The Bills' points with Ty Johnson and without Ty Johnson
, ,
Against what the season’s ratings expected, the Bills were 2.9 points over with Ty Johnson and 9.4 points over without Ty Johnson over their last 35 games, 31 games against 4. The plain averages are 30.0 and 33.5, a difference of 3.5. All 10 division games were with Ty Johnson. All 3 games after a bye were with Ty Johnson. All 25 games on a full week’s rest or more were with Ty Johnson.
Ty Johnson, running back: 400 rushing yards, 95 carries and 53 receptions in 31 of those 35 games.
points a game over expected
A −6.4 difference in points over expected with Ty Johnson vs without Ty Johnson is a bit unusual: shuffle which games were with Ty Johnson and it comes up 9 times in 100.
A −3.5 difference in points with Ty Johnson vs without Ty Johnson is about what you’d expect.
The games
| Side | Date | Game | Result | points | Expected |
|---|---|---|---|---|---|
| Dec 1, 2024 | vs SF | W | 35 | 30.3 | |
| Dec 8, 2024 | @ LA | L | 42 | 25.6 | |
| Dec 15, 2024 | @ DET | W | 48 | 24.8 | |
| Dec 22, 2024 | vs NE | W | 24 | 31.0 | |
| Dec 29, 2024 | vs NYJ | W | 40 | 29.4 | |
| Jan 5, 2025 | @NE | L | 16 | 30.3 | |
| Jan 12, 2025 | vs DEN | W | 31 | 25.9 | |
| Jan 19, 2025 | vs BAL | W | 27 | 27.5 | |
| Jan 26, 2025 | @ KC | L | 29 | 25.3 | |
| Sep 7, 2025 | vsBAL | W | 41 | 27.3 | |
| Sep 14, 2025 | @NYJ | W | 30 | 30.9 | |
| Sep 18, 2025 | vs MIA | W | 31 | 29.5 | |
| Sep 28, 2025 | vs NO | W | 31 | 27.0 | |
| Oct 5, 2025 | vsNE | L | 20 | 25.7 | |
| Oct 13, 2025 | @ ATL | L | 14 | 27.8 | |
| Oct 26, 2025 | @ CAR | W | 40 | 24.1 | |
| Nov 2, 2025 | vsKC | W | 28 | 25.0 | |
| Nov 9, 2025 | @MIA | L | 13 | 29.2 | |
| Nov 16, 2025 | vs TB | W | 44 | 27.3 | |
| Nov 20, 2025 | @ HOU | L | 19 | 21.2 | |
| Nov 30, 2025 | @ PIT | W | 26 | 26.2 | |
| Dec 7, 2025 | vs CIN | W | 39 | 31.9 | |
| Dec 14, 2025 | @NE | W | 35 | 22.8 | |
| Dec 21, 2025 | @ CLE | W | 23 | 26.0 | |
| Dec 28, 2025 | vs PHI | L | 12 | 25.8 | |
| Jan 4, 2026 | vsNYJ | W | 35 | 32.3 | |
| Jan 11, 2026 | @ JAX | W | 27 | 23.6 | |
| Jan 17, 2026 | @DEN | L | 30 | 23.1 | |
| Sep 13, 2026 | @HOU | W | 36 | 24.7 | |
| Sep 17, 2026 | vsDET | W | 41 | 25.2 | |
| Sep 27, 2026 | vs LAC | W | 24 | 26.3 | |
| Oct 4, 2026 | vsNE | L | 26 | 24.4 | |
| Oct 12, 2026 | @LA | W | 37 | 24.4 | |
| Oct 18, 2026 | @ LV | W | 33 | 27.3 | |
| Nov 1, 2026 | vsBAL | W | 36 | 25.4 |
Method
- Expected is an offense and a defense a team, fit over every game of its season between two teams in the table with the game being read left out, plus 0.9 in 2024, plus 1.1 in 2025, plus 0.2 in 2026 for the home team (the edge the fit finds once strength is in), nothing at a neutral site.
- Each expected number is then fit to the season’s games: a game bears out about 68% in 2024, 69% in 2025, 23% in 2026 of the difference the model predicts.
- The shuffle keeps the season’s shape: the games are cut in two where one date leaves the least variance, each game is read against its stretch, and the sides are shuffled only inside each stretch.
- Who played is read from the snap counts, which name every player who took a snap on offense, defense or special teams.