Gameweek 6 Verdict — FootyPulse

Premier League · Gameweek 6 Verdict

Gameweek 6 Verdict: 5/10 correct as upsets dominate the round

Published 13 October 2026 · FootyPulse AI Analysis

The AI model went 5-for-10 in Gameweek 6, matching exactly 50% accuracy as five fixtures defied the pre-match probabilities. Season accuracy now sits at 40% (23 correct from 57 fixtures), with draws and surprise home wins undermining otherwise confident calls.

The scoreboard

Gameweek 6 delivered five correct predictions from ten fixtures—a 50% hit rate that dragged season accuracy down slightly to 40%. Through 57 Premier League matches, the model has now logged 23 correct calls, with upsets and draws continuing to punish probability-based forecasting.

Biggest upset

Manchester United 1-1 Tottenham Hotspur proved the gameweek's most jarring result. The model backed United at home with 58% confidence, only for Spurs to salvage a point at Old Trafford. A draw—the outcome the data deemed least probable—neutralised what looked like a straightforward home banker on paper.

Win probability — Manchester United 58%, Draw 24%, Tottenham Hotspur 18%

Best correct call

Chelsea 5-1 Bournemouth stands as the gameweek's most precarious correct pick. The model favoured Chelsea with just 38% confidence, the lowest of any winning call this round. What the data flagged as a coin-flip turned into a five-goal statement, rewarding the model's marginal lean toward the home side despite minimal pre-match conviction.

Win probability — Chelsea 38%, Draw 35%, Bournemouth 27%

Every fixture

  • Arsenal 2-1 Leeds United – Predicted home, actual home ✅
  • Ipswich Town 2-1 Fulham – Predicted home, actual home ✅
  • Aston Villa 2-2 Brentford – Predicted away, actual draw ❌
  • Sunderland 0-2 Brighton & Hove Albion – Predicted away, actual away ✅
  • Chelsea 5-1 Bournemouth – Predicted home, actual home ✅
  • Manchester United 1-1 Tottenham Hotspur – Predicted home, actual draw ❌
  • Hull City 2-2 Everton – Predicted home, actual draw ❌
  • Crystal Palace 3-1 Nottingham Forest – Predicted away, actual home ❌
  • Liverpool 0-3 Manchester City – Predicted away, actual away ✅
  • Coventry City 4-1 Newcastle United – Predicted away, actual home ❌

Five misses—three draws and two inverted results—highlight the volatility inherent in single-gameweek samples. Palace overturning Forest and Coventry dismantling Newcastle were both cases where the model leaned away and the underdog delivered. See the pre-match preview for the original probabilities. The data remains consistent over larger sample sizes, but individual rounds will always carry variance. Onward to Gameweek 7.

Half-time re-predictions this gameweek

At half-time the AI re-reads each match and re-predicts. Here is how those calls did against the pre-match calls, on the same set of Gameweek 6 matches.

Pre-match

5/10

50% correct

At half-time

6/10

60% correct

    Where half-time changed the call

    Aston Villa 2–2 Brentford

    Pre-match: Brentford ✗ · Half-time: Draw (2–2) ✓

    Why the call changed

    • xG 0.7 v 0.6
    • Shots on target 4 v 2
    • Possession 43% v 57%

    Ipswich Town 2–1 Fulham

    Pre-match: Ipswich Town ✓ · Half-time: Draw (1–1) ✗

    Why the call changed

    • xG 0.1 v 0.5
    • Shots on target 1 v 1
    • Possession 36% v 64%

    Manchester United 1–1 Tottenham Hotspur

    Pre-match: Manchester United ✗ · Half-time: Draw (1–1) ✓

    Why the call changed

    • xG 0.1 v 1.0
    • Shots on target 1 v 2
    • Possession 43% v 57%

    Hull City 2–2 Everton

    Pre-match: Hull City ✗ · Half-time: Everton (0–1) ✗

    Why the call changed

    • xG 0.6 v 0.9
    • Shots on target 3 v 3
    • Possession 43% v 57%

    Crystal Palace 3–1 Nottingham Forest

    Pre-match: Nottingham Forest ✗ · Half-time: Crystal Palace (3–2) ✓

    Why the call changed

    • xG 1.6 v 1.9
    • Shots on target 4 v 4
    • Possession 48% v 52%

    Liverpool 0–3 Manchester City

    Pre-match: Manchester City ✓ · Half-time: Draw (1–1) ✗

    Why the call changed

    • xG 0.4 v 0.1
    • Shots on target 2 v 1
    • Possession 46% v 54%

    This is a single gameweek — far too small a sample to read anything into, in either direction. See the running pre-match vs half-time comparison on our accuracy page for the season picture.

    Share this verdict:

    Prediction figures in this article reflect the data available at the time of publication and may have shifted since. See the live Gameweek 6 fixtures for the current numbers on each match.

    See how the AI called it before kickoff:

    Read the Gameweek 6 preview →