Loading...
Save percentage, goals against average, wins. Those are the stats most fans know. But they only tell part of the story. Here's a plain-English breakdown of every goalie stat that matters, from the basics to the advanced metrics that separate a good goalie evaluation from a lazy one.
TL;DR
SV% is the most useful basic goalie stat because it isolates the goalie better than GAA or wins. But it still has blind spots. GSAA builds on SV% by factoring in workload. Advanced metrics like high-danger SV% and expected goals against go further by accounting for shot quality. If you only look at one number, make it GSAA. If you want the full picture, pair it with high-danger save percentage.
These are the stats you see on every broadcast and box score. They are a starting point, not the finish line.
The percentage of shots a goalie stops. If a goalie faces 30 shots and allows 2 goals, their SV% for that game is .933.
SV% = (Shots Faced - Goals Allowed) / Shots Faced
NHL average: ~.905 to .910. This is the most useful basic stat because it focuses on what the goalie actually controls: stopping pucks.
The average number of goals a goalie allows per 60 minutes of play. Lower is better, but this number is heavily shaped by the defense in front of the goalie.
GAA = (Goals Allowed x 60) / Minutes Played
NHL average: ~2.80 to 3.00. A goalie on a strong defensive team will post a lower GAA than an equally skilled goalie on a weak one.
The goalie's record. A win goes to the goalie of record when their team wins. OTL means the goalie was in net for an overtime or shootout loss.
Record = Games won - Games lost in regulation - OT/SO losses
The most misleading goalie stat. Wins depend on how many goals the team scores. A goalie can play great and lose 1-0, or play poorly and win 6-5.
GAA counts goals allowed, but goals allowed depend on how many shots a goalie faces and what kind of chances the defense gives up. A goalie who faces 35 shots per game will almost certainly allow more goals than one who faces 25, even if the first goalie is better.
SV% strips out the volume question. It asks: of the shots this goalie faced, what percentage did they stop? That makes it a better measure of individual performance. It still has flaws (it treats all shots equally), but it is a significant step up from GAA for comparing goalies across different teams.
Quick comparison
SV% is a good starting point, but it treats a breakaway the same as a point shot. These metrics go deeper.
Goals Saved Above Average. Takes the difference between a goalie's SV% and league average, then multiplies by shots faced. A GSAA of +20 means the goalie saved 20 more goals than an average goalie would have on the same shots. This is the best traditional stat for goalie evaluation because it rewards both quality and volume.
GSAA = (Goalie SV% - League Avg SV%) x Shots Faced
A model-based estimate of how many goals a goalie "should" have allowed based on the location, type, and danger level of every shot they faced. If a goalie allowed 100 goals but their xGA was 115, they saved about 15 more goals than expected. This is the foundation of GSAx (Goals Saved Above Expected).
Accounts for shot quality, not just shot volume
Save percentage on shots from the inner slot and other high-danger areas. This is where games are decided. League average for high-danger SV% is roughly .810 to .820. Elite goalies consistently post .830 or higher. This stat reveals who actually makes the hard saves and who benefits from facing mostly low-danger perimeter shots.
The best test of a goalie's ability on tough chances
Here is a rough tier breakdown for evaluating a goalie's SV% over a full season or a large sample of games. These numbers shift slightly year to year, but the ranges have been stable for the past decade.
Elite
.920+
Vezina-caliber territory. Only a handful of goalies hit this mark over a full season. These goalies carry their teams.
Above Average
.915 to .920
Solid starter range. A team with a goalie in this zone has a reliable backbone between the pipes.
League Average
.905 to .915
Roughly where most NHL starters land. Not a weakness, not a strength. Perfectly fine with a good team in front.
Below Average
.900 to .905
Starting to become a liability. Workable on a dominant team, but this goalie is not stealing games.
Struggling
Below .900
Over any meaningful sample, sub-.900 is a serious problem. Either the goalie is having a rough stretch or they are not an NHL-caliber starter.
Check the current goalie leaderboard to see where this season's goalies fall.
A save percentage above .920 is elite. The .915 to .920 range is above average. League average sits around .905 to .910 in most seasons. Anything below .900 over a meaningful sample is a red flag. Keep in mind that SV% can fluctuate early in the season, so sample size matters before drawing conclusions.
Goals against average is heavily influenced by the team playing in front of the goalie. A netminder behind a strong defensive group faces fewer shots and fewer high-danger chances, which lowers their GAA regardless of individual skill. Two goalies can play at the same level but post very different GAAs because of team context. SV% and GSAA do a better job isolating the goalie's actual performance.
GSAA accounts for workload. Two goalies can both post a .918 SV%, but if one faced 1,800 shots and the other faced 1,200, the high-volume goalie saved more goals above average. GSAA captures that difference by multiplying the gap between a goalie's SV% and league average by total shots faced. It rewards both quality and durability.
Start with SV% as a baseline, then look at GSAA to account for workload. If you have access to expected goals data, check high-danger save percentage and GSAx (goals saved above expected) to see how a goalie handles tough chances. Quality start percentage tells you about consistency. No single stat tells the full story, but combining SV%, GSAA, and high-danger SV% gives you a solid picture.