Now that fans of the Edmonton Oilers have had a few weeks to process what went wrong with the team, they’re focusing on a few factors, but one of them is NOT Connor McDavid and Leon Draisaitl’s defensive play.
In a Cult of Hockey poll on Twitter, I asked fans what the single most important factor in Edmonton’s play-in round defeat to the Chicago Blackhawks was.
Just 2.8 per cent said it was McDavid and Draisaitl’s defensive play.
The majority, 53.1 per cent, said it was the Top 4 d-men struggling, while 26.3 per cent pointed to weak goaltending, and 17.8 per cent pointed at Coach Dave Tippett’s decision to break up the NHL’s top two-way line of Leon Draisaitl, Ryan Nugent-Hopkins and Kailer Yamamoto, the Dynamite Line.
- We’ve got astute fans in Oil Country. My own first pick would also be the disappointing and weak play of the Top 4 d-men, Darnell Nurse, Oscar Klefbom, Ethan Bear and Adam Larsson, with Kris Russell also having struggles of his own, including deflecting the puck into the Edmonton. In a close second place, I would personally put Tippett’s decision to break up the Dynamite Line. How do you let a team like Chicago beat you without forcing them to stop the top two-way line in the NHL?
- Next I’d have goaltending, as Mike Smith and Mikko Koskinen generally failed to come up with big saves at the key moments. Finally, I’d list McDavid and Draisaitl’s defence. Neither of them was solid on defence, and I and other fans blasted both during the playoffs for their major errors. But I would not now single them out in this regard, as the entire team was a shambles on defence, not getting stick on puck, not charging the points to get into shooting lanes, and especially failing repeatedly to cover the slot in front of Edmonton’s net.
- On Twitter, I could only pick four options for folks to vote on. The top write-in candidate was “overall team defence,” which is probably correct, but rather non-specific. In terms of team defence, I think it’s fair to single out the Top 4 d-men and the goalies as having particular woes in the Edmonton end. All of those players performed much, much better in the regular season in terms of their defensive play.
- As for McDavid and Draisaitl, they were at about the same level in their own end in the playoffs as they’d been all year, which is to say not great, not even good, with too many big errors in slot coverage. That said, Draisaitl’s defensive play had picked up noticeably during the regular seasons after he was switched onto the Dynamite Line in late December. I suspect if Tippett had stuck with that excellent two-way trio, which scored 30 goals in the regular season and gave up just nine at even strength, we would not now be talked about Draisaitl’s defensive woes in the playoffs at all. They would have had the puck in the offensive end too much, and scored too many big goals, for that to be of much notice.
- If you read this blog and listen to the Cult of Hockey podcast, you’ll know that I have often complained of defensive lapses on the Oilers, and was negative about McDavid and Draisaitl’s defensive game in December. Overall, though, in rating their even strength contributions this year, the two were excellent two-way players, creating much more on the attack than they gave up in their own end — and that trend continued in the playoffs.
- Various stats are used to argue that McDavid and Draisaitl were weak defensive players this year. Fans of on-ice stats will point out that McDavid and Draisaitl’s Corsi, Fenwick and Expected Goals For percentages, according to Natural Stat Trick, were just 48%, 48.1% and 47.9% for McDavid, and just 47.5%, 47.7% and 47.4% for Draisaitl. Those numbers don’t scream out: NHL superstars. But I’d argue these numbers are low-information numbers, as misleading as they are relevant to rating the two-way play of players. How are they low-information? They accurately tell us that the Oilers were slightly outshot and were expected to be slightly outscored when McDavid and Draisaitl on the ice, but they don’t dig into McDavid and Draisaitl’s actual individual contributions. They tell us how the team did, but not what the individual players were doing. They lack that crucial information.
- For example, Draisaitl was on the ice for 80 goals for and 77 goals against at even strength this year, according to NHL.com, but those numbers don’t tell us what contribution, if any, he made to the goals for. They don’t tell us what mistakes, if any, he made on the goals against. Through videotape review, we tracked these individual actions at the Cult of Hockey, with me doing the initial review most nights and Bruce McCurdy going over my work and make corrections. We found that on those 80 goals, Draisaitl made 64 major contributions, where he was either the key shooter or passer on the play. On those 77 goals against, he made 28 major mistakes, most often leaving open his check in the defensive slot. His even strength plus minus was +3, but his individual plus-minus was +36, the highest number on the Oilers. Next best were Kailer Yamamoto and Connor McDavid, both +26.
- Of course, our numbers aren’t perfect signifiers of even strength prowess either. When you look at the Cult of Hockey’s individual goals plus-minus numbers, two pieces of crucial context are missing: 1) the quality of competition each player faced and 2) the amount of offensive opportunity and defensive responsibility assigned to each player. In the NHL, we expect forwards to put up more goals and assists than d-men, simply because they have more offensive opportunity. The same rule applies to our “major contributions to goals.” Forwards will generally have more of them because they have more opportunity.
- In the defensive end, d-men and centres will generally have more “major mistakes on goals against” because they have more defensive responsibility than centres, and centres have more defensive responsibility than wingers. To make sense of these numbers, it’s best to compare centres to centres, wingers to winger and d-men to d-men, which is why we list them that way.
- We can’t compare Draisaitl’s +36 to other centres in the NHL, as we don’t do video review of all NHL goals. But he led Edmonton centre/wingers, with McDavid at +26, Ryan Nugent-Hopkins +14, Gaetan Haas +1 and Riley Sheahan +4.
- If you got by on-ice numbers like Corsi, Fenwick and Expected Goals, Haas’s numbers were 48.31%, 49.1%, and 50.2%. But if you put weight in those numbers and think they’re telling you he slightly outperformed McDavid and Draisaitl at even strength, you’re not grasping that these on-ice numbers are misleading you because they lack vital information.
- Things get even more crazy when you rely on official NHL plus-minus numbers to rate individual players, as some NHL writers evidently do when deciding on their MVP votes. I’ve heard a few of them mention Draisaitl’s low official NHL plus-minus number of -7 in their argument not to vote for him. But in the NHL’s official total they include 12 short-handed goals against Edmonton when Draisaitl was on the power play. They also include 12 empty not goals against the Oilers when Draisaitl was on the-ice. This distorts the NHL’s official plus-minus number.
- The NHL includes the short-handed goals against because that’s how it’s always been done since Montreal Canadiens coach Dick Irvin Sr. and hockey’s first analytic expert Allan Roth teamed up in the 1940s to come up with the plus-minus stat. The stat was a huge conceptual leap forward in terms of hockey analysis, and it was the best that could be done before video review of goals and other key plays was possible. But Irvin and Roth’s decision to include shorthanded goals against was a mistake that has yet to be corrected. The NHL should do so now.
- In 2019-20, if short-handed goals against was dropped, Draisaitl’s official plus-minus would be +3, which is somewhat more indicative of his even strength play. Again, I would argue his +36 goals for individual plus-minus is most indicative of his two-way play, but without league-wide numbers for individual goals plus-minus we lack crucial context in using that number to compare him to other top centres. Some NHL teams almost certainly have this kind of information-rich data, or something like it, but they’re not sharing it.
At the Cult of Hockey
LEAVINS: 9 Things, including Holland’s search to improve the blueline
McCURDY: The in-depth details of the Joakim Nygard transaction
STAPLES: How to transform the D-corps in 3 (not so) easy steps
McCURDY: The return of Jesse Puljujarvi imminent?