Earlier in the week I wrote about Goaltending being the biggest Oiler story of the first 60 games, and I think it has been. I mean there has been a complete roster turnover and the quality of starts they were getting went from the bottom of the league to near the top (although only for a short time). Looking forward to the final quarter of the season the Oilers desperately need to find some form and build into next year. In terms of goaltending I think they have the necessary parts to do that, but there is a lot more to that goal than just stopping more pucks.
The next step in terms of building towards next year with these final 20+ games is going to be fixing their mounting possession problems. And this will be no small task. Unlike the goaltending, the Oilers aren’t trending in the right direction. The Coach knows it, the players know it, and something has to change right away to fix it.
POSSESSION
To start with, there is no perfect Possession Stat. It doesn’t exist. We have to make do with the best that we have, and in terms of the availability of the data we will be using Corsi to talk Possession. The reality is that we could be adding in zone time and zone entry information but those are either not available or require an incredible amount of work to record. Perhaps one day in the future we will be able to quickly break down a player/team in terms of how much time they spend in the offensive zone after a dump-in and how often that results in shots on net, but that information is not available without EXTREME time consumption.
That doesn’t mean that what we have isn’t useful either, it just means that we cant answer all of the questions we want to ask with it. Right now Corsi gives us a pretty good impression of where the play is taking place in a game. If you are interested in what direction we can go when we do have more information then check out what Tyler Dellow is doing with that data regarding Taylor Hall this year here.
For anyone still unfamiliar, Corsi For is the total number of shot attempts (shots, misses, blocked shots) that are directed at the opposition net. Corsi Against is the same thing but on your own net. This can be a team number or an individual number, it can be per X minutes of ice, and it can be a percentage. It is very much like the very flawed +/- stat except it records a lot more activities on the ice so the biggest flaw in the system is removed. When someone brings up Corsi, they are likely referring to 5v5 numbers, but they do exist for other game states.
WHERE ARE THE OILERS?
I hope you’re sitting down…They’re near the bottom of the League. Oh, you were kind of expecting that? Nevermind then.
Yes, the Oilers have been atrocious this year when it comes to the Corsi. As of today (per stats.hockeyanalysis.com) the Oilers sit 28th in the NHL in terms of Corsi% at 44.5%. That means that over the course of these 60 games, the Oilers’ share of shot attempts has only made up 44.5% of the recorded events. That’s really bad. The only teams with worse ratios are the Sabres and the Leafs.
If you’re thinking the same way that I did at first then what may have crossed your mind is that the Oilers play a wide open brand of hockey and they just need to tighten things up.
Not so much. They definitely need to tighten things up, but when I think of “Wide Open Hockey… I tend to think that means that should result in a large amount chances (read: shot attempts) for the team who is playing that style. In terms of total Corsi For on the year to date, the Oilers are 29th in the NHL with 2254. San Jose is leading the pack with 2942, if you thought maybe there wasn’t much separating the teams.
So much for that River Hockey theory. You want “Wide Open Hockey… with lots of chance going both ways? Start watching the Islanders, actually. Lots of chances for (2642), lots of chances against (2748).
The reality is that the Oilers are not spending a lot of time in the offensive zone creating chances, and worse, they are giving up a lot. In terms of raw Corsi Against, Edmonton is again in 28th spot in the NHL. This time, interestingly enough, the teams below them are the Canadiens and the Leafs.
WHAT IS GOING ON?
Specifically? I’m not sure. There are a lot of things going on with these Oilers and with this coaching staff. When Dallas Eakins first came in to Edmonton he was touting his “Swarm… Defense. He didn’t like some key ideas that the previous Coach had emphasized and really wanted to run this system instead. The results were catastrophic right out of the gate. The team lost 11 of 14 games in October and were dead in the water with more than 60 games remaining.
The “Swarm… came in like a lion and out like a lamb and by November the Rookie Coach abandoned his plan to adopt a more traditional defensive structure. You can read about it here on David Staples’ piece from the Journal’s Cult of Hockey. The Oilers were giving up some prime shooting opportunities as a result of their interpretation of the new system and pucks were going into the back of the net.
At the time (the end of October) Dallas Eakins admitted that he essentially gave up on the swarm so he could teach a more traditional Defense to his group because they clearly lacked the necessary training. And could you blame him at the time? The Oilers went into the season with a group that had made some progress the year before and a goalie that had posted a .920 sv% as well. All of a sudden Eakins institutes the swarm and the team goes down the toilet in a big way. It’s only natural to assume he was the problem.
SO WAS THE SWARM THE PROBLEM?
I don’t think so. The problem, as it turns out, was maybe more so the Goalie than anyone thought. Even after the switch back to “Traditional… Defense the goaltending was still abysmal and the goals kept going in against them. Is it not possible that the goaltending was SO BAD that it actually gave the impression the system was broken? The Oilers weren’t world beaters under the original system, but they weren’t terrible either.
Michael Parkatti has been tracking and posting Corsi numbers on his site BoysOnTheBus.com all season after every game, so if you are so inclined you can see the big breakdown for both the team and individual players on a per game basis. It’s truly interesting stuff. Because of his hard work we can see how our possession metric was doing over time rather easily.
We know that the Swarm was done by the end of October so we have a month’s worth of data that we can use for how the Oilers were doing with the Swarm, goaltender performance removed. Here are the averages for the Swarm and Post-Swarm Oilers.
SWARM CF 40.4 CA 45.8 C% 46.8 Differential -5.4
POST-SWARM CF 38.6 CA 48.6 C% 44.3 Differential -9.9
Right off the hop we can see that neither cluster did particularly well, both below the 50% mark. However, after Eakins abandoned his system the Oilers started spending less time in the opposition zone and more time in their own. Indeed the difference between what the Oilers were doing and what their opponents were doing almost doubled!
BUT…
Now the Post-Swarm cluster is significantly bigger than the Swarm cluster so we should really break things down further. Since the Swarm Oilers were the October Oilers it is natural to keep breaking things down by month.
OCTOBER CF 40.4 CA 45.8 C% 46.8 Differential -5.4
NOVEMBER CF 35.9 CA 43.8 C% 45 Differential -7.9
DECEMBER CF 42.0 CA 48.1 C% 46.6 Differential -6.1
JANUARY CF 38.5 CA 51.9 C% 42.6 Differential -13.4
FEBRUARY CF 36.3 CA 53.5 C% 40.4 Differential -17.3
Um…what happened after Christmas??? That is all kinds of bad. The Oilers really go right off the rails there in a big way. Looking at the monthly breakdown we can see that October remains the high point in terms of Corsi % but November and December aren’t bad. It’s not until January and February that this turns into a complete disaster.
In fact, if we take a look at the numbers in 20 game segments then the middle segment is actually stronger than the first.
GAMES 1-20 CF 38.3 CA 45.9 C% 45.5 Differential -7.6
GAMES 21-40 CF 41.6 CA 45.2 C% 47.9 Differential -3.6
GAMES 41-60 CF 37.4 CA 52.8 C% 41.5 Differential -15.3
Now this is really interesting. Looking at things this way really changes the way I view Eakins’ switch from the Swarm to something different. First and foremost, there is a huge chunk of games post-swarm where the Corsi metrics improve (games 21-40). 47.9% would still place the Oilers firmly in the bottom 3rd of the NHL but it was an improvement overall from where they started. Still we see that those final 20 games have been a complete gong show, though.
WHAT DOES THIS MEAN GOING FORWARD?
As we discussed previously, the Oilers have been gifted with good goaltending as of late. That goaltending has carried them through their worst stretch of hockey all year. The good news is that this team should be capable of turning that poor play around and at least return to previous levels.
The biggest problem that faces them possession wise will be figuring out what the heck happened over these past 20 games or so and correcting them. The good news is that Dallas Eakins has plenty of time to figure that out. The Olympic break is effectively 2 weeks long but the Oiler wont play for almost a week after that, so they have lots of practice time.
This practice time is going to be used like a mid-season Training Camp. Through Joanne Ireland, Dallas Eakins said, “I am comfortable with where we’re at system-wise now. We’re ready to take another step forward with some of the stuff we want to instill but we’re not going to take that step until after the break.…
I’m not so sure I would be all that comfortable right now, but we know that there are aspects of Eakins’ system that he has purposefully held back on, likely because of that horrific start to the year. The Coach clearly intends to bring more of that complexity into the way the Oilers play.
With a solid foundation in net teams can overcome being outshot and outshot-attempted every night but it is difficult. There’s a reason the best teams have solid Corsi numbers and it’s very basic. Good teams outshoot the other and spend more time in the offensive zone. As the Oilers transition into a team that drives the offensive play we will see their Corsi% increase.
If the Oilers want to build for next year then what happens in these final 20 games will be paramount. The club needs to pull out of a nose dive and excel at something they have failed to do all year. I’m comfortable believing they will get good enough goaltending to win games, but the skaters will have their work cut out for them.
One final thought. The Coach came in with a system in mind and felt compelled to change it almost immediately. In the 20 or so games that followed those changes the team did pretty well but something seems to have changed again and now the group is a mess. Now that the noise of poor goaltending isn’t an issue I think it would benefit the club if Eakins returned back to his original vision for the system.
Follow me on Twitter @Archaeologuy