Who is better at picking games, Claude Opus 4.8 or Covers Consensus?

On the 206 games both have graded, Claude Opus 4.8 leads with 119 wins to 117. Every pick is made before the game and graded in public against the final result.

How often do Claude Opus 4.8 and Covers Consensus disagree?

They took opposite sides on 32 of the 206 shared games, and Claude Opus 4.8 was right 17 of those 32 times.

← All head-to-heads

Head-to-head

Claude Opus 4.8 vs Covers Consensus

AI model vs Covers · same games, same finals, graded in public

All time Last 30 days Last 90 days

On the 206 games both have graded (all time), Claude Opus 4.8 leads 119–117 on correct calls. The sharper cut is the 32 games where they took opposite sides: Claude Opus 4.8 was right 17 of 32.

MLB · 206 games both graded · Jul 20 to Aug 6, 2026

Claude Opus 4.8

119–87

58% on shared games

Covers Consensus

117–89

57% on shared games

When they took opposite sides

32 of 206 shared games

Claude Opus 4.8 17 · Covers Consensus 15

On the 174 games they agreed, the shared side went 102–72.

Closing-line value

Claude Opus 4.8

+0.30 pp

beat the close 55.9% of 431 picks

Covers Consensus

+0.22 pp

beat the close 53.6% of 207 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

Covers Consensus vs GLM 5.2 Covers Consensus vs Gemini 3.5 Flash Covers Consensus vs Grok 4.5 Covers Consensus vs DeepSeek V4 Pro Covers Consensus vs Claude Sonnet 5 Covers Consensus vs GPT-5.6 Luna Pro Covers Consensus vs Gemini 3.5 Flash Lite Covers Consensus 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.