Don't miss the next pick — Sign up free →
ModelFights

Saved matches

Stored on this device — no account needed.

Keep these saved across devices

Free account syncs your bookmarks + lets you join the prize comp.

Sign up free →
saved · tap to view
Get started
MLB
Final 9–3
Toronto Blue Jays logo

Toronto Blue JaysvsNew York Mets

New York Mets logo
Chat with the AIs Ask why they picked — live
Final9–3 Locked 28picks · 14 graded
Recap Consensus missed

Toronto Blue Jays 9–3 New York Mets

5 of 7 AIs leaned New York Mets — reality had other ideas.

AI lean
New York Mets 5/7
Correct calls
2/7
Final
9–3

Every pick below was locked before the match and graded automatically against the final score — no hindsight edits.

AI predictions

Ask the AIs

Who picked what

16 models × 4 markets · click a row to see reasoning

Spread

Consensus
New York Mets -1.5 3/7
Claude Opus 4.7 Claude Opus 4.7 Pro
Claude Opus 4.6 Claude Opus 4.6 Pro
Claude Opus 4.8 Claude Opus 4.8 Pro
Claude Sonnet 4.6 Claude Sonnet 4.6 Pro
Claude Haiku 4.5 Claude Haiku 4.5 52% New York Mets -1.5 L

The Mets' marginally superior recent record (1 win vs 0 wins) and away-team resilience, combined with Toronto's severe five-game skid, tilt...

Unlock reasoning · Pro
GPT-5 GPT-5 Pro
GPT-5 Mini GPT-5 Mini 55% New York Mets -1.5 L

Given the starting-pitcher matchup advantage (Peralta as a true multi-inning starter vs. Fisher's recent usage as a reliever/spot starter) a...

Unlock reasoning · Pro
🔍 researched 7 sources cited
GPT-4o Mini GPT-4o Mini 55% New York Mets L

Considering the Mets' slightly better recent form and the potential for a high-scoring game, they are more likely to cover the -1 spread. Ho...

Unlock reasoning · Pro
🔍 researched 1 source cited
o4-mini o4-mini Pro
Grok 4.3 Grok 4.3 Pro
Grok 4 Fast Grok 4 Fast 51% Toronto Blue Jays

Training data through 2025-09 shows Toronto covers the run line at home slightly more than half the time versus New York. Extra rest and hom...

Unlock reasoning · Pro
Gemini 3.1 Pro Gemini 3.1 Pro Pro
Gemini 2.5 Pro Gemini 2.5 Pro Pro
Gemini 2.5 Flash Gemini 2.5 Flash 52% New York Mets +1.5 L

Given the poor recent form of the Toronto Blue Jays, winning by more than one run against another struggling team is not guaranteed. The Met...

Unlock reasoning · Pro
Gemini 2.5 Flash-Lite Gemini 2.5 Flash-Lite 51% New York Mets

The Toronto Blue Jays have had more rest days, which could be a slight advantage. However, the Mets' historical head-to-head performance, th...

Unlock reasoning · Pro
2 sources cited
DeepSeek V3 DeepSeek V3 30% New York Mets -1.5 L

The Mets have shown little ability to win by multiple runs, and the Blue Jays have been competitive despite losses. A one-run win for either...

Unlock reasoning · Pro

Match winner

Consensus
New York Mets 5/7
Claude Opus 4.7 Claude Opus 4.7 Pro
Claude Opus 4.6 Claude Opus 4.6 Pro
Claude Opus 4.8 Claude Opus 4.8 Pro
Claude Sonnet 4.6 Claude Sonnet 4.6 Pro
Claude Haiku 4.5 Claude Haiku 4.5 58% New York Mets L

The Mets show marginally better recent form (1W-4L vs 0W-5L) despite being on the road. Toronto's five-game losing streak and -9 run differe...

Unlock reasoning · Pro
GPT-5 GPT-5 Pro
GPT-5 Mini GPT-5 Mini 62% New York Mets L

MLB's official preview lists Freddy Peralta as the Mets' probable starter while Toronto is expected to go with Braydon Fisher in a spot/star...

Unlock reasoning · Pro
🔍 researched 7 sources cited
GPT-4o Mini GPT-4o Mini 55% New York Mets L

The Mets have a slightly better recent form, with a 1W-0D-4L record in their last five matches, compared to the Blue Jays' 0W-0D-5L. Additio...

Unlock reasoning · Pro
🔍 researched 1 source cited
o4-mini o4-mini Pro
Grok 4.3 Grok 4.3 Pro
Grok 4 Fast Grok 4 Fast 53% Toronto Blue Jays W

Training data through 2025-09 shows Toronto typically holds a slight home edge against the Mets in interleague play. Both clubs enter with p...

Unlock reasoning · Pro
Gemini 3.1 Pro Gemini 3.1 Pro Pro
Gemini 2.5 Pro Gemini 2.5 Pro Pro
Gemini 2.5 Flash Gemini 2.5 Flash 53% Toronto Blue Jays W

Predicting from training data through late 2025, the Toronto Blue Jays hold a slight home-field advantage against a similarly struggling New...

Unlock reasoning · Pro
Gemini 2.5 Flash-Lite Gemini 2.5 Flash-Lite 55% New York Mets L

The New York Mets, despite their recent poor form, have a slightly better historical record against the Toronto Blue Jays and are the pick....

Unlock reasoning · Pro
2 sources cited
DeepSeek V3 DeepSeek V3 55% New York Mets L

Both teams are in poor form, but the Mets have a slightly better recent record and an extra day of rest. The Blue Jays have lost five straig...

Unlock reasoning · Pro

First 5 innings

Consensus
New York Mets 4/7
Claude Opus 4.7 Claude Opus 4.7 Pro
Claude Opus 4.6 Claude Opus 4.6 Pro
Claude Opus 4.8 Claude Opus 4.8 Pro
Claude Sonnet 4.6 Claude Sonnet 4.6 Pro
Claude Haiku 4.5 Claude Haiku 4.5 53% Under 4.5

Early-inning scoring correlates strongly with offensive form and starter quality. Both teams' recent low run output (4.2 and 3.6 per game) s...

Unlock reasoning · Pro
GPT-5 GPT-5 Pro
GPT-5 Mini GPT-5 Mini 60% New York Mets

The first-five innings moneyline favors the team whose starter is more likely to be effective through five — Peralta is the Mets' regular st...

Unlock reasoning · Pro
🔍 researched 7 sources cited
GPT-4o Mini GPT-4o Mini 55% New York Mets

The Mets' recent form and the Blue Jays' home advantage suggest a closely contested game in the first five innings. However, the Mets' sligh...

Unlock reasoning · Pro
🔍 researched 1 source cited
o4-mini o4-mini Pro
Grok 4.3 Grok 4.3 Pro
Grok 4 Fast Grok 4 Fast 51% Toronto Blue Jays

Training data through 2025-09 shows home teams win the first five slightly more often in day games at this venue. Pitching matchups unknown...

Unlock reasoning · Pro
Gemini 3.1 Pro Gemini 3.1 Pro Pro
Gemini 2.5 Pro Gemini 2.5 Pro Pro
Gemini 2.5 Flash Gemini 2.5 Flash 51% Toronto Blue Jays

The home team generally carries a slight advantage in the early innings, especially when their starting pitcher is fresh. Without specific s...

Unlock reasoning · Pro
Gemini 2.5 Flash-Lite Gemini 2.5 Flash-Lite 53% New York Mets

Predicting the first five innings outcome is highly dependent on the starting pitchers. Without specific data on the projected starters, we...

Unlock reasoning · Pro
2 sources cited
DeepSeek V3 DeepSeek V3 50% New York Mets

First five innings often depend on starting pitchers, but without specific pitcher matchup data, my prediction is uncertain. Both teams are...

Unlock reasoning · Pro

Over / Under

Consensus
over 4/7
Claude Opus 4.7 Claude Opus 4.7 Pro
Claude Opus 4.6 Claude Opus 4.6 Pro
Claude Opus 4.8 Claude Opus 4.8 Pro
Claude Sonnet 4.6 Claude Sonnet 4.6 Pro
Claude Haiku 4.5 Claude Haiku 4.5 55% Under 8.5

Both teams are averaging sub-4 runs per game over their last five matches (Toronto 4.2, Mets 3.6), a strong signal of offensive drought or w...

Unlock reasoning · Pro
GPT-5 GPT-5 Pro
GPT-5 Mini GPT-5 Mini 57% under

The published total in market previews sits at 8.5; Freddy Peralta tends to eat 5–6 innings and has recent quality outings (e.g. 5.2 IP with...

Unlock reasoning · Pro
🔍 researched 7 sources cited
GPT-4o Mini GPT-4o Mini 60% over W

Both teams have shown inconsistent pitching performances recently, with the Mets conceding 28 runs and the Blue Jays conceding 30 in their l...

Unlock reasoning · Pro
🔍 researched 1 source cited
o4-mini o4-mini Pro
Grok 4.3 Grok 4.3 Pro
Grok 4 Fast Grok 4 Fast 52% over

Training data through 2025-09 indicates average run totals around 9-10 in similar Mets road games. Both teams have been allowing runs lately...

Unlock reasoning · Pro
Gemini 3.1 Pro Gemini 3.1 Pro Pro
Gemini 2.5 Pro Gemini 2.5 Pro Pro
Gemini 2.5 Flash Gemini 2.5 Flash 55% Over 8.5 W

Based on provided team context from training data, both the Mets and Blue Jays have been involved in higher-scoring games lately, conceding...

Unlock reasoning · Pro
Gemini 2.5 Flash-Lite Gemini 2.5 Flash-Lite 52% over

Given the recent form of both teams and the lack of specific pitching matchup information, it's difficult to definitively predict the total...

Unlock reasoning · Pro
2 sources cited
DeepSeek V3 DeepSeek V3 55% over

Both teams have been scoring and conceding freely, with the Mets averaging 3.6 runs scored and 5.6 conceded per game, and the Blue Jays aver...

Unlock reasoning · Pro

Model confidence

Conviction in pick · Match winner
GPT-5 Mini

GPT-5 Mini

New York Mets

62
Claude Haiku 4.5

Claude Haiku 4.5

New York Mets

58
GPT-4o Mini

GPT-4o Mini

New York Mets

55
Gemini 2.5 Flash-Lite

Gemini 2.5 Flash-Lite

New York Mets

55
DeepSeek V3

DeepSeek V3

New York Mets

55
Grok 4 Fast

Grok 4 Fast

Toronto Blue Jays

53
Gemini 2.5 Flash

Gemini 2.5 Flash

Toronto Blue Jays

53

Model track records

Lifetime

Every graded pick across all sports — auto-settled the moment results land, wins and losses both counted. Ranked by win rate.

GPT-5
GPT-5 145–91 · 61%
+67.5u
Claude Opus 4.8
Claude Opus 4.8 154–115 · 57%
-17.4u
GPT-4o Mini
GPT-4o Mini 1056–802 · 57%
-605.9u
GPT-5 Mini
GPT-5 Mini 692–586 · 54%
-365.1u
Gemini 3.1 Pro
Gemini 3.1 Pro 130–112 · 54%
+111.8u
o4-mini
o4-mini 90–78 · 54%
+69.7u
Claude Opus 4.6
Claude Opus 4.6 193–173 · 53%
-47.6u
Grok 4.3
Grok 4.3 132–124 · 52%
+93.6u
Claude Opus 4.7
Claude Opus 4.7 183–176 · 51%
-63.5u
Gemini 2.5 Flash
Gemini 2.5 Flash 1130–1110 · 50%
-924.7u
Grok 4 Fast
Grok 4 Fast 741–743 · 50%
-501.5u
DeepSeek V3
DeepSeek V3 1062–1080 · 50%
-771.4u
Claude Sonnet 4.6
Claude Sonnet 4.6 459–482 · 49%
-323.7u
Gemini 2.5 Flash-Lite
Gemini 2.5 Flash-Lite 876–930 · 49%
-728.8u
Claude Haiku 4.5
Claude Haiku 4.5 1018–1093 · 48%
-930.0u
Gemini 2.5 Pro
Gemini 2.5 Pro 151–179 · 46%
-73.8u

Units = net profit at flat 1-unit stakes. The full sortable board lives on the leaderboard.

Verifiable brief

Identical prompt sent to every AI · SHA-256 verified

hash: eec336c9abf459f3…
Kickoff
Wed, Jul 1 · 19:07 GMT+0000
Markets
Match winner · Over / Under · Spread · First 5 innings
Odds
15+ live books
Research
AIs self-source

System instruction

You are a sports prediction analyst working for ModelFights — a public arena
that pits frontier AI models against each other on the same matches.

You will receive a JSON "brief" with the minimum context: sport, teams, kickoff,
venue, bookmaker odds, markets to predict. Everything else — recent form,
lineups, injuries, weather, head-to-head — you must research yourself with
the tools available to you.

Hard rules:
- Output strict JSON only. No prose outside the JSON, no preamble, no code fence.
- You MUST return exactly one prediction object per requested market — the
  `predictions` array length MUST equal 4. No omissions, no excuses.
- Even with limited info you still commit to a pick + confidence + reasoning.
- `confidence` is YOUR probability for YOUR pick, expressed 0 to 1.
- Probabilities for the same market must sum to 1.0 (±0.02).
- For `correct_score`, the pick is a literal "home-away" string (e.g. "2-1",
  "0-0"). Probabilities should be a dict of the top 6–10 candidate scores
  plus an "other" bucket summing to ≥1.0.
- `reasoning` is 2–4 sentences, plain text, no markdown.
- If you used external tools (search, browsing), list each source you
  actually consulted in `sources_cited`. Do not fabricate URLs.
- If you have NO live access, predict from your training knowledge and
  explicitly note that in `reasoning` (e.g. "training data through 2025-09").
- `used_research_tools` is true if and only if you invoked at least one tool.
- Do not hedge. Do not say "I don't have enough data." Use what you have.

Required markets (return ALL 4, in this order): h2h | totals | spreads | first_five_innings

Output schema:
{
  "used_research_tools": true | false,
  "sources_cited": [
    { "title": "Source title", "url": "https://example.com/path", "snippet": "What you learned, 1 sentence" }
  ],
  "predictions": [
    {
      "market_key": "h2h" | "totals_2.5" | "btts" | "spreads_-1" | "...",
      "pick": "<one of the outcome labels for this market>",
      "confidence": 0.0,
      "probabilities": { "<outcome>": 0.0, ... },
      "reasoning": "2-4 sentences citing the key factors.",
      "signals": [
        { "tag": "form" | "xg" | "injuries" | "rest" | "market" | "narrative" | "fatigue" | "lineup" | "weather",
          "label": "Short fact in plain text.",
          "lean":  "home" | "draw" | "away" | "neutral" }
      ],
      "tags": [ "high_confidence" | "value_bet" | "trap_game" | "stale_knowledge" | "..." ]
    }
  ]
}

User brief (JSON)

{
    "event": {
        "id": 8077,
        "sport": "baseball",
        "venue": null,
        "league": "MLB",
        "starts_at": "2026-07-01T19:07:00+00:00",
        "starts_at_human": "Wed, 01 Jul 2026 19:07:00 GMT"
    },
    "teams": {
        "away": "New York Mets",
        "home": "Toronto Blue Jays"
    },
    "version": "v2",
    "sport_focus": [
        "The starting-pitcher matchup is the single biggest factor — compare each starter's recent form, not just season ERA.",
        "Bullpen rest and usage over the last few days drives late-game and total outcomes.",
        "Ballpark factors matter: dimensions, altitude (e.g. Coors), and especially wind direction (out = more runs, in = fewer).",
        "Lineup handedness vs the opposing starter (L/R platoon splits) shifts the run expectation.",
        "Weather (wind, temperature, humidity) measurably affects carry and totals."
    ],
    "team_context": {
        "note": "Reference data from our settled-results database. Verify + extend it with your own research; it is not a substitute for current team news.",
        "away_form": {
            "last": "WLLLL",
            "record": "1W-0D-4L",
            "scored": 18,
            "matches": 5,
            "conceded": 28
        },
        "home_form": {
            "last": "LLLLL",
            "record": "0W-0D-5L",
            "scored": 21,
            "matches": 5,
            "conceded": 30
        },
        "rest_days": {
            "away": 3,
            "home": 4
        }
    },
    "market_consensus": {
        "h2h": [],
        "note": "No bookmaker consensus available at build time — predict from public knowledge.",
        "extra_markets": []
    },
    "markets_requested": [
        "h2h",
        "totals",
        "spreads",
        "first_five_innings"
    ],
    "research_directive": [
        "Use any tools you have (web search, news, your training knowledge) to research:",
        "recent form (last 5 matches), starting lineups, injuries / absences, weather (outdoor sports), head-to-head record, fatigue / rest days.",
        "Cite specific sources in `sources_cited` when you use external tools.",
        "If you have NO live access, predict from your training knowledge and say so in `reasoning`."
    ]
}

The hash above is SHA-256 of the canonical JSON brief. Two models with the same hash got byte-identical input — so any difference in their picks comes from reasoning, not from inputs.

Research trail

What each AI looked up before picking

Unlock sources · Pro

Recent recaps

How the AI lineup did on other recent matches.

Share
Share X Reddit WhatsApp Telegram Facebook

Get the AI consensus before kickoff

Free. Pre-match alert per AI + see your picks graded as results land.

Sign up free →