Model Dutch Blitz mechanics in Race Outlook

[mwickett]

Summary

• Derive per-player cards-played and blitz-pile samples from actual round data
• Use mechanics-aware sampling in the Monte Carlo forecast when cards/blitz data is available
• Add next-round blitz-out win probability and 20+ point swing probability
• Surface Blitz-out path and 20+ pt swing readouts in Race Outlook

Stack notes

• Stacked on Blend player history into Race Outlook #268 (mwickett/history-blended-race-outlook)
• This is the Dutch Blitz mechanics layer after the current-game Race Outlook and historical profile blend

Claude review

• Claude reviewed the PR3 diff and found no blocking issues
• Addressed its major fidelity finding by enforcing the existing GAME_RULES.MIN_CARDS_FOR_BLITZ minimum whenever a simulated round is a blitz
• Added a regression test showing swing probability is intentionally independent from next-round win probability

Verification

• npm run lint
• npm test
• RESEND_API_KEY=re_dummy npx next build

---

Integration/testing PR: #270 contains the full stack for end-to-end testing.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Actions	Updated (UTC)
blitzer	Ready	Preview, Comment	Jun 27, 2026 7:41pm

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: d8a7b05839

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Ensure simulated rounds include a blitzer

When each player's mechanics are sampled independently here, a future round can have zero outcomes[i].blitzed (for four players with 20% blitz rates this happens about 41% of the time). The app's scoring rules reject such rounds via validateGameRules in src/lib/validation/gameRules.ts, which requires at least one blitzPileRemaining === 0, so the mechanics forecast spends many simulations on impossible non-blitz rounds and skews next-round danger/blitz-out odds; resample or otherwise condition the round on at least one blitzer.

Useful? React with 👍 / 👎.

[mwickett]

Closed as superseded by merged integration PR #270.