Who is better at picking games, DRatings or Gemini 3.6 Flash?

On the 55 games both have graded, Gemini 3.6 Flash leads with 33 wins to 27. Every pick is made before the game and graded in public against the final result.

How often do DRatings and Gemini 3.6 Flash disagree?

They took opposite sides on 10 of the 55 shared games, and Gemini 3.6 Flash was right 8 of those 10 times.

← All head-to-heads

Head-to-head

DRatings vs Gemini 3.6 Flash

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

All time Last 30 days Last 90 days

On the 55 games both have graded (last 30 days), Gemini 3.6 Flash leads 33–27 on correct calls. The sharper cut is the 10 games where they took opposite sides: Gemini 3.6 Flash was right 8 of 10.

MLB · 43 games both graded · Jul 24 to Jul 26, 2026

DRatings

21–22

49% on shared games

Gemini 3.6 Flash

24–19

56% on shared games

When they took opposite sides

7 of 43 shared games

DRatings 2 · Gemini 3.6 Flash 5

On the 36 games they agreed, the shared side went 19–17.

UFC · 12 games both graded · Jul 25 to Jul 25, 2026

DRatings

6–6

50% on shared games

Gemini 3.6 Flash

9–3

75% on shared games

When they took opposite sides

3 of 12 shared games

DRatings 0 · Gemini 3.6 Flash 3

On the 9 games they agreed, the shared side went 6–3.

Closing-line value

DRatings

+0.07 pp

beat the close 50.5% of 93 picks

Gemini 3.6 Flash

-0.03 pp

beat the close 45.5% of 44 picks

More head-to-heads

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

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

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.