Why Shots Predict Goals Better Than Goals Do: Reading a Player's Hot Streak

A skater who scored on a quarter of his shots last week is not a quarter-of-his-shots scorer. We measured four NHL seasons of our own box scores: how many shots a player takes holds steady, how many go in swings and settles back. Here is how to read a hot or cold run before a prop.

The short version

Goals come from shots. How many shots a player takes is a habit of his role, his ice time and his line, and it changes slowly. How many of those shots go in depends on bounces, goalies and luck over any short stretch, and it changes fast. So when a player is on a scoring run, the first question is whether he is shooting more or finishing more. Shooting more tends to last. Finishing more tends not to.

What the league looks like

In the 2024-25 NHL regular season, 10.7% of skater shots on goal went in. Among skaters with 100 or more shots, the 10th percentile scored on 5.2% of them and the 90th percentile on 17.3%. Real finishing skill exists, which is why the elite snipers sit near the top of that range every year. But the spread over a full season is about 12 points, and a 10-game stretch can swing far wider than that.

Volume repeats, finishing mostly does not

We split the 2024-25 season in half for the 260 skaters with 50 or more shots in each half. A player's shots per game in the first half and the second half correlate at 0.79: a player who shot a lot early kept shooting a lot. His shooting percentage in the two halves correlates at 0.38. Half a season of finishing tells you something, but far less than half a season of shot volume.

What happens after a hot 10 games

Across four NHL regular seasons, 2021-22 through 2024-25, we found 395 stretches of 10 games where a skater took at least 20 shots and scored on 8 or more points more of them than his season rate. In those runs he scored on 24.2% of his shots. In his next 10 games he scored on 11.1%, below his season mark of 13.0%, which itself includes the hot run. His shots per game barely moved, from 2.60 to 2.50. The goals dried up because the finishing came back down, not because he stopped shooting.

And after a cold 10 games

The same holds in reverse. We found 703 stretches where a skater scored on 6 or more points fewer of his shots than his season rate. He finished 4.4% of them in the cold run and 14.6% in his next 10 games, with shots per game moving from 2.67 to 2.48. A player taking his usual shots through a dry spell was still the same shooter.

How to read a streak before a prop

Look at two numbers side by side. First, shots per game over the last 10 against the games before: if it rose, the player has a bigger role or more ice time, and that tends to carry forward. Second, his finishing rate over the same two stretches: if the goals came from a finishing rate far above his norm, expect it to drift back. The sportsbook line already prices most of this, so treat it as context for the number you are looking at, not as a pick.

The same idea in other sports

Basketball works the same way with shot attempts and minutes against points per shot, and baseball with at-bats against batting average. Playing time and attempts are steady; what happens on each attempt is noisy over 10 games. We have only measured the hockey numbers above, so we will not quote figures for the others, but the same two-number check applies.

Where to see it on SharpCapper

The Player props screen on the Edge Board, a Pro and Elite feature, now shows volume and finishing for NHL, NBA, WNBA and MLB players. Open any player and you see shots, minutes or at-bats over the last 10 games beside the games before, the finishing rate for both, and one plain sentence when a run is unusually far from the player's norm. It sits beside each sportsbook's line and price and the player's hit rates. It never names a side to bet: past rates drift back toward the price, so it is research, not a pick.