Does defense vs position matter for NHL shots on goal?

In one season of our logs, it lined up with a difference. Against the opponents most generous to their position, forwards finished 0.309 shots higher, relative to their own history, than against the stingiest. For defencemen, the gap was 0.220 shots.

Those are results from the 2025/26 NHL regular season, using the Steezanomics game-log store frozen on 16 October 2026. They are not comparisons with a posted line. The cause is unknown.

What we compared before each game

For every examined skater-game, we calculated two things using only games dated before that date:

1. The opponent’s shots allowed per game to the skater’s group: forwards or defencemen.
2. The skater’s own shots per appearance in that regular season so far.

We ranked the clubs from fewest to most allowed, then compared the skater’s result with his own prior average. For example, a player who had averaged two shots and then recorded three counts as +1 shot. A player who had averaged four and recorded three counts as -1 shot.

The allowance totals use the same definitions as NHL Defense vs Position: add up the opposing group’s shots in each game, then average those game totals. We do not average individual skaters’ averages. Forwards combine centers and left and right wings, using the position recorded for that game. Goalies are excluded. The live hub carries the current season; the historical figures here use its definitions with frozen 2025/26 rows.

Rank 1 means fewest allowed. The generous group starts at the fifth-highest allowance; the stingy group ends at the fifth-lowest. We include clubs tied at those boundaries. In the qualifying 2025/26 shot comparisons, a group therefore contained five or six clubs on a date, rather than breaking a tie alphabetically. Every qualifying date had measured rates for all 32 clubs.

The opponent had to have 20 earlier games, and the player needed 10 earlier appearances that season. The player also had to finish the examined game with at least 10 minutes. Earlier appearances under 10 minutes still count in his prior average. Twenty is this study’s opponent-history floor; it is not the minimum used by the live table.

That last minutes condition uses a completed game’s information. This is a description of that historical sample, not a rule for choosing players before puck drop.

Forwards: a 0.309-shot gap

The forward comparison covered 20,423 qualifying appearances by 501 players, from 17 November 2025 through 16 April 2026. The table below shows the two extreme opponent groups within that sample. Opponent membership was recalculated before every date, rather than taken from the final standings.

The generous-minus-stingy gap was 0.309 shots, and the share above the player’s own average differed by 8.7 percentage points. Equal-to-average games stay in the denominator: there were 36 in the generous group and 25 in the stingy group. The remaining 1,626 and 1,904 were below the player’s prior average.

The generous group contained 470 distinct forwards across 314 team-games; the stingy group contained 466 across 309. A player or club can appear in both groups at different times during the season.

Defencemen: a 0.220-shot gap

The defenceman comparison covered 11,222 qualifying appearances by 261 players, over the same 17 November 2025 to 16 April 2026 window.

That is a 0.220-shot gap and a 6.7-percentage-point difference in the share above the player’s average. Each group included 24 games exactly on that average; 921 and 1,039, respectively, were below it.

The generous group contained 250 distinct defencemen across 314 team-games; the stingy group contained 246 across 313. These are skater-games sharing opponents and game conditions, not thousands of independent games.

Points showed smaller differences

For points, we repeated the comparison with the opponent’s points allowed to that group and the player’s own prior points average. We did not reuse the shots ranking. The qualifying player population and dates were unchanged.

The points gap was 0.091 per forward-game and 0.057 per defenceman-game. Fifth-club ties expanded a points group to as many as six clubs for forwards and seven for defencemen on some dates.

For forwards, the generous group had 16 games equal to the prior average and 1,878 below; the stingy group had 17 equal and 2,028 below. For defencemen, those counts were 15 equal and 1,249 below, then 13 equal and 1,292 below.

The points groups contained 469 forwards across 313 team-games and 462 across 307, respectively. The defenceman counts were 248 across 321 and 249 across 313.

What made it into the sample

The frozen 2025/26 source contains 47,231 skater appearances by 940 players, across 1,312 games, from 7 October 2025 to 16 April 2026. SOG, points and minutes were present in every skater row. Every club had 82 regular-season games in the source.

We applied the conditions in the order shown below. The counts are identical for the shots and points studies because neither field was missing.

All earlier regular-season appearances inform a player’s own baseline, including appearances before the opponent qualifies. Same-date games never enter either prior. Playoff games and other seasons are excluded throughout.

As a check on definitions, all 128 club/group/stat combinations for shots and points matched the hub’s allowance rates, game counts and ranks when its method is run on the full 2025/26 season. For example, in this frozen 2025/26 season, Ottawa allowed 17.60 shots per game to forwards, rank 1 of 32, while Toronto allowed 23.74, rank 32. To defencemen, Carolina allowed 5.88, rank 1, and Toronto 8.70, rank 32. These final-season examples explain the table; they were not used to classify earlier games.

What the difference does and does not tell us

The direction was the same for both position groups and both stats in this one season. That is a reason to inspect the opponent’s allowance alongside the player’s own history.

It does not show that the allowance caused the result. Players, minutes, home ice, opponents and game conditions can differ between the two groups. Subtracting a player’s own average helps describe his result against his history, but it does not hold all those conditions constant. We have not established how much of the gap would repeat in another season.

Nor did we compare with sportsbook lines or prices. A player finishing above his own average may still finish below a prop line. These counts establish no betting return and no next-game probability. For the distinction between an average and a threshold, read what stats matter for NHL shots on goal props. For the same comparison in basketball, read does defense vs position matter for NBA props.

Start with NHL Defense vs Position. Choose forwards or defencemen, then shots on goal or points. Read the season and game count beside the allowance, remember that rank 1 is the fewest allowed, and compare that opponent context with the skater’s own history.