From d240cd5d940fb2515a30d0a2858a756b7d5faf92 Mon Sep 17 00:00:00 2001
From: Claude <noreply@anthropic.com>
Date: Sun, 5 Apr 2026 20:35:48 +0000
Subject: [PATCH] Fix flaky Champions Stage stochastic test
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

The makeTeams(8) helper creates only a 70-pt Elo spread (1800→1730).
With the Champions Stage bracket math this gives team-0 a ~19.6% win
rate — right at the 0.2 threshold, causing the test to fail ~63% of
the time in CI despite 1000 iterations.

Use 100-pt steps (1800→1100) instead, giving team-0 a ~40% win rate
and raising the assertion threshold to 0.25 for a clear safety margin.

https://claude.ai/code/session_019w21Nkf5TvTZHH6oVHaQXR
---
 .../__tests__/cs-major-simulator.test.ts          | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)

diff --git a/app/services/simulations/__tests__/cs-major-simulator.test.ts b/app/services/simulations/__tests__/cs-major-simulator.test.ts
index 78861b6..f57efad 100644
--- a/app/services/simulations/__tests__/cs-major-simulator.test.ts
+++ b/app/services/simulations/__tests__/cs-major-simulator.test.ts
@@ -217,14 +217,21 @@ describe("simulateChampionsStage", () => {
   });
 
   it("top Elo team wins more often than last (stochastic)", () => {
-    const teams = makeTeams(8);
-    // team-0 has highest Elo, team-7 has lowest
+    // Use a clear Elo spread so team-0 has a decisive advantage.
+    // makeTeams(8) only gives a 70-pt gap (1800→1730) which puts the
+    // true win rate right at the 0.2 threshold — extremely flaky.
+    // 100-pt steps (1800→1100) give team-0 a ~40% win rate, well clear of 0.25.
+    const teams = Array.from({ length: 8 }, (_, i) => ({
+      id: `team-${i}`,
+      elo: 1800 - i * 100,
+      rank: i + 1,
+    }));
     let wins = 0;
     for (let i = 0; i < 1000; i++) {
       const result = simulateChampionsStage(teams);
       if (result.placements.get("team-0") === 1) wins++;
     }
-    // Should win significantly more than 12.5% of the time (1/8 random)
-    expect(wins / 1000).toBeGreaterThan(0.2);
+    // Should win well above 12.5% (1/8 random baseline)
+    expect(wins / 1000).toBeGreaterThan(0.25);
   });
 });