Analytics Profile: Zdeno Chara

This is the fourth of sixteen player profiles. If you missed the introduction, you can find that here. Special thank you to Corsica, Evolving Wild, Micah Black McCurdy, for the data and visualizations

Shot Creation : B

3 year WAR/82 : 0.21

3 year RAPM Z-Score : 0.57

Zdeno Chara has been a slightly above average shot creator over the last three seasons. A simple visit to Corsica might not show it, but advanced models dictate that the captain is useful at creating shots.

The Bruins have shot the puck at a slightly higher rate with Chara off the ice over the last three seasons, but the margin is so small that it’s not very noteworthy. And given Chara’s usage, having a null impact on shot rates is somewhat impressive. The WAR and RAPM models pick up on his zone starts and quality of competition, which leaves him at slightly above average in this aspect.

Shot Suppression : B-

3 year WAR/82 : -0.06

3 year RAPM Z-Score : 0.12

Similar to the Bergeron profile last week, the world might end for this one. It appears, at his current state, that Chara is only a slightly above average shot suppressor. Please don’t shoot the messenger, but outside of his regularized adjusted plus-minus there isn’t really a positive stat for his shot suppression.

Similar to shot creation, the Bruins give up slightly fewer shots against per hour without Chara, but the difference isn’t too noteworthy. The concern for we is his wins above replacement in this realm. Over the last four seasons, the WAR model has Chara at slightly below replacement level each year in this aspect of the game.

Finding Chara below average here seems like a stretch. Between the RAPM, relative shot rates, and his reputation as a whole, I think Chara earns a B- here.

Offensive Shot Quality : C-

3 year WAR/82 : -0.04

There’s really not much to say about Chara’s offensive shot quality. He’s had a null impact on expected shooting percentages on a team that’s struggled from a shot quality perspective over the last three seasons.

The WAR model does a better job of taking team effects into account, and finds him below average in this aspect. The Bruins have netted 0.85% more of their unblocked shot attempts with Chara on the ice over the last three seasons, but that is certainly due to the quality of the forwards he is playing with.

Defensive Shot Quality : B+

3 year WAR/82 : 0.17

As one would expect, Chara’s strength is on defense. Although we just discussed he might not be as good as many think at preventing shots, his ability to decrease the probability each one goes in seems to have a higher upside.

Although this hasn’t been true, Bruins goaltenders are expected to save 0.20% more unblocked shots with Chara in front of them over the last three seasons. Due to the voodoo that is goalies, they’ve actually saved less behind Chara, but that isn’t the captain’s fault.

Author’s Comments

So this is surely the most controversial profile in this series. For those who follow the analytics community, you know that quality of competition is a big debate. The argument “against” quality of competition is that it washes away with time. And while that is a true statement, and is even true for a player like Chara, we should be trying to adjust for quality of competition on a shift by shift basis. That’s where models like RAPM and WAR come in.

Someone suggested in the Bergeron profile that the WAR model should be redone because Bergeron was poor defensively. The model looks very highly on Bergeron, finding him to be one of the best forwards in the league, but assigned the credit to his offensive play. Considering Chara was consistently above replacement in shot creation, and consistently below in shot suppression, the model is assigning credit to the wrong aspects of the game.