Archive

July 12, 2026 · comparison

On 12 July 2026, before either semi, we locked eight prediction boards: ChatGPT, Claude, Gemini, and Grok — each once on a fast / cheap model tier, once on a max-reasoning tier.

Same three games. Tools and web search allowed either way — but no betting odds. After the final on 19 July, we score both boards and ask: does the heavy model buy better football picks, or is it expensive vibes?

The three games

SF1Tue 14 Jul

France vs Spain

Dallas Stadium

SF2Wed 15 Jul

England vs Argentina

Atlanta Stadium

FinalSun 19 Jul

SF winners

NY / NJ Stadium

Third place (18 Jul) is not scored. Finalists must match each model's own semi winners.

Board A

FAST — cheap / fast model

Fastest tier (Mini Light, Flash, Haiku-class, etc.). Tools and non-odds web OK.

ModelSF1France–SpainSF2Eng–ArgFinalScoreChamp
ChatGPT
France
2–1
England
2–1
France
2–1 · Eng
France
Claude
Spain
2–1
Argentina
2–1
Argentina
1–0 · Spain
Argentina
Gemini
France
2–1
Argentina
2–1
Argentina
2–1 · Fra
Argentina
Grok
Spain
2–1
Argentina
1–0
Spain
2–1 · Arg
Spain

Board B

DEEP — max reasoning

Highest-reasoning / “think hard” tier. Same info rules: research OK, no betting odds.

ModelSF1France–SpainSF2Eng–ArgFinalScoreChampvs FAST
ChatGPT
France
2–1
England
2–1
France
2–1 · Eng
FranceSame
Claude
France
2–1
England
2–1
France
2–1 · Eng
FranceChanged
Gemini
France
2–1
Argentina
2–1
France
2–1 · Arg
FranceChanged
Grok
France
2–1
England
2–1
France
1–0 · Eng
FranceChanged

Locked 12 Jul · before either semi

Cheap models disagree. Heavy models converge on France.

  • ChatGPT — DEEP = FAST (identical board).
  • Claude — FAST≠DEEP (Argentina path → France path).
  • Gemini — FAST≠DEEP (same semis; final flips Argentina → France).
  • Grok — FAST≠DEEP (Spain path → France path).
FAST champsFranceArgentinaArgentinaSpain
DEEP champsFranceFranceFranceFrance

Settings

What “fast” and “deep” meant

ChatGPT

Fast
GPT-5.4 Mini Light · web OK, no odds
Deep
GPT-5.6 Sol Ultra · high reasoning + web

Claude

Fast
Haiku 4.5 · low effort
Deep
Fable 5 Max · web + extended thinking

Gemini

Fast
3.1 Pro (Low) · Planning Mode
Deep
3.5 Flash (High) · Planning Mode

Grok

Fast
Single-pass snap · no tools
Deep
Deliberate second pass · no odds

After 19 Jul

How we score

Correct winner

01

+1 per match (SF1, SF2, Final). ET counts; pens = that side wins.

Exact score

02

+2 bonus per match (reg + ET score; ignore pen digits).

Then compare

03

FAST board · DEEP board · Δ (DEEP − FAST) · mean FAST vs mean DEEP.

Why bother

Everyone argues about reasoning models in the abstract. Here the unit is public and rude: three scores on a calendar. Same constraints, two compute regimes, one tournament clock.

If DEEP wins, interesting.
If FAST ties or beats DEEP, also interesting — and cheaper.

Results after the final whistle.

Related: the context rocket equation — why every LLM turn is a full relaunch of your payload — and the API cost calculator if you want numbers on “thinking harder.”