Javonte Williams
RB
Dallas Cowboys
- Teammate George Pickens (a top target) is questionable.
A hurt top target or lead back nudges the projection up a little, since their work tends to shift to teammates. QB and offensive-line flags are shown for context but do not change the number.
This is how RBs have played against the Colts in the 2026 season.
RB1 = the highest-scoring RB against Colts in each game, not a fixed depth-chart spot.
| Wk | Team | Player | Car | RuYd | RuTD | Tgt | Rec | ReYd | ReTD | FPts |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 |  BAL | Derrick Henry | 24 | 144 | 3 | 1 | 1 | 19 | 0 | 35.3 |
| 2 |  KC | Kenneth Walker III | 24 | 117 | 0 | 10 | 6 | 61 | 0 | 23.8 |
| 3 |  HOU | Woody Marks | 5 | 15 | 1 | 1 | 1 | 6 | 0 | 9.1 |
| 4 |  WAS | Jacory Croskey-Merritt | 9 | 15 | 0 | 4 | 2 | 35 | 0 | 7.0 |
| 5 |  PIT | Jaylen Warren | 15 | 101 | 1 | 6 | 6 | 87 | 1 | 36.8 |
| 6 |  TEN | Tony Pollard | 19 | 76 | 0 | 1 | 0 | 0 | 0 | 7.6 |
| 7 |  MIN | Jordan Mason | 15 | 68 | 0 | 2 | 0 | 0 | 0 | 6.8 |
| 8 |  JAX | Bhayshul Tuten | 16 | 104 | 1 | 2 | 2 | 55 | 1 | 29.9 |
| Average | 15.9 | 80.0 | 0.8 | 3.4 | 2.3 | 32.9 | 0.3 | 19.5 | ||
| Median | 15.5 | 88.5 | 0.5 | 2.0 | 1.5 | 27.0 | 0.0 | 16.4 | ||
| High | 24 | 144 | 3 | 10 | 6 | 87 | 1 | 36.8 | ||
| Low | 5 | 15 | 0 | 1 | 0 | 0 | 0 | 6.8 | ||
Averages are raw, outliers are never removed.
How Javonte Williams should do vs the ColtsSeason log, the group you pick, and a projection
Season so far, compared with what the group you pick below has done against the Colts, plus a projection for this game.
What is this table?
Rows at the top are every game Javonte Williams has played this season, using the same stat columns as the table above.
Compare against picks which group of Colts opponents to line them up with: the best RB they faced in each game (1), the second best (2), the third best (3), or all of the top three. Nobody knows which slot Javonte Williams will land in this week, so this is your call, not a guess.
Colts avg is the average line of that group. Difference is that minus Javonte Williams's own average: green means the defense gives that slot more than Javonte Williams usually produces.
Projected is Javonte Williams's recent-weighted average scaled by how generous the Colts were to the slot you picked. Change the slot and the projected points change, because each slot has a different matchup adjustment. Picking Top 3 uses the usual-role projection.
| Wk | Opp | Car | RuYd | RuTD | Tgt | Rec | ReYd | ReTD | FPts |
|---|---|---|---|---|---|---|---|---|---|
| 1 | NYG | 12 | 41 | 1 | 5 | 5 | 31 | 1 | 24.2 |
| 2 | WAS | 12 | 30 | 0 | 4 | 3 | 20 | 0 | 8.0 |
| 3 | BAL | 19 | 98 | 1 | 3 | 2 | 5 | 0 | 18.3 |
| 4 | HOU | 19 | 62 | 3 | 5 | 5 | 21 | 0 | 31.3 |
| 5 | TB | 12 | 45 | 1 | 3 | 2 | 17 | 0 | 14.2 |
| 6 | GB | 11 | 41 | 1 | 2 | 2 | 9 | 0 | 13.0 |
| 7 | PHI | 25 | 155 | 1 | 0 | 0 | 0 | 0 | 21.5 |
| 8 | ARI | 24 | 56 | 0 | 2 | 1 | 1 | 0 | 6.7 |
| Williams avg | 16.8 | 66.0 | 1.0 | 3.0 | 2.5 | 13.0 | 0.1 | 17.1 | |
| Colts RB1 avg | 15.9 | 80.0 | 0.8 | 3.4 | 2.3 | 32.9 | 0.3 | 19.5 | |
| Difference | -0.9 | +14.0 | -0.3 | +0.4 | -0.3 | +19.9 | +0.1 | +2.4 | |
| Projected as RB1 | 19.4 | 76.3 | 1.2 | 3.5 | 2.9 | 15.0 | 0.1 | 18.1 | |
Difference is what the defense allows minus what Williams averages (green = defense gives up more). Projected stats scale the player's averages by the matchup adjustment, so they are a rough guide.
Start / Sit · Javonte WilliamsGood confidence · 20.8 proj pts7/10 start
- Javonte Williams averages 17.1 PPR points over 8 games (recent-weighted 15.7) and has played as the team's RB1.
- The Colts have allowed 19.5 points to RB1s over 8 games (+21% vs league). The model applies 73% of that gap because the sample is small, giving a +16% adjustment.
- Last season's average (15.2) is blended in at 20% weight because this season's sample is small.
- Game environment +14%: Vegas total 51.5, implied 28.5 points for DAL (league avg 22.6); Favored by 5.5: leaning run-heavy late; Indoors: no weather effect, a small boost for passing and kicking.
- Opposing-defense injuries (+1%): Opposing starter Akeem Davis-Gaither (LB) is questionable, which helps the passing game. Opposing starter Justin Walley (CB) is questionable, which helps the passing game.
- Teammate injuries (+0%): Teammate George Pickens (a top target) is questionable, which can mean extra looks for Williams.
- Projection: 20.8 points.
- For reference, a typical weekly starter-level RB scores about 13.1 and a top-tier one about 20.9.
- A statistical model from box scores, Vegas lines, weather and injury tags. It cannot see role changes or in-game events.
Prediction & chance to scoreJavonte Williams vs the Colts
Prediction for Javonte Williams
Nobody knows which slot Javonte Williams will land in this week, so pick the one you expect. The projection changes because each slot faces a different average from the Colts.
I predict about 18.1 points for Javonte Williams if playing as a RB1. Javonte Williams has averaged 17.1 over 8 games, and the Colts have given up 19.5 to RB1s (+21% vs the league average of 16.1).
How this projection is built
1. The player. Javonte Williams scored: Wk 1 24.2, Wk 2 8.0, Wk 3 18.3, Wk 4 31.3, Wk 5 14.2, Wk 6 13.0, Wk 7 21.5, Wk 8 6.7. Average 17.1; weighted toward recent games it is 15.6.
2. The defense. Over 8 games, the Colts allowed 19.5 points to their opponent's RB1. The league average for that slot is 16.1, so that is +21%.
3. Trusting small samples less. With 8 defensive games, only 73% of the gap is applied, capped at +/-30%. That gives a +16% adjustment (x1.16).
4. Result. 15.6 x 1.16 = 18.1. Switching slots only changes step 2 and 3: the player's own numbers stay the same, but each slot faces a different defensive average.
It is a statistical model, not knowledge of injuries, game script or weather.
Chance to score
Type a point total to see how likely Javonte Williams is to reach it as a RB1.
Why this percentage?
The projection (18.1) is the middle of the range. Fantasy scores swing a lot from game to game, so the model spreads possible scores around it by about 8.4 points, leaning on Javonte Williams's own game-to-game swings once there are several games (and on typical RB swings before that).
Scores are modelled as right-skewed, since a big game is possible but a very negative one is not. The percentage is the share of that spread at or above 18.
With only 8 games of history, treat this as a rough gauge. It gets sharper as the season goes on. It is never lower than 1% or higher than 99%.
InsightsPlain-English takeaways from the table
- RB1s have averaged 15.9 carries, 80.0 rushing yards and 2.3 receptions against the Colts across 8 games.
- The Colts allow 19.5 PPR points to RB1s, +21% versus the league average of 16.1 (ranked 3 of 32, where 1 is most generous). That is a favorable matchup.
- Outlier: Week 1, Derrick Henry (35.3 pts, 24 car, 144 rush yds) sits far above the other games and pulls the averages up. Average with it vs without it: pts 19.5 vs 17.3; car 15.9 vs 14.7; rush yds 80.0 vs 70.9. The table above keeps the raw averages.
- Outlier: Week 3, Woody Marks (5 car, 15 rush yds) sits far below the other games and drags the averages down. Average with it vs without it: car 15.9 vs 17.4; rush yds 80.0 vs 89.3. The table above keeps the raw averages.
- 4 of 8 games reached 10+ PPR points. Scores range from 6.8 to 36.8 (median 16.4).
- RB1s scored a touchdown in 4 of 8 games (50%).
Insights are generated by rules from the table above and recalculate every week.