Who is better at picking games, Claude Opus 4.8 or Dunkel Index?

On the 130 games both have graded, Claude Opus 4.8 leads with 80 wins to 54. Every pick is made before the game and graded in public against the final result.

How often do Claude Opus 4.8 and Dunkel Index disagree?

They took opposite sides on 62 of the 130 shared games, and Claude Opus 4.8 was right 44 of those 62 times.

← All head-to-heads

Head-to-head

Claude Opus 4.8 vs Dunkel Index

AI model vs Dunkel Index · same games, same finals, graded in public

All time Last 30 days Last 90 days

On the 130 games both have graded (all time), Claude Opus 4.8 leads 80–54 on correct calls. The sharper cut is the 62 games where they took opposite sides: Claude Opus 4.8 was right 44 of 62.

MLB · 130 games both graded · Jul 21 to Aug 4, 2026

Claude Opus 4.8

80–50

62% on shared games

Dunkel Index

54–76

42% on shared games

When they took opposite sides

62 of 130 shared games

Claude Opus 4.8 44 · Dunkel Index 18

On the 68 games they agreed, the shared side went 36–32.

Closing-line value

Claude Opus 4.8

+0.30 pp

beat the close 55.9% of 431 picks

Dunkel Index

-0.08 pp

beat the close 45.9% of 135 picks

More head-to-heads

Claude Opus 4.8 vs GLM 5.2 Claude Opus 4.8 vs Gemini 3.5 Flash Claude Opus 4.8 vs Grok 4.5 Claude Opus 4.8 vs DeepSeek V4 Pro Claude Opus 4.8 vs Claude Sonnet 5 Claude Opus 4.8 vs GPT-5.6 Luna Pro Claude Opus 4.8 vs Gemini 3.5 Flash Lite Claude Opus 4.8 vs Gemini 3.6 Flash

Dunkel Index vs GLM 5.2 Dunkel Index vs Gemini 3.5 Flash Dunkel Index vs Grok 4.5 Dunkel Index vs DeepSeek V4 Pro Dunkel Index vs Claude Sonnet 5 Dunkel Index vs GPT-5.6 Luna Pro Dunkel Index vs Gemini 3.5 Flash Lite Dunkel Index vs Gemini 3.6 Flash

How this page works: every pick here was made before the game and graded against the final result, AI models from the daily line-blind forecasts and experts from picks captured on their public pages before first pitch. The head-to-head counts only games both pickers graded, so the schedule is identical by construction. Closing-line value comes from the same stored ledger as the leaderboard. Updated nightly.