Matthew Savoie was acquired by the Edmonton Oilers in the summer of 2024 from Buffalo in exchange for Ryan McLeod and Tyler Tullio. The rookie has put up three goals and four assists in 26 games played so far this season.
Savoie has spent roughly 109 of his 310 minutes of 5v5 ice time with Connor McDavid and while Savoie is not yet putting up the points we might have hoped for him, especially with such a potent line mate, I wouldn’t consider him to be a disappointment either. I personally have appreciated his speed and hard work in the corners.
On the stats side
The reason I am taking a look at Savoie today is because it recently came to light that, according to the fine creators of advancedhockeystats.com, Matthew Savoie has a 0% WAR.
WAR stands for “wins above replacement” and in this case is a combination of a player’s offence, defence, power play, penalty kill, finishing, goals, first assists, and penalties. It is meant to represent how many more wins a team has with a specific player than if that player were replaced by the next man up.
It is a complex, and non-standard stat, you will not find it on the NHL website and there is no one way to calculate it. It was originally invented for baseball and there is a part of me that believes it should stay there where a win can be attributed more confidently to an individual’s stats. That being said, the idea behind the stat is extremely enticing and is essentially what every sports statistician is chasing: a way to accurately quantify a player’s value.
I trust the creators of advancedhockeystats.com, they put out great content and have, perhaps, gone further in hockey analytics than anyone else whose results are publicly available. However, the actual formula they use to calculate WAR is not public and even if it were, it is likely calculated from raw data that needs to be crunched before being usable—that kind of crunching can be a lot of work.
Looking at the actual stats from the card now, each stat is a percentile. Savoie is in the 67th percentile in EV offence which means he has better offensive numbers than 67% of forwards. For me that is better than expected—his xGF/60and GF/60 are respectable among the Oilers.
His defence is unfortunately in the dreaded third percentile. Savoie does have the second worst xGA/60 on the team at 3.28 and the fifth worst GA/60 at 3.58.
Another sore spot is his finishing, which is largely based on goals above expected. Do you score most of the goals that are great chances? Do you score some goals on the low-probability chances? Savoie does have one of the worst 5v5 goals above expected in the league at -3.6. The good news is that that is very likely to change; Savoie’s career data wouldn’t fill a thimble and this could easily be attributed to bad luck.
Lastly is his low goals and first assists, both of which are based on /60 rates. Savoie is also well on the low side in that category. He is trending for about 10 goals this season and technically zero primary assists since he doesn’t yet have one.
On the bright side, Savoie draws far more penalties than he takes and has played against relatively difficult opponents which is reflected in the competition stat.
Bet like a pro here with Bet99—Ontario’s go-to for elite odds and nonstop action.
And outside Ontario, Canadians bet better
here with Bet99—bringing top-tier odds and action from coast to coast.
Why the last section doesn’t matter
One thing to keep in mind here, and this is extremely important, is that there is not much data on Savoie yet, having only played 31 total games in the NHL so far. This WAR is generally calculated as a weighted average of the last three years of play, which essentially don’t exist for Savoie. One reason stats are so much more powerful in baseball is because of how much data they have—they play twice as many games as in hockey!
Every statistician in the world would tell you that Edmonton is likely to play better in the relatively near future, simply because they are playing so much worse than they have in the recent past. Edmonton currently gets scored on third most in the league and scores the 12th most goals. There is a blatant parallel between Savoie’s and Edmonton’s offensive and defensive rankings. Savoie’s individual stats will get better as Edmonton’s do, I can promise you that.
Another thing to remember is that Savoie is still very young. While he may not be putting up spectacular numbers yet, he can still become a better player.
In closing, recently Vancouver Canucks Head Coach Adam Foote said that one shouldn’t take stats too seriously because they may “come from people that could be in Russia in a basement having cigarettes and cashews.” His point being that they aren’t close enough to the game to actually know what’s going on and, in this case especially, I believe he is right. Savoie’s poor stats are far more due to an extremely small sample size and the team itself doing poorly. Sure he might have some things to improve on but I like what I see so far.
stats from NHL.com, moneypuck.com, naturalstattrick.com, and advancedhockeystats.com
Like this:
Loading…
