What If Getting More Zone Exits Won’t Help?

This last year I’ve been tracking zone transitions for the Oilers defencemen. I finished 31 games and will be planning on getting all 82 done next year (with help). One of the things I wanted to do this summer is analyze the data to see how these stats impact game-flow. Recently, I looked at Zone Defence and corsi. This time I take a look at how Zone Exits impact shot attempts.

In review, last time we figured out that Oiler defencemen who were better at preventing Controlled Zone Entries also had better corsi. We also saw that in games where the Oilers, as a team, prevented Controlled Zone Entries, they had better shot attempt differentials.

So What About Zone Exits?

Naturally, we should expect to see the same with Zone Exits. Indeed, if we plot Controlled Zone Exits with corsi we see…

… wait a minute. That doesn’t make any sense. There’s no correlation here at all. Even if I remove players with less than 5 games tracked, there’s still no correlation. Oiler defencemen who were better at getting the puck out with control did not necessarily have better corsi.

Okay, but there’s a lot of factors that impact a player’s shot attempt statistics. Surely if we look at games where Oilers D had better breakouts, we’ll see that they had more shot attempts. Here we plot Controlled Zone Exits by all Oiler defencemen per game with corsi.

Huh. Nothing. Slightly negative actually. (If we convert to Controlled Exit %, it becomes slightly positive, but still an extremely weak correlation… R² of 3%.)

I’ll save you guys a post full of graphs and just summarize the overall conclusion when I repeated this looking at corsi for, corsi against, and the percentage of Controlled Zone Exits instead of just the rate. In games where Oilers D were breaking out of the zone more effectively, they did not have any more shot attempts for or any fewer shot attempts against. Controlled Zone Exits by defencemen made no major difference in the overall shot attempts in the game.

Should They Just Dump The Puck Out Then?

What happens if we repeat this same experiment but with Uncontrolled Zone Exits? Well, it turns out that there is a correlation there. Oilers defencemen who dump the puck out more have a worse corsi. Mostly, this is due to more shot attempts against. The correlation is much weaker with shot attempts for. Also, in games where the Oilers D as a whole dump the puck out more, they have more shot attempts against (though it’s fairly weak correlation).

Enough With The Graphs! What Does It Mean?

So let’s sum this up. For Oilers defencemen…

1. Preventing opponents from entering the zone with control of the puck: much better corsi
2. Dumping the puck out of the zone more often: somewhat worse corsi (mostly more shot attempts against)
3. Exiting the puck with control more often: no difference in corsi

That last one really makes no sense.

Jen Lute Costella noted that getting the puck out of the zone with control led to going on the offensive 88% of the time. That should result in more shot attempts for. Yet, no matter how you slice it for the Oilers, the correlation is, at best, weak.

Any Explanations?

It’s possible there are other mitigating factors. Keep in mind that I’m not tracking all zone exits, only the ones done by defencemen. Maybe making a good zone exit pass, while an important skill, is not enough for a defencemen to move the needle in overall game flow. Other skills, like preventing controlled zone entries, may have a greater impact that overshadows it.

If this is true, then it has interesting implications for how we interpret defensive stats. It means that we can’t look at a defenceman’s corsi (whether as a percentage, corsi-for, or relative to team) and infer their puck-moving ability from that statistic. Just because a defenceman has good shot attempt numbers doesn’t mean they can move the puck out of their zone. The same is true if they have poor shot attempt numbers.

Practically, what this means is that, while Zone Exits are important, you shouldn’t overvalue that statistic TOO much. Take someone like Kris Russell. He reportedly led the Flames in defensive zone exits per game. However, he was a possession black hole. His Zone Exit skill alone wasn’t enough to make up for that.

Any OTHER Explanations?

Another possibility is that this is not an NHL wide phenomena. I don’t know if Zone Exits and corsi are correlated league-wide. Maybe it’s just the Oilers that show this oddity.

That is an even more interesting possibility. That means that problem is not the Oilers exiting the zone. The problem is converting those Controlled Zone Exits into successful Zone Entries. If this is where the Oilers are struggling, and these findings hold true, then getting better puck moving from the defence will only have a marginal impact on the Oilers ability to get more shot attempts.

Why the Oilers haven’t been able to convert their Zone Exits in Zone Entries is something I don’t have the answer to. My first thought was that the Oilers are dumping the puck into the offensive zone too often and not carrying it in. However, the only data that I’ve seen tracking Oilers Zone Entries (by J.D. Burke) has indicated that they carry the puck in about 40% of the time, which is around league average.

I have some ideas, but this is something I’ll have to look into more closely and follow for next season.

Summary

While defending the defensive blue line effectively and preventing Controlled Zone Entries leads to better shot attempt statistics, this has not been true of Zone Exits. The Oilers have not been able to convert their Control Zone Exits into better corsi.

I don’t know if this is specific to the Oilers or not. If it’s not, that means that we can’t infer anything about a defenceman’s puck-moving ability based on their corsi. If it is Oilers specific, that means that the Oilers are doing something else wrong. They need to figure out what that something is or getting more Controlled Zone Exits from the back end may not make any difference to their overall fate next season.

Save