This week on Frozen Tool Forensics we are going to take a quick detour from our recent focus on skaters and point production. Today we are looking at on goalies.

In most standard leagues we use bunch of goalie stats that goalies don"t necessarily have a lot of control over. Much of the value comes from volume like for wins and saves categories. Sure a 'good' goalie is likely to play more, but ultimately these are coaching decisions and team quality. The same goalie on two different teams could have very different deployment situations depending on the coach's preference and the quality of their backup. The team's strength can have a dramatic impact on the win totals, and on the shots against every game. So sometimes even in instances even a decent goalie who plays a high volume won't have as many wins or saves. Goals against average suffers some of the same issues as a team that lets a lot of shots against is going to give up more goals.

The one category that a goalie seems to have some control over is save percentage. This is still not a perfect category though as the quality of shot against clearly matters, and that is dictated by the team. When we are attempting to analyze save percentage then one thing we can take a look at is the goalie's short-handed save percentage. Typically, these result in much higher shot quality against, and therefore lower average save percentages. Additionally, since short-handed opportunities are a much smaller sample size, an individual's save percentage can see big fluctuations. These are usually not repeatable though do have an impact on a goalie's overall numbers.

As a quick case study let's look at Magnus Hellberg. He has played 10 games and let in 23 goals on 243 shots against. That gives him an overall save percentage of .907. He has a .919 short-handed though, which is very high. The average save percentage on the penalty kill this season is .862. If he performs at league average in his starts going forward and everything else maintains, his overall save percentage would drop to .899.

This is the idea we are going to tackle today. Which players are being most impacted by their performance on the penalty kill.

Given my personal schedules this week data was pulled on 2/21. That means some of the goalies will have played an additional game by the time of this release so these numbers will have some slight variations to currently posted data. The general trends will remain consistent though.

The following table contains basic player information and then in gray, season totals for goals against, shots against, and save percentage at all strengths. The next white section is the same data, but specifically for penalty kill (PK). The final gray column is season adjusted save percentage assuming a league average penalty kill save percentage. The list is sorted by largest change in season save percentage. 

As you can see by the far-right column, we aren't talking about great goalies turning into pumpkins or something. Generally, goalies who were in the .920's or teens stayed there, but clearly at a lower save percentage. The biggest impact is going to be on those borderline goalies and

Filip Gustavsson isn't necessarily a borderline guy given these percentages but because of his deployment. He has been in a battle with Marc-Andre Fleury all season. He seems to have a bit of an edge at the moment, but knowing that some of that success is because of an unsuitable PK save percentage does give me slight pause. If his deployment has been somewhat tenuous even given his over success, will he cede starts back to Fleury if he has a few poor games on the PK?

Cam Talbot has also been borderline this season. His inconsistent play and injuries have made it tough to rely on him. Fortunately, he has re-entered the lineup and with Anton Forsberg injured he is likely to get the volume going forward. Unfortunately, his already questionable save percentage looks due for a dip going forward. That is something to keep in mind when making the decision to roster him.

James Reimer is sort of the worst-case scenario here. He is below .900 even with an inflated PK save percentage. Kaapo Kahkonen was already starting to steal starts, and then Reimer went and got injured and Kahkonen played pretty well. If Reimer's play wasn't enough to hang onto the job before, potentially worse play going forward spell disaster. He did have a good game upon his return, but the long term trend isn't all that positive for Reimer.

The same theory is true in reverse. If goalies are having a particularly bad time on the penalty kill, they may be in line for a rebound in that area.

Spencer Martin being assigned to the minors with Thatcher Demko's imminent return makes this sort of moot here, but good to know he deserved better.

Philip Grubauer is definitely on the plus list across the board. He has already won the net back from Martin Jones with his first streak of four straight starts this season. He has put up good save percentages in four of the last six (which is a dramatic improvement for him). His .900 save percentage on the season is certainly trending up, and he deserves a chunk better if he could just a league average PK percentage.

It is worth noting that Spencer Knight deserves a little bit better, but he likely isn't getting the volume any time soon to be really taking advantage of it.

Jack Campbell and Ukko-Pekka Luukkonen are interesting here. Both have had inconsistent seasons but appear to be their respective team's starter, and might be due for slightly improved save percentages down the stretch. Campbell has a history of pretty wild fluctuations, and his improved season numbers still aren't great. Hopefully the goal support from Edmonton can give him some wins. Luukkonen has had fewer prolonged bouts of terribleness, but has a limited track record. If choosing between the two I would pick Luukkonen at the moment.

That is all for this week

Do your part to support organizations working to make hockey for everyone.

Black Girl Hockey Club

Hockey is For Everyone

Athletes for Hope

Premier Hockey Federation