How many games before an NHL player’s stats mean anything?
About 30 games for shots, and longer for points and goals. We checked about 500 NHL regulars in each of the last two seasons. After 20 games, a skater’s shots average this season was a better guide to the rest of his season than last season’s only 47 times in 100. For points and goals it was about 40.
Hockey settles much slower than basketball, where a scoring average is worth trusting after about ten games. A shot either goes in or it does not, a few times a night, and a month of that is still a small pile. Here is what we measured, the one stat that is different, and how to use both seasons at once.
What we measured
For every skater who played at least 40 regular-season games in a season and at least 40 the season before, we took his average over his first few games and asked a plain question. Which was closer to what he averaged for the rest of that season: those first few games, or his whole previous season?
The table shows how often this season’s early average was the closer one. Fifty means a coin flip. Under fifty means last season was the better guide.
| After this many games | 5 | 10 | 15 | 20 | 30 |
|---|---|---|---|---|---|
| Shots on goal | 32% | 43% | 46% | 47% | 54% |
| Points | 26% | 33% | 40% | 41% | 46% |
| Goals | 22% | 30% | 37% | 40% | 45% |
| Assists | 24% | 33% | 36% | 41% | 45% |
| Ice time | 53% | 57% | 60% | 62% | 65% |
Share of skaters whose average after that many games was closer to the rest of the season than last season’s average. Based on 499 skaters in 2024-25 and 500 in 2025-26, each with 40 or more regular-season games that season and the season before. Red is where last season was the better guide. Differences of a few points either side of 50 are inside the noise.
Two things stand out. Shots take about 30 games: last season is the better guide through 20, and this season only pulls ahead after that. Points, goals and assists are slower still. At 30 games last season was still the closer guide for all three, more often than not. Both season pairs pointed the same way.
The exception: ice time
Ice time is the one number where this season takes over early. From about five games, a skater’s ice time this season was the better guide, and by 20 games it was closer than last season’s 62 times in 100.
That makes sense. A coach changes a skater’s role between seasons: a new line, a power-play spot won or lost, a trade. His habit of shooting, how many shots he takes for each minute he plays, changes much less. So early on, lean on this season’s ice time and last season’s shooting.
What one big night is worth
After a single game, a skater’s shots number missed his rest-of-season average by a full shot or more 44 times in 100. Last season’s average missed by that much about once in 100.
A made-up but typical case. A forward averaged 2.4 shots on goal last season and opens with 4, 5 and 4. His next line goes to 3.5. If you bet the over at -110, you risk $11 to win $10, so you need to be right 52.4% of the time to break even. The number pushing you toward that over is a three-game average, which beat last season’s as a guide only 27 times in 100.
That does not make the under a good bet. The book saw the same three games and the same 2.4. It means three games are not the reason to bet either side.
Use both, not one
The better answer was often neither number alone. For shots, an even mix of last season and this season was closer than either from about ten games on, all the way to 30. For points and goals, last season alone was the better guide through about ten games, and the mix took over between 15 and 20.
Through November 10 of this season, the typical club had played 18 games. So most of the first two months sit inside the window where a skater’s own numbers so far are the weaker guide. A rough rule that fits the table: for shots, give last season half the say for the first 30 or so games. For points and goals, lean on last season alone for the first ten, then go to an even mix. For ice time, mix the two at first and let this season take over by about 20 games.
Where this stops being true
- It is about regulars. Everyone in the study played 40 or more games in both seasons. Rookies and skaters who barely played last year have no last season to lean on. For them, early games are all you have, and they are noisy.
- Role changes break it. A skater traded into a top-six job, or moved onto the first power-play unit, has a last season that describes a different job. Watch his ice time: it shows the change first.
- It compares averages, not bets. Being closer to a skater’s average is not the same as clearing a book’s line, and nothing here is a hit rate.
- Two season pairs is a modest sample. The direction was the same in both, and the shares near 50 could move a few points with another season.
What ET’s shots read does with this
ET’s shots-on-goal read is built from the same two pieces this page separates. Its shot rate, shots on goal per 60 minutes on the ice, is this season blended with last, so last season keeps a say while this season’s games pile up. Its ice time comes from a skater’s recent games, the one number this study says to trust early. Every read shows both, beside the book’s own chance at the line. The record of how those reads have landed is public on the NHL page.
Every research tab, the public record, and what we will not claim yet.
Questions
How many games until NHL player stats are reliable?
About 30 for shots on goal, and longer for points and goals. Across about 500 regulars in each of the last two seasons, a skater’s shots average after 20 games was closer to the rest of his season than last season’s average only 47 times in 100. It passed half at about 30 games. For points and goals it was still about 40 in 100 at 20 games.
Should I use last season’s stats for early-season NHL props?
Yes, alongside this season’s. For shots, an even mix of the two was closer to the rest of the season than either alone from about ten games on. For points, last season alone was the better guide through ten games, and the mix took over at about fifteen.
Does a big opening night mean anything in the NHL?
Very little. After one game, a skater’s shots number missed his rest-of-season average by a full shot or more 44 times in 100. Last season’s average missed by that much about once in 100.
Which NHL stat settles fastest?
Ice time. A skater’s ice time this season was the better guide from about five games, and by 20 games it was closer than last season’s 62 times in 100. Roles change between seasons. Shooting habits change much less.
Where can I see a skater’s season so far?
In the ET Parlays NHL Lab. The Player Lab page graphs one skater’s every game against any line you set, with his season profile, splits and game log.
Nothing here is betting advice, and no research changes the fact that these are negative-expectation bets. Bet only what you can afford to lose. If it stops being fun, take a break.