Taking a look at Flames game scores after six games

The Calgary Flames have gotten off to a fairly good start in the standings, winning four of their first six games of the season. There have been some strong individual performances and some weaker ones.

As we sit in a brief lull in the schedule, let’s dig into individual Flames performances thus far using the statistic known as game score.

About game score

Game score was conceptualized by Dom Luszczyszyn, currently of The Athletic, in an attempt to capture player performance in one comparable number. Here’s how he summed it up for Hockey Graphs:

It’s not a perfect stat – it’s meant to be a rough measure after-all – but I do think it works well for its intended purpose and is effective at what it does. It’s meant to answer “who had the best game” by adding proper perspective to a combination of a player’s total contributions and into an easily understood all-in-one stat.

For skaters, game score uses goals, primary assists, secondary assists, shots on goal, blocked shots, penalty differential, faceoffs, 5-on-5 corsi differential and 5-on-5 goal differential. For goaltenders, it uses saves and goals against.

The rundown

Here’s each Flames player’s average game score through six games. Anybody who was on the Flames last season has their 2016-17 game score in brackets for comparison’s sake. We’ve used the same comparative labels that we used in last season’s game score looks.

Great (0.95+):

1.450 – Mike Smith
1.138 – Johnny Gaudreau [0.857]
1.063 – Matthew Tkachuk [0.786]

Obviously these guys are outliers and bound to come down. But generally speaking, these have been the three most consistently good players for the Flames.

Good (0.45 to 0.95):

0.875 – Dougie Hamilton [0.810]
0.875 – T.J. Brodie [0.400]
0.813 – Mark Giordano [0.691]
0.751 – Sean Monahan [0.662]
0.685 – Michael Frolik [0.688]
0.566 – Mikael Backlund [0.777]
0.529 – Kris Versteeg [0.488]
0.500 – Jaromir Jagr (3 games)
0.470 – Freddie Hamilton (1 game) [0.109]

Hey look, the Flames have three defensemen in a cluster! Brodie’s made a big jump early on; his even strength numbers are a little bit better than last season’s, but his power play points inflate his number some. Jagr makes an appearance just a smidge below Versteeg, which makes a lot of sense given his usage and ice time. It’ll be interesting if his average score creeps up with more time playing with Monahan and Gaudreau.

Fine (0.15 to 0.45):

0.246 – Travis Hamonic
0.243 – Micheal Ferland (5 games) [0.312]
0.164 – Matt Stajan (5 games) [0.242]
0.160 – Curtis Lazar (4 games) [0.445]

Hamonic doesn’t get power play time, so this placement kinda makes sense. Ferland seems way low based on who he plays with. And here begins the march of the depth players.

Bad (-0.15 to 0.15):

0.121 – Michael Stone [0.297]
0.038 – Matt Bartkowski [-0.053]
0.027 – Troy Brouwer [0.166]
-0.059 – Tanner Glass (5 games)
-0.128 – Sam Bennett [0.280]

It’s the depth players, and Bennett. Bennett has zero points and leads the team in penalty minutes, which somewhat explains his bad numbers. Stone and Bartkowski seem low but have gotten slammed possession-wise (a product of their iffy defensive play).

Awful (below -0.15):

-0.250 – Eddie Lack (1 game)
-0.270 – Garnet Hathaway (1 game) [0.139]

Gigantic, blinking “sample size” warnings here. Lack played one game in relief during garbage time, while Hathaway played fourth line minutes and then was shunted to the AHL to open a roster spot for Jagr.

  • buts

    Off topic, an oiler nation commenter said that drysaddles concussion was courtesy from the boyfriend of a girl he was trying to pick up at a cactus club bar in chuck city…..must have been a flames fan 🙂

    • Nick24

      Even strength is much more important though, and if Stone is supposedly a top-4 defenseman, he should be more than effective against 3rd pairing competition. Just because he hasn’t been on the ice for a goal against doesn’t mean that he’s positively contributing to the PK. He seems like the worst of the four regular defensemen on the PK by expected metrics and chances, too.

      • The Fall

        Even strength is not ‘much more important’. Almost 25% of all goals are scored during a man advantage and that trend is going up. This season is also proving to be an outlier as far as penalty calls — Much like after the league cracked down on interference.

        A savvy coach looking for an edge might look at the historical precedents of rule changes and spend even more practise time woking on special teams, as there is likely to be a huge increase of time 5v4 / 4v5 this season.

    • OYYC

      Player Game Score = (0.75 * G) + (0.7 * A1) + (0.55 * A2) + (0.075 * SOG) + (0.05 * BLK) + (0.15 * PD) – (0.15 * PT) + (0.01 * FOW) – (0.01 * FOL) + (0.05 * CF) – (0.05 * CA) + (0.15 * GF) – (0.15* GA)

      So the players who get points with PP time tend to have higher game scores. The guys who play on the PK tend to have lower game scores. Truly riveting stuff. I’ll be saving my pennies to take an advanced analytics course for sure now!

  • BringtheFire 2.0

    Jagr’s number can increase on corsi alone. And in general whatever line he’s on can hold the offensive zone for a shift, but so far they haven’t been able to get scoring chances. Just good cycling. As Jagr gets his footing there’ll be goals.

    I think he can be the guy who helps Bennett most, on and off the ice.

    • BendingCorners

      The only 5-5 components are Corsi and goal differential. I’m not sure of the weighting. I see three names at the bottom of the list that to the eye look like they should be ranked higher – Bennett, Bartkowski, Stone. Bennett’s problem is he skates into traffic and doesn’t look for a good passing option so maybe he deserves his rating but he does a fair job on the forecheck and in his own zone. Bartkowski and Stone are undervalued by Game Score because they don’t score and don’t drive play up the ice, they just stop the other team from scoring , especially on the PK. Even Brouwer hasn’t looked bad so far (although that may be just relative to expectations) and probably should rank higher than the three “fine” forwards in the list above. Sample size will help adjust the numbers to a more representative value, but won’t help the biases built in to the formula. We need some kind of fudge factor called “eye test” to adjust the numbers with.

      • Nick24

        If Stone and Bartkowski don’t score or drive play then how do they stop the opposition from scoring? It’s more likely that Mike Smith has been covering up for a lot of mistakes. Of the players in the “bad” area, Sam Bennett is likely the only one who sample size may positively impact. Brouwer and Stone are at about what we can expect, as it falls in line with their career averages. While Bratkowski is actually trending above his normal impact, so it would be reasonable to expect him to dip down a bit.
        The tough thing about eye test is that it’s subjective. My eye test has the group with the “bad” game score adequately graded. Someone else’s may see them as being worse. There are some people out there who see Gaudreau as being awful. The problem is figuring out who’s eye test is closest to reality.

      • The Doctor

        Couldn’t agree more about Bennett skating into traffic and not looking for a good passing option. He needs to clean that up, it’s easily fixed as long as he’s coached properly and uses his head & a bit of hockey common sense.

  • The Doctor

    I like Bennett a lot and think we should stick with him, but he has been playing kinda stupid at times. I mentioned this before, but he needs to simplify his game IMO. Instead of trying to be a hero and trying to barge through or deke out defensemen, he needs to focus on simple, basic stuff like keeping possession, stopping and distributing the puck etc. There are a bunch of times where he’s been outnumbered on a rush and he’s tried to barge through two defenders and has predictably not succeeded and turned the puck over. That does the team no good. Earth to Sam: the offensive kung fu that worked in juniors doesn’t work in the NHL. The players, and in particular the defensemen, are too good. You’re in the big leagues now. Play like you’re in the big leagues. Play smart.

  • Sven

    Wouldn’t you love to know where Mark Jankowski would be if he started with the club in Edmonton?

    and does anyone really think he could be worse than Tanner Glass?

    • BringtheFire 2.0

      I just realized that we have maybe 20-30 points on his stick in the AHL and we’re content to leave him there and win anyway.

      He’d be a number one centre in Montreal.

      I think we should all be very grateful that these are the kinds of problems we have! 😀

  • Derzie

    For me, the game scores are pretty close to the eye test. Again, the outliers are the D specialists. Hamonic and to a lesser degree Stone. are under-scored. Hamonic is defensively good as ES and PK but Stone has had a great PK run but has been terrible at even strength. Kulak would help but some is still on Stone.

  • deantheraven

    I’m warming up to this stat but I’m sick of hearing arguments in corsi . Ther’s been very little to correlate Win-wins and corsi-wins so far this season. In fact, most corsi wins have been losses. Give me goals, shots/saves- for and -against. That’s quantifiable data that reads well.