Wrecked him? Damn near killed him.
I Want the Yak Attack Back
I am contemplating Nail Yakupov's future, and to do so, I have a roll of Tums next to me.
I am not unbiased in writing this post - I desperately want Yak to succeed. Aside from my 23 year old Oil Drop jersey, the only other Oilers paraphrenalia I own right now is an OilersNation "#YAKCITY" t-shirt.
I want Yak's love of the game to be a source of energy for his teammates instead of a source of derision from the Cherry's of the world. I want him to rub it in the face of the xenophobes. I want him to scare goalies ten times every game with that cannon of a shot. And I want him to help the Oilers hoist the Cup.
Imagine the celly, boys, imagine the celly.
But that future looks remote right now. Are the naysayers right? Is Yak a flop?
Jonathan Nails It
An interesting thing stood out for me the other day when I was reading this article by Jonathan Willis over at The Cult of Hockey.
Nice little article, and I particularly liked his analysis of how the change in year on year shooting and save percentage affected Yak's (horrendous) plus-minus:
Now, let’s apply Yakupov’s 2013-14 percentages to those shot totals. At a 7.5 shooting percentage, his goals for drop to 34, down 16 from the year before. With a 0.882 save percentage, the goals against figure rises to 77, a jump of 27. Without so much as a one-shot difference, that percentage fluctuation has turned ap layer [sic] who was even into one who now has a minus-43 rating.
But what actually stood out for me is the chart Jon posted in that article, showing Yak's 10-game rolling average of Corsi and Goals. Interesting shape to the Corsi curve: despite all the articles at that time about how "Yak was struggling", the Corsi was distinctly positive AND IMPROVING the first third of the season.
Positive. Improving. Not words you heard used in many sentences describing Yak this season.
But that chart shape was very distinct. So I dug into the data a little more (trust but verify!) to see what I could find. Sure enough - to me, it looks like a real thing.
If it Looks like a Duck ...
Yak's Corsi WAS positive, and it WAS improving through the first 30ish games of the season. That trend culminated with a tremendous three (out of five) games with 5x5 CF% of 83% against the Coyotes, 12 CF 100% (!!) CF% against the Flames, and 64% against the Big Bad Bruins.
Then something happened.
I'm wracking my brain, but I don't remember what that something is or was or could have been - this was weeks after the whole Larionov incident, the PP goals against, the MacT 'Yakupov is going to fail' headline.
But something happened, and whatever it was, it wasn't good.
Yak's Corsi trend turned seriously south the last two weeks of Dec, culminating in a 6 CA 0% (!!!) CF% against the Flyers on Dec 28th.
After that, he seemed to recover a bit, especially in early January, but that positive trend from the first few months was gone. You can see this in the chart below, where I've plotted Yak's 5x5 CF%, and thrown on some regression lines for the entire season along with the three visually separated seasonal segments (click for super large version).
So what happened during those last two weeks of 2013?
Maybe it's just a random thing, but those changes look too pronounced to me to think they're just random variation. Have a quick look at Hall, Perron, and even the ever-streaky Gagner's numbers, and you won't see that dramatic a Corsi range.
Was it the Western Conference effect? I don't think so. Yak's CorsiRel follows the same pattern as his Corsi - meaning that it's not just against the opposition that his play declined, it's relative to his own teammates, and that's a huge tell that something changed in his game. The games against the West also started a bit earlier than Yak's watershed.
Yak also posted radically different CF against the same teams from either conference mostly in line with this trend e.g. 64% Dec vs 21% Feb against the Bruins; 61% Nov vs 27% Jan against the Hawks. Against the Sharks even stranger - 48% in Nov, 55% in January right after his 0% Corsi game, and then 10.5% in late January.
Was it score effects? Maybe, but Yak's season average Corsi 5x5 (44.9%), 5x5 Close (44.2%), and 5x5 Tied (45.8%) are all remarkably close. If anything, the 'Tied' number suggests the opposite of score effects.
Not that you should read too much into Corsi in individual games. But there it is. Using the trends, to me it looks like a real thing.
So again I ask ... what happened during those two weeks in December?
What wrecked Our Yak?
More importantly, how do we unwreck him?
If I'm the Oilers, I'm phoning Yak right now and asking him to recount his year. Health. Medicine. Diet. Exercise regime. Video games. Girls.
And I'm watching every minute of Yak's games during the first third of the season, and then every second of every one his shifts in those mostly catastrophic next nine games. Because if they can figure it out, figure out what changed, figure out what happened, if they can get the dynamic young positive-Corsi-trend player from the first 31 games and put it together with a reasonable shooting % - watch out.
Figure it out, guys. The future of the team (and my t-shirt) may very well depend on it.