FootballBlogger.org Week 3 Report
Week 3 covered 16 games. The market’s selected prediction was correct in 8 games and wrong in 8, matching the favorite’s 8–8 win/loss record.
The average Brier score was 0.2600. Lower Brier scores are better. The 50/50 baseline is 0.2500, so this week’s result was 0.0100 worse than that baseline. In other words, the predictions did not beat a simple 50/50 approach this week.
Best Prediction
The best prediction was KC @ MIA. Kansas City won 24–10 after receiving an 85.5% prediction probability. Its Brier score was 0.0210.
Other strong results included:
- ARI @ SF: San Francisco was predicted at 77.5% and won. Brier score: 0.0506.
- LAC @ BUF: Buffalo was predicted at 75.5% and won. Brier score: 0.0600.
- NYJ @ DET: Detroit was predicted at 72.5% and won. Brier score: 0.0756.
Biggest Miss
The biggest miss was SEA @ WAS. Seattle was given a 76.5% probability but lost 33–31 to Washington. The Brier score was 0.5776, the highest of the week.
Other notable misses included Philadelphia losing to Chicago after being predicted at 68.5%, and Atlanta beating Green Bay despite Green Bay receiving a 69.5% probability.
There were no missing snapshots. The permanent snapshot closest to kickoff was available for all 16 games.
Bet the House Benchmark
Bet the House is a deliberately naive fictional benchmark. It starts with $10 per eligible game, divides the entire weekly starting bankroll equally across all eligible market favorites, and uses the final permanent snapshot midpoint as the assumed purchase price. Fees are not included.
For Week 3:
- Starting bankroll: $160.00
- Hypothetical wagers: 16
- Record: 8 wins and 8 losses
- Amount wagered: $160.00
- Weekly P/L: -$38.49
- Ending bankroll: $121.51
- Return: -24.06%
The benchmark’s largest winning wager was MIN @ TB, which produced +$9.05. Other notable wins included TEN @ NYG at +$8.02 and NE @ JAX at +$6.81.
The losing wagers each lost the full $10.00 stake. Those losses included the favorites in SEA @ WAS, PHI @ CHI, ATL @ GB, and several other games where the selected favorite did not win.
This benchmark is fictional. No real money is wagered, and it is not betting advice. Results are reported whether they are good or bad, and the rules are not changed retroactively.
What We Learned This Week
Week 3 was an even split for favorites and predictions: 8 wins and 8 losses. The strongest calls were confident favorites such as Kansas City, San Francisco, Buffalo, and Detroit.
But several substantial upsets, especially Washington’s win over Seattle, pushed the average Brier score above the 0.2500 50/50 baseline. The fictional Bet the House benchmark also had a losing week, ending at $121.51 from a $160.00 starting bankroll.