As Alexandra discussed in her latest post, the Syracuse Crunch currently sit on the top of the AHL leaderboard in many categories, both offensive and defensive. They're having a(nother) heck of a season, and it's been huge fun to watch. Unfortunately, the goalies in Syracuse are not having the best season. It's hard to say exactly why, and it doesn't seem to be a huge anchor weighing the Crunch down at the moment. But it bears further examination.
Dustin Tokarski: 0.907 SV% (25th in the league), 2.33 GAA (8th), 977:50 TOI (17 GP)
Riku Helenius: 0.884 SV% (46th), 2.83 GAA (31st), 594:06 TOI (11GP)
Right now, the Crunch are allowing a league-low 25 shots against (SA) per game, and limiting shots is generally a pretty good way to keep goals against (GA) low, yet they're 8th in the league in goals against per game (2.69). It's fairly easy, given these numbers, to lay blame on the goalies.
There are, however, stats that begin to test out the relationship between a goaltender's statistics and his team's play. One of these is the quality start (QS). (I talked more about this stat and how it was developed here.) The quality start metrics were developed for the NHL and were based on win percentages at certain levels of goaltender play. I borrowed this concept and applied this season's AHL averages to determine how the Syracuse goalies stack up. It's a quick and dirty translation, of course, but it gives some idea of where things stand.
The quality start metric is intended to determine how often a goaltender's play gave his team a good chance to win, and its converse, the "blow up" (BU) is intended to show how often the goaltender's play made it hard for his team to win. The third component, the "bail out" (BO) helps us tell how often a team is able to make up for a below average performance from the goaltender. Taken together, these are a measure of consistency that helps to put highly variable stats like save percentage and GAA into a bit of context.
As of Monday morning, December 17, the AHL's league save percentage for goalies with more than 5 games played was 0.910 and the league average GAA was 2.67. Thus, any game in which the goaltender performed better than these averages is counted as a quality start. In practice this means a QS was awarded for any game in which the goalie saved at least 91.1% of the shots against OR stopped more than 88.4% of shots while giving up fewer than 3 goals. These are the games where the goaltender gave his team a good chance to win.
A blow up was awarded when the goaltender saved less than 85% of shots OR between 85 and 87% of the shots against while giving up 5 or more goals (that is, when they faced more than 38 shots). [Note: This is the NHL formula. Without figuring out win percentages for various levels, it's unclear if this over or undercounts blow ups.]
A bail-out was awarded when the team won without a quality start.
Tokarski: 15 starts
9 QS (60%) and 4 BU (27%), 2 of which were BOs (13%) (2 games fit none of these categories)
Tokarski also came on in 2 relief efforts, one with a 1.000 SV% and one with a 0.769.
- While the QS rate is high, indicating that Tik gives the team a good chance to win most of the time, the BU rate is also high. That's a mark of an inconsistent goalie. It has been either hot or cold for him.
Helenius: 11 starts
5 QS (45%) and 3 BU (27%), one of which was a BO (on Tokarski's relief effort). (3 games fit none of these categories).
- Helenius's QS rate is respectable, though not outstanding, and it reflects his struggles early on in the season. But once again, there's that 27% Blow Up rate, which is enough to raise some questions about his reliability.
I would caution that we shouldn't draw strong conclusions on this data. There have simply been too few games played for the randomness inherent in hockey to even out. In addition, the Crunch tend to take a lot of penalties and extra time spent a man down often raises goals against. As the AHL doesn't distinguish even-strength SV% from special teams SV%, it's unknowable how much all those penalties are affecting the goalie stats.
But if you have watched Syracuse Crunch games and felt that the goaltenders have been inconsistent, doing either very good or rather poorly, the data bears that impression out. And this inconsistency is slightly concerning, though I wouldn't get alarmed about it, by any means. It's not currently a real problem for the team. Currently...
The worry is that offensive output is fairly variable as well, when taken on the whole. Right now, the Crunch are doing very well at generating both shots and goals, which is easier to sustain than high scoring on few shots. I haven't sat down to figure out the team's shooting percentage (and it's not available through the AHL) so it's hard to say how sustainable current performance is. If offensive output drops for a prolonged period, this inconsistency could become a problem.