Goals saved above average compares the goals a goalie allowed with the goals a league-average goalie would have allowed on the same number of shots. This season’s leader is Jesper Wallstedt of the Minnesota Wild at +11.5, by our calculation from the NHL’s goalie table, games through Wednesday, November 4, 2026.
GSAA answers one narrow question: how many goals did this goalie keep out compared with an average NHL goalie facing his shot count? A positive number is better than average. A negative number is worse. The unit is goals, so it is easier to read than a save percentage gap. Wallstedt’s +11.5 means about eleven and a half goals kept out that an average goalie would have allowed, across his 15 games.
How to calculate goals saved above average
The calculation has three steps. Each one uses numbers you can find in a standard goalie table.
Step 1, league save percentage. Add up every save by every goalie, then divide by every shot against. This season the NHL table holds 74 goalies, who faced 13,555 shots and made 12,066 saves. That gives a league save percentage of .8902.
Step 2, an average goalie’s goals against on his shots. Multiply the goalie’s shots against by (1 – league save percentage). This is the number of goals a league-average goalie would have allowed facing the same volume.
Step 3, GSAA. Subtract the goalie’s actual goals against from the Step 2 figure. What is left is goals saved above average.
The worked example uses Wallstedt’s line:
| Step | Wallstedt |
|---|---|
| Shots against | 387 |
| Saves | 356 |
| Save percentage | .920 |
| League save percentage | .8902 |
| Goals an average goalie would allow on 387 shots | 42.5 |
| Goals he allowed | 31 |
| GSAA | +11.5 |
An average goalie at .8902 lets in 42.5 goals on 387 shots. Wallstedt let in 31. The difference, 42.5 minus 31, is his +11.5.
GSAA leaders and trailers this season
The table covers goalies with five or more games, 56 of them, and shows the five best and the three worst by GSAA.
| Goalie | Team | Games | Shots against | Goals against | Save % | GSAA | GSAx |
|---|---|---|---|---|---|---|---|
| Jesper Wallstedt | Minnesota Wild | 15 | 387 | 31 | .920 | +11.5 | +12.1 |
| Linus Ullmark | Ottawa Senators | 8 | 182 | 10 | .945 | +10.0 | +8.3 |
| Jacob Fowler | Montréal Canadiens | 9 | 267 | 20 | .925 | +9.3 | +9.8 |
| Joel Hofer | St. Louis Blues | 11 | 316 | 26 | .918 | +8.7 | +5.1 |
| Pyotr Kochetkov | Carolina Hurricanes | 11 | 230 | 18 | .922 | +7.3 | +9.0 |
| Kevin Lankinen | Vancouver Canucks | 9 | 219 | 43 | .808 | -18.9 | -15.0 |
| John Gibson | Detroit Red Wings | 11 | 235 | 33 | .860 | -7.2 | -5.9 |
| Karel Vejmelka | Utah Mammoth | 12 | 281 | 37 | .872 | -6.1 | -7.6 |
Kevin Lankinen of the Vancouver Canucks sits last at -18.9, with a .808 save percentage over 9 games. The qualifying sample runs from 5 to 16 games, so one bad night can move a goalie several goals in either direction.
GSAA and GSAx
GSAA counts every shot as equal. A wrist shot from the blue line and a rebound from the crease each carry the same league-average chance of going in. Goals saved above expected (GSAx) fixes that by weighing each shot with an expected-goals model. Here the model is MoneyPuck’s: GSAx is the expected goals of the shots a goalie faced minus the goals he allowed.
When the two numbers agree, the shot mix was close to average. Wallstedt is at +11.5 GSAA and +12.1 GSAx, on 43.1 expected goals against. Those figures tell the same story.
When they split, shot quality explains the gap. Joel Hofer of the St. Louis Blues is at +8.7 GSAA but +5.1 GSAx, on 31.1 expected goals against. By the model, more of the shots he stopped were easier ones, so his GSAA flatters him by a few goals. Neither number is a forecast: both describe the games already played.
GSAx comes from the expected goals model. The most dangerous shots on their own are counted in high-danger scoring chances. For the shot volume a team allows in front of its goalie, Corsi counts every attempt. Our game-by-game calls are on the NHL predictions page.