Can you trust an NFL player's L5 hit rate early in the season?
Not on its own. On our NFL week 3 board, 435 of 496 player lines had exactly two games from this season behind them, so an L5 was mostly last season's games: 452 of the 496 L5s reached back into last season, some into last January's playoffs. On 35 lines the L5 and this season's two games disagreed by 50 points or more. Read the L5 next to this season's split and ask what changed.
An L5 hit rate looks like recent form. In September it mostly is not. Three weeks into an NFL season, most players have two games on this year’s record, so the last five games are mainly last season’s.
We counted exactly how much, on our own board for NFL week 3.
How many games are behind the number
496 player lines on the week 3 board, across receiving yards, receptions, rushing yards and passing yards.
| Games this season | Lines |
|---|---|
| 2 | 435 |
| 1 | 55 |
| 0 | 6 |
452 of the 496 L5s, 91%, reached back into last season. The median player on these lines had 16 games from last season on file. For a player with two games this season, an L10 is eight earlier games to two of this season’s.
That is not a flaw in the number. A recent-form window should be the last five games a player actually played, not the last five dates on a calendar. It just means an L5 in September is mostly a statement about last year.
When the two seasons disagree
On 35 lines, the L5 and this season’s two games disagreed by 50 points or more. They go both ways:
- 14 lines had an L5 of 60% or better with zero of two hits this season.
- 21 lines had an L5 of 40% or worse with two of two this season.
Two real examples from the board:
Cooper Kupp, receptions over 2.5. L5: 60%. The five games behind it were 5, 4, 6, 2 and 2 catches. The first three, all hits, were last January and February’s playoff games, the last of them the Super Bowl. The two misses are this season. The L5 said 60%. This season said zero for two.
Travis Kelce, receiving yards over 51.5. L5: 40%, off 6, 36, 12, 71 and 101 yards. The three misses closed out last season. The two hits, 71 and 101, are this season’s only games. The L5 said 40%. This season said two for two.
Neither number is the right one. Two games is a small sample in either direction. The point is that “L5: 60%” and “L5: 40%” were each hiding a split that a reader deciding in week 3 needs to see.
What to read instead
- Put the L5 next to this season’s split. If they disagree by a lot, you are looking at two different stories and the L5 is averaging them.
- Ask what changed. New team, new quarterback, a bigger or smaller role. Target share and snap share tell you more in week 3 than a five-game hit rate built mostly on last year.
- Treat last season as a prior, not as form. For a player in the same role on the same team, it is often the biggest sample you have. It just is not recent.
The same trap in baseball, and why the size of the window matters more than people think, is in should you trust L5 and L10 prop trends. The full research order we use is in how to research a player prop.
See the split on a real player
On a player’s page in the Lab, the window buttons under the chart carry the year, and each season gets its own hit rate chip under it. Flip from L5 to this season and the sample shrinks to what the player has done this year. Cooper Kupp’s receptions shows the whole problem on one screen: three hits from last year’s playoffs, two misses this September.
What we counted
Our board for the NFL week 3 games of 27 and 28 September 2026, as published on Friday 25 September. 496 player lines, one per player per market, in receiving yards, receptions, rushing yards and passing yards. For each line we read the hit rate block the board itself shows: games this season, games last season, and the last five games played across both. Postseason games count as games played, which is how Kupp’s playoff games entered his L5.
No sportsbook was contacted. Numbers come from our own board for NFL week 3, 2026.
Quick answers
How many games are behind an NFL hit rate in week 3?
Usually two from this season. On our week 3 board, 435 of 496 player lines had exactly two games this season, 55 had one and 6 had none. Anything longer, an L5 or an L10, is filled out with last season's games.
Does an L5 hit rate include last season?
On our board, yes, deliberately: a recent-form window is the last five games the player actually played. In week 3 that reached into last season on 452 of 496 lines, and for some players into last season's playoff games.
What should I look at instead of the L5 in September?
The L5 next to this season's own split, and whatever changed since last season: team, quarterback, snap share, target share. On 35 lines on our week 3 board, the L5 and this season's two games disagreed by 50 points or more.
When do NFL hit rates become reliable?
There is no week when a number flips from unreliable to reliable. What changes is how much of the window is this season. By week 6 an L5 is mostly this season's games, and by then a hit rate is at least describing the current team.
Is last season's hit rate useless?
No. For a player in the same role on the same team it is often the biggest sample you have. The trouble is reading it as if it were this season's, which is what an unlabelled L5 invites in September.
Want tonight's board?
The Lab prices every player prop on the slate, shows the form and the matchup behind each line, and posts a short card before first pitch. Every pick is graded in public afterwards, win or lose. It is free and there is nothing to sign up for.