Last 7, 15, 30, Season: What Do Hit-Rate Windows Actually Mean?

Open almost any player prop and you’ll see a row of numbers sitting next to each other: L7, L15, L30, and season. They look like four different stats, but they’re actually the same stat — a hit rate — measured over four different stretches of games. Learning to read them side by side, instead of grabbing whichever one looks best, is what separates a useful glance from a misleading one.

What each window is actually showing you

A window is just a sample size. L7 is the hit rate calculated from a player’s most recent 7 graded games. L15 and L30 do the same thing over 15 and 30 games. Season uses every graded game so far this year. None of these numbers is more “correct” than the others — they’re the identical calculation (Overs divided by graded games) run over different amounts of history.

The tradeoff is the same one that shows up anywhere sample size matters. A short window like L7 reacts fast: if a player has been unusually hot or cold lately, it shows up immediately. A long window like season absorbs more games, which smooths out streaks but also means it’s slower to reflect anything that’s changed recently — a new role, a return from injury, a shift in how a player is being used.

Neither is wrong. They’re answering slightly different questions: L7 asks “what has this player been doing right now?” Season asks “what has this player done overall?”

A simple example

Say a player’s strikeout prop has gone Over in 5 of his last 7 graded games — a 71% L7 hit rate. Zoom out to his full season, though, and the Over has landed in a smaller share of his starts — call it 45%. Same player, same prop type, two different numbers. (These are illustrative round numbers to show how the windows can diverge, not a posted line — real windows depend on the actual lines each of those games was graded against.)

Neither number is lying. The L7 window is honestly reflecting a hot recent stretch. The season number is honestly reflecting the fuller picture, streak included. Reading only the 71% and ignoring the 45% — or the reverse — is where the story gets distorted, not the math itself.

Why the windows disagree — and why that’s useful

A short window and a long window landing on very different numbers isn’t a bug. It’s information. It tells you a player’s recent form has moved away from their season-long baseline, in one direction or the other. Whether that shift reflects something real (a lineup change, a healthier body, a softer or tougher recent slate of matchups) or is just short-sample noise is exactly the question worth sitting with before you lean on either number alone.

The opposite case matters too: when L7, L15, L30, and season are all clustered close together, that agreement is itself a signal — it says the player’s recent stretch looks a lot like their full-season norm, for whatever that’s worth.

How to read the windows together

The habit worth building is comparison, not selection. Instead of asking “which window makes this prop look best,” ask “do the short and long windows agree, and if not, why not?” A recent window that’s drifted sharply from the season number is worth a second look at context — matchup, role, health — before you treat it as the headline. A recent window that lines up with the season number is a steadier, if less exciting, read.

You can see all four windows lined up together for real players and real lines on the player props hub, and check how a given game’s slate has actually played out in the game archive. If you’re weighing several legs at once and want to see how their hit-rate windows interact across a full ticket, that’s what the parlay analyzer is built to show.