
- **比赛信息**: 第 24 场 · K组
- **开球时间**: 2026-06-18 10:00 北京时间（原始：2026-06-17 20:00 local）
- **场馆**: Mexico City Stadium · Mexico City
- **贝叶斯数据更新时间**: 2026-06-17T16:27:36

## Key Takeaways / 关键判断与结论

- 模型当前最倾向于“客队胜”，概率为 54.5%，领先第二选项 30.2 个百分点。
- 综合置信度为低；模型分歧为 25.4%，分歧越低代表贝叶斯与蒙特卡洛结论越一致。
- 当前没有可标准化的 Polymarket 胜平负对照，市场概率不进入模型，只作为外部参考。
- 本次贝叶斯上下文来源为 audited-live，更新时间为 2026-06-17T16:27:36。
- 本次预测已使用手动/外部阵容覆盖默认首发，贝叶斯与蒙特卡洛共享同一套首发名单。

## Synthesis / 综合概率

- Uzbekistan 胜: 24.2%
- 平局: 21.3%
- Colombia 胜: 54.5%
- 综合置信度: 低
- 方法分歧: 25.4%

## Bayesian Analysis / 贝叶斯分析

- 后验胜平负: Uzbekistan 17.1% / 平局 15.8% / Colombia 67.1%
- Elo: 1465 vs 1693
- lineup: Uzbekistan 4-2-3-1 lineup confirmed from manual; key starters: Akmal Mozgovoy, Azizjon Ganiev, Abbosbek Fayzullaev（幅度 6.5%）
- lineup: Colombia 4-2-3-1 lineup confirmed from manual; key starters: James Rodriguez, Jaminton Campaz, Jefferson Lerma（幅度 6.5%）
- simulation factors: Weather condition: drizzle（幅度 2.5%）

## Bayesian Evidence Audit / 贝叶斯证据审计

- Elo available: 可用
- Lineups trusted: 不可用
- Injuries trusted: 不可用
- Recent form trusted: 可用
- Weather available: 可用
- Pitch available: 可用
- Referee available: 不可用
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[lineup]: 首发阵容已通过大名单匹配；官方/手动确认权重较高，专家预测首发仅低权重入模。；置信度 0.78
- 入模[weather]: 小雨/毛毛雨只作为很弱的低比分信号，不再自动推断为沉重场地。；置信度 0.42
- 原始伤停线索 3 条仅保存审计；未匹配或低可信内容不入模。

## Monte Carlo / 蒙特卡洛模拟

- 模拟次数: 10000
- 90分钟胜平负: Uzbekistan 31.4% / 平局 26.8% / Colombia 41.8%
- 平均进球: 1.13 - 1.35
- 90分钟结果: Uzbekistan 31.4% / 平局 26.8% / Colombia 41.8%

## Simulation Parameters / 模拟参数与外部因素

- 随机种子: 1309429543
- 模拟规则: 小组赛，90分钟允许平局
- 天气: drizzle; 温度 16.6°C; 风速 5.9 km/h; 湿度 85%; 来源 open-meteo
- 场地: 草皮 unknown，质量正常
- 旅行/休息: 主队休息 7 天，客队休息 7 天，主队旅程 0 km，客队旅程 0 km
- 因素影响: 未启用明显天气/场地调制，按中性条件模拟。

## Injury Mapping / 伤停映射

- 暂无可映射伤停，Monte Carlo 阵容保持默认。

## Lineups / 阵容来源与匹配状态

- 主队 Uzbekistan: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Abduvohid Nematov, Abdukodir Khusanov, Abdulla Abdullaev, Avazbek Ulmasaliev ., Bekhruz Karimov, Abbosbek Fayzullaev, Akmal Mozgovoy, Azizbek Amonov ., Azizjon Ganiev, Dostonbek Khamdamov, Eldor Shomurodov (Istanbul Basaksehir
- 客队 Colombia: 4-2-3-1 / manual / confirmed / confidence 0.78
  - 首发: Alvaro Montero, Daniel Munoz, Davinson Sanchez, Devier Machado, Jhon Lucumi, James Rodriguez, Jaminton Campaz, Carlos Andres Gomez, Jefferson Lerma, Jhon Arias, Jhon Cordoba

## Diagnosis / 综合诊断

- Monte Carlo is higher on home win by 14.3pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on draw by 11.0pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Bayesian is higher on away win by 25.3pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.

## Market Comparison / Polymarket 概率对照

暂无 Polymarket 对照。

## Engine Raw Output / 引擎原始输出

# Match Prediction Report
## World Cup 2026 Group Stage: Uzbekistan vs Colombia

- **Venue**: neutral
- **Referee**: TBD
- **Confidence**: LOW
- **Shared Baseline**: group / context quality 0.67

## Summary

| Outcome | Bayesian | Monte Carlo | Averaged |
|---------|----------|-------------|----------|
| Uzbekistan Win | 17.1% | 31.4% | 24.2% |
| Draw | 15.8% | 26.8% | 21.3% |
| Colombia Win | 67.1% | 41.8% | 54.5% |

## Layer 1: Bayesian Analysis

### P0 (Elo Prior)
- Uzbekistan Elo: 1465
- Colombia Elo: 1693
- P0: Uzbekistan 17.0% / Draw 16.0% / Colombia 66.9%

### Likelihood Updates
- [home] Uzbekistan 4-2-3-1 lineup confirmed from manual; key starters: Akmal Mozgovoy, Azizjon Ganiev, Abbosbek Fayzullaev (±6%) — lineup
- [away] Colombia 4-2-3-1 lineup confirmed from manual; key starters: James Rodriguez, Jaminton Campaz, Jefferson Lerma (±6%) — lineup
- [draw] Weather condition: drizzle (±2%) — simulation factors

### P1 (Final Bayesian)
- Uzbekistan: 17.1%  |  Draw: 15.8%  |  Colombia: 67.1%

## Layer 2: Monte Carlo (FM-Style)

- Simulations: 10,000
- Avg Goals: Uzbekistan 1.13 / Colombia 1.35 / Total 2.47
- Avg Shots: Uzbekistan 16.7 / Colombia 18.6
- Conversion: Uzbekistan 6.8% / Colombia 7.2%
- Avg Cards: 3.7

### Top Scorelines
- Uzbekistan 1-1 Colombia: 13.3%
- Uzbekistan 0-1 Colombia: 11.5%
- Uzbekistan 1-0 Colombia: 9.4%
- Uzbekistan 1-2 Colombia: 8.8%
- Uzbekistan 0-0 Colombia: 7.6%

## Layer 3: Synthesis

**Confidence: LOW** (max divergence: 25.4%)
- Method weights: Bayesian 50% / Monte Carlo 50%
- Rationale: Methods diverge materially; averaged estimate is shown, but confidence is low and diagnostics should drive interpretation.

### Divergence Analysis
- Monte Carlo is higher on home win by 14.3pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Monte Carlo is higher on draw by 11.0pp. Likely generative simulation effect from squad duels, tempo, or factors (0 factor effects).
- Bayesian is higher on away win by 25.3pp. Likely evidence-layer effect (3 evidence items) rather than player-chain simulation.

### Consensus Estimate
- Uzbekistan: 24.2%  |  Draw: 21.3%  |  Colombia: 54.5%

## Disclaimer / 免责声明

本报告仅供数据分析和赛前讨论参考，不构成投注、交易或投资建议。请理性看待体育赛事结果，远离非法赌博。

欢迎光临爱吱声 (http://www.aswetalk.net/bbs/)