Wednesday, February 18, 2009


The bounces can be cruel

Kovalev led the club in scoring with 84 points last season, but he is tied for second with only 13 goals and 26 assists in 57 games in the current campaign.

He has also looked listless on the ice, rarely showing the fire that made him the team's most popular player with the fans last season.

"The team doesn't need him the way he's playing now," said Gainey.

"To score goals and help other people score, you have to do a lot of different things. To stay on task, to do the mundane, small, detail, mandatory things to keep the team afloat is a harder thing for some players than for others."

To call this outcome predictable would be overstating it, but it was absolutely inevitable that all things Kovy would not be as sunny and joyful this year. Alert readers may recall that in my September post on scoring expressed as rates (per 60 min TOI) and game states (EV, PP, PK), I noted that Kovalev produced at an inhuman rate on the PP last season, to the tune of 8.07PPPts/60 when no other player was above 6.00PPPts/60.

To recycle an old analogy, this is kind of like batting .400 in mid-June. It's impressive as hell and doesn't happen unless you're extremely talented, but it's almost certainly not sustainable -- you're bound to go through a stretch when even your hard line drives are right at someone's glove.

Anyway, at right are some numbers for last year's hard working, leading-by-example Kovalev compared with this year's fire-less, non-detail-oriented Kovalev.

As you will notice, these particular numbers are awfully similar. Virtually identical point production at even strength. Slightly more frequent EV shots on goal this season, leading in turn to a slightly better shots differential and Corsi number. Pretty similar SV% behind him while he's on the ice. The EV Shooting% while he's on has actually dropped from 10.0% last season to 7.3% this season, but the additional Shots For (and earning points on a higher % of scoring plays while on) have kept his production level. And maybe most surprising, slightly more frequent PP shots on goal while Kovalev is on the ice.

Now here, at right, is the objective (or if you prefer, numerical) problem. For whatever reason, the shots on the powerplay are simply not going in (the Habs as a whole have dropped from 16.5% on the PP last season to 9.7% this season).

Bob Gainey has forgotten more than I've ever known about hockey, and has an extremely good record at every job he's ever had in the game. But I think this is nevertheless worth highlighting: just how strongly results colour our perceptions of the "little things".

Alexei Kovalev may well be unwilling to get his nose dirty and have trouble staying "on task", etc. -- but if that's the case now, it was probably also the case last season. Whatever his struggles are, they have had either a negligible or zero impact (relative to last season) on his 2-way EV play; his ability to create shots at EV and on the PP; and his ability to score at EV. They have only manifested themselves numerically in his (and his team's) ability to finish while on the PP. Which -- clearly -- seems like a mighty odd thing to see as a sole consequence of lackluster play.

If I were to look at these figures without having watched the Habs play at all this season (which is very nearly the case), I'd have to conclude that the biggest difference in #27 was that he got a lot of love from the hockey gods on the PP last season, and now he's getting quite a bit of hate.

But whatever. I don't cheer for the Habs, so I don't care if they bench a player whose primary problem seems to be Bad Luck, Compounded by Prior Good Luck.


The power play is really letting Montreal down this season, evidenced by the drop in PP % you mentioned, while there is virtually no change in their production at even strength. As a team they're on pace for 145 goals at 5 on 5 this season compared to 148 last season. The only big change in personnel on the power play is Streit and I certainly didn’t his loss would have this big of an impact.

It's interesting, what they do with Kovalev. All those numbers you put up, Montréal's management are aware of. The way they kept using Kovalev show this.

And that's why I think the story is getting overblown. He is a valuable (if non-exceptional) contributor.

He did get into a kind of funk since the ASB though (his CORSI was actually positive before). That's why he is now "benched"; dude is in a funk. But I doubt he has more value on the trade market than on the team. And I think Gainey knows that.

They do have Tanguay coming back any day now.

I think the interesting thing is that the drop in PP% is entirely a shooting percentage thing, not a fewer shots (e.g. failure to get set up in the zone) thing.

Some of that might be fewer plays where an A+ chance is created, but a lot of it almost has to be bad luck. That and the roster changes, i.e. losing Streit.

I dunno, it may not be manifesting itself in the numbers, but in the games I've seen, Kovalev's just looked horrible. Maybe he's playing hurt, maybe his decision-making is artificially made to look worse because he's the only guy to key on for that PP now, or maybe he just isn't playing as hard as he did last year for any of 187 reasons, but there's something visible, and it was visible long before the storyline was presented in the media.

This much I know: that power-play is shit, regardless of what Kovalev looks like right now. Fingers crossed that Schneider makes something happen there, because without that PP in order, their EV and PK improvements are probably going to go for naught, especially with Carey Price playing like latter-days Jose Theodore right now.

Kovalev is a world-class dog-fucker.

Maybe that fact was obscured by his PP success last year, leading some Habs fans to overlook what has been blindingly obvious to fans in every other city Kovalev has played in.

Doesn't anybody remember how desperate the Habs were to motivate/kill/trade Kovalev two season ago?

He merely reverted to form this season. That's about as surprising as Peter Forsberg being hurt or Andrew Raycroft being a bad goalie.

Alex Kovalev is probably the player in the league who most clearly demonstrates the way that fans project their preconceptions of what a hockey player should be onto their assessments of the player. For some people, Kovalev’s insane stick-handling and jaw-dropping shot makes him L’Artiste, “the most talented player in the past 15 years.” But to others, the fact that he doesn’t score 100 points a year, despite those hands means he’s lazy, only cares about himself, doesn’t try, etc…

It’s pretty obvious to me, though, that the reason Kovalev hasn’t ever scored 100 points is not because he doesn’t try, or because he’s “mercurial” or “enigmatic.” It’s because he isn’t actually good enough to be a 100-point player. Kovalev has amazing hands, but he’s really, really slow. He can’t create space for himself. He needs his teammates to gain the zone and get him the puck. And he’s not all that solid defensively. None of that has to do with effort: he’d still be slow if he took pucks in the face every night, dropped the gloves, and had a nice Canadian name. Having great hands is nice, but it’s not the only skill that a hockey player needs: people see his hands and either think he’s the best player in the league or a lazy, selfish malingerer. He’s neither. He’s not a dog-fucker: he’s an incomplete hockey player who’s a bit too slow to be elite.

As far as perceptions go, Kovalev is the anti-Ryan Smyth. To a certain sort of hockey fan, Ryan Smyth (or rather, the Ryan Smyth of a few years ago) is the perfect hockey player. He scores ugly goals! He takes pucks in the face and gets back up! He’s played for Canada a million times! He’s missing teeth! Overemphasizing this kind of thing got Ryan Malone an insane payday last summer. The fact that Smyth does these things is nice, but it doesn’t mean he’s the best player in the league. (Steve Begin takes pucks in the face too. He’s a healthy scratch for a Habs team that’s lost 10 of 13.)

Matt D for prez.

So here's the question: Kovalev is clearly undervalued in Montreal (as in, he's a starter, not a bench player). Does moving him to Edmonton help the team, assuming Jagr isn't coming?

Excellent post. Do you feel this same way about Iginla? (Due to lower percentages his goal production is down compared to last season despite a higher corsi and almost identical shot amounts and distance.)

I'm totally with you on the good luck turning to bad luck, but can it really make this dramatic of a difference? Kovalev's going to drop by about 30 power play points alone this season at this rate.

Just watching the Habs power play, they're not getting primo chances anymore. And Streit has played damn well for the Isles when up a man (which doesn't happen often).

As for Kovalev’s PP shooting numbers, my sense is that he’s taking much lower quality shots. Last year, Markov was really controlled the PP from the left point. Teams were as concerned with covering Streit’s point shot as they were with covering Kovy, which opened the PK box up for lots of cross-box, backdoor passes from Markov to Kovalev at the right faceoff circle. Lots of his goals came from one-timers and cross-crease passes. And when Kovalev got the puck, he had space to create a shot for himself, since penalty-killers were worried about covering Streit.

This year, teams aren’t worried about the point shot, and they’re collapsing down around Kovalev. I can’t remember seeing him get a cross-box pass from Markov once all year. But the PP is running through Kovalev more than Markov, so he’s taking just as many shots. It’s just that those shots come with two guys in his face. They’re not one timers, and he never gets time and space to walk in off the boards and pick his spots.

So my impressionistic conclusion is that his SH% numbers reflect the fact that he’s taking worse shots, and not all of that is his fault. (Of course, he could learn how to pass the puck rather than hold it looking for a shot that’s never going to come, but without a real point shot threat, he may figure there’s not much point.) Now, that conclusion could be a combination of confirmation bias and changing luck, I don’t know. But the Habs PP sure isn’t mesmerizing anyone with puck movement this year.

@ Sunny: short answer, No. The Corsi figure might be the same, but last season he was 3rd/9 among Flames forwards (min. 10:00 ESTOI/gm). This year he's 8th/9. Roughly speaking, last year he spent more time on the ice with poor teammates. Also, he's not taking as many shots; he's down .6 shots per game (15%), which I think (at this point in the season anyway) qualifies as non-trivial.

I'm totally with you on the good luck turning to bad luck, but can it really make this dramatic of a difference? Kovalev's going to drop by about 30 power play points alone this season at this rate.

This big a difference? Probably not, but I'll repeat that half of the difference (maybe more?) is that he and the Habs saw everything go in last year.

Generally speaking, my view is that there are a lot of plays every game that look like pretty mediocre (or poor) scoring chances when they hit the goalie in the crest or miss the net, but would (or do) look like a well-deserved goal when they go in off the inside of the post.

The point that I really wanted to underline with this post is not *Kovalev is still teh awesome!!1!1!*, but that unsustainably good and bad results both have a disproportionate effect on our perceptions of how well a player is playing. Kovalev's PP results last year were so good that (A) the bar was set impossibly high for this year, especially with the loss of Streit, and (B) they obscured the fact that, as has been the case for most of his career, he's a pretty damn average player at evens.

You're looking at the whole season's numbers, so the "listless" Kovy isn't an accurate picture. I never, and I don't think Gainey or Carbo did either, think that Kovy was listless for the first half of the year, though he clearly did not have chemistry with his linemates the previous year as well. It's the team's recent slide that has coincided with Kovy dropping into some kind of existential funk. His effectiveness is waning when the team needs him the most.

That's the problem, and it won't show up in season long stats, aside from the PP difference. Yesterday, against Washington, the PP went 3 for 4 with Schneider in and Kovy out. Go figure.

I also think Montreal's GF/GA ratio was much improved pre-slide even strength compared to last year, but you're concentrating on just GF... the 5 on 5 play for Montreal is much improved, just maybe not in GF. The slide is not the fault of a bad PP... though it is not helped by it, either. The bad PP was the same when the team was in 1st at the start of the year and essentially 2nd for most of the year (4th in standings).

Excellent post Matt. Besides your stats his whole handling has been bizarre to say the least. Kovalev seems to be the new whipping boy in Montreal, the 'Avery' of the East one might say.

Great post, and great comments by Matt D.

Markov having the two great options on the right side (Souray/Streit and Kovy) had a lot to do with the PP success, but this much of a drop off has to be mostly luck.

Unfortunately (for Habs & Habs fans) that's not the way Carbo sees it so he literally sends the 4th line out for PPs, presumably to "stir things up." What a mess.

re: ryan smyth
Bob Gainey was once called the best all around hockey player in the world by the head of the russian hockey federation.That,s saying alot considering all the players who were playing at the time.

i'd like to 2nd (or 4th or 5th or whatever) all the praise here- fantastic article.

if i may, though, i'd like to put forth a methodological question: how did you decide where 'luck' begins- that is, where not only your ability but anyone's ability to hypothesize causal mechanisms is exhausted? when i got to the juicy twist in your article, the first thing that jumped to my mind (having seen deplorably few habs games completely this year myself) was that kovy might not getting the same quality of shots with the current power play as he did in the streit season (the same thing matt d said). but that's the point where you decided to fall back on luck (i.e. inexplicability) as the necessary conclusion. why did you choose to end your analysis when you did? and, more broadly speaking, how do you decide where the borders of luck are when you do this kind of quantitative analysis? you personally, or you speaking on behalf of any organization you wish to speak for.

[apologies if this sounds at all bitchy- it's nothing but pure and docile curiosity.)

Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?