How to Read Shooting Percentage in Hockey

Shooting percentage is the percentage of a player’s or team’s shots on goal that become goals. It measures conversion—not how often every attempted shot hits the net, and not shot speed. The NHL labels it S%. Its official glossary excludes missed and blocked shots from the calculation.

Shooting percentage = goals ÷ shots on goal × 100

A player with 20 goals on 200 shots on goal shoots 10%: one goal for every ten recorded shots on goal across that sample.

What counts in the denominator?

Shots on goal include goals and shots saved by the goalie. Goals are already part of the shot total; do not add them again. Misses and attempts blocked by opposing skaters do not count. A shot that hits only the post or crossbar and stays out is a miss, according to the NHL glossary.

For example, suppose a player records:

• 2 goals;
• 18 shots saved;
• 5 missed shots;
• 5 blocked attempts.

That is 20 shots on goal and 30 total attempts. The player’s shooting percentage is 2 ÷ 20 × 100 = 10%, not 2 ÷ 30. Goals per attempt would be a different metric.

Shootout percentage is also separate: the NHL calculates it using all shootout attempts, including misses and other failed attempts. Do not mix that figure with ordinary S%. For the related in-game edge case, see how penalty shots count as shots on goal.

When combining games, seasons or players, divide total goals by total shots on goal. Do not simply average the percentages: a two-shot game should not have the same weight as a ten-shot game. With zero shots on goal, the percentage is mathematically undefined—not evidence of poor finishing.

What is a good shooting percentage?

An NHL league baseline is useful, but it is not a universal player target. For the 2024–25 regular season, Hockey-Reference lists 3.01 goals and 28.1 shots against per team per game. Across the league, shots against equal shots for, giving an approximate overall conversion rate of 10.7% from those rounded averages. The source excludes shootout-deciding goals from its goals average. (NHL league averages)

That is a league-wide, shot-weighted rate—not the simple average of every player’s S%. Thus, 15% was above that broad league baseline, while 20% was substantially above it. Neither threshold, by itself, establishes a player’s finishing talent or total offensive value.

Consider three high-scoring forwards from that same season:

Player, 2024–25 regular season	Goals	Shots on goal	S%
Leon Draisaitl	52	240	21.7%
David Pastrnak	43	319	13.5%
Nathan MacKinnon	32	320	10.0%

Source: NHL regular-season skater statistics, filtered to 2024–25.

Pastrnak’s percentage was much lower than Draisaitl’s, yet his greater shot volume helped produce 43 goals. MacKinnon scored 32 at 10%. Conversion and volume must be read together.

For youth, junior or recreational players, use a same-league, same-season comparison where possible. An NHL aggregate should not become a development target for players facing different opponents and playing different roles.

Why a higher percentage does not automatically mean a better shooter

Shot quality changes the comparison. A close-range opportunity and a long, angled shot do not carry equal scoring probabilities. MoneyPuck’s expected-goals methodology considers location, shot type, the preceding event, manpower and whether the net is empty. Ordinary S% adjusts for none of these.

Deployment matters. Compare forwards with similar roles, or defensemen with defensemen, before treating a percentage gap as a skill gap. Separate 5-on-5 and power-play results when possible. The NHL glossary explains that even-strength or 5-on-5 comparisons reduce distortion from unequal special-teams usage; it also confirms that empty-net goals enter players’ overall totals.

Small samples swing sharply. Two goals on ten shots is 20%; one goal on ten is 10%. That ten-percentage-point difference is just one goal. The NHL’s sample-size guidance warns that short stretches can reflect good or bad fortune rather than underlying ability. MoneyPuck likewise distinguishes observed goal conversion from estimated shooting talent using a player’s longer record.

Before calling a percentage “good,” check the shot count, the player’s previous seasons, the game situation and the quality of the chances. Expected goals can help with the last question, but it remains a model estimate. Read S% as a description of what went in—not a complete explanation of why, or a guarantee of what comes next.