Did The Oilers Have Bad Injury Luck Last Year?

Injuries at the NHL level can have tremendous impact on the team. In 2009-10, the Oilers’ fall to last place in the NHL was blamed in part on injuries to Ales Hemsky and Nikolai Khabibulin. In the comments section here, various parties pointed to those injuries and asked how a team like the Penguins would do without Sidney Crosby and Marc-Andre Fleury (one hears that less often these days, for multiple obvious reasons).

Were the Oilers been particularly unlucky when it came to injuries in 2010-11?

At first glance, the answer would appear to be ‘yes.’ According to the team’s official website, the Oilers lost 281 man-games to injury last season, a number that sounds staggering and equates to three or four players missing each and every game of the season.

Of course, to determine whether that number is actually a lot, one needs a frame of reference. $100 is ‘a lot’ if we look at the prices of chocolate bars, but considerably less if we’re out to buy a house. How can we get that frame of reference?

The method typically used is by comparison to other NHL teams. The problem with this method is that different players get injured more or less frequently – if Andrew Cogliano misses five games in a year, he had an abnormally injury-filled season, but if Ales Hemsky misses five games in a year it was a great season. Because of this, various teams employing different players should expect a differing number of man-games lost.

A better method, the method I’m going to use, is to compare a player’s total number of games missed to their three-year averages. Thus, if a player misses 20 games after averaging perfectly healthy seasons the last three years, then that would be 20 unexpected missed games. If, on the other hand, a player appears in all 82 games after averaging 10 games a season lost to injury, then the team could count itself lucky to get those ‘extra’ 10 games. Naturally, this isn’t perfect – a normally healthy player might suffer a serious injury and thus have his average skewed upward – but random injuries happen too and should be accounted for to some extent.

How do the Oilers fare by this method? We’re going to consider them by position, but before we do the number supplied by the Oilers’ website needs to be pared down a bit.

For starters, the Oilers’ itemized list of injuries shows, not 281 missed games as the top of the page states, but rather 278. Since I need the itemized list for this analysis, and that’s the number with supporting data, that’s the figure I’ll use. Additionally, I’m going to remove the 18 games lost by Taylor Chorney, as I’m only using players that started the season on the roster (in other words, we aren’t using Chorney’s 18 injured games, but we aren’t using Ryan O’Marra’s zero missed games either).

Next, there is the matter of the rookies. We have no way of knowing what an average season looks like for Linus Omark or Taylor Hall, so we can’t run this analysis on them. Those players are listed below.

Rookies

Player Games 3-Yr. Avg.
Taylor Hall 15
Jordan Eberle 13
Magnus Paajarvi 1
Linus Omark 0

Forwards

Player Games 3-Yr. Avg.
Gilbert Brule 38 6
Shawn Horcoff 33 12
Ales Hemsky 32 26
Sam Gagner 12 7
Jean-Francois Jacques 12 20
Colin Fraser 7 0
Steve MacIntyre 3 14
Andrew Cogliano 0 0
Ryan Jones 0 10
Liam Reddox 0 1
Zack Stortini 0 2
Total 137 98

Based on three-year averages, we would have expected the Oilers to lose a little fewer than 100 games; instead they lost 137. The main culprits for the deviation were Gilbert Brule and Shawn Horcoff. Neither is a terribly surprising addition to the list, but both had been relatively healthy over the previous three seasons.

Defense

Player Games 3-Yr. Avg.
Ryan Whitney 45 14
Jim Vandermeer 15 13
Theo Peckham 11 10
Kurtis Foster 5 26
Ladislav Smid 4 17
Tom Gilbert 3 0
Jason Strudwick 0 1
Total 83 81

On the blue line, the Oilers missed almost the exact number of games we would expect, although the distribution was a little different from three-year averages. Rather than losing Foster and Smid for extended stints, the Oilers lost Ryan Whitney for more than half the year.

Goaltenders

Player Games 3-Yr. Avg.
Nikolai Khabibulin 11 32
Devan Dubnyk 0 0
Total 11 32

In net, the Oilers had a fortunate year. Devan Dubnyk was unsurprisingly healthy, but Nikolai Khabibulin only missed about a third as many games as he’s averaged over the last three seasons.

Summary

Based on three year averages, we would have expected the Oilers’ veteran roster players to miss 211 games. They missed a total of 231, indicating a slightly higher than expected total but nothing extremely surprising.

Even the distribution was relatively even. The losses of Horcoff and Whitney hurt, but the players who experienced exceptionally healthy seasons – Smid, Foster and the team’s starting goaltender – help to balance that out.

I think we can say that 2010-11 was a below-average season for the Oilers on the injury front, but not by a lot.

    • justDOit

      Interesting article but I am not sure much this really tells us. I have always found the total man games lost to injury a silly thing to reference.

      Clearly having Steve MacIntyre more healthy then expected does not offset losing Hemsky.

      What I think would be more interesting is doing a study comparing the quality of man games lost due to injury. Since losing Hemsky has a far greater impact to the team than losing MacIntyre I think you need to weight it. You could use advanced stats of some sort to come up with a weight.

  • book¡e

    JW – I seem to remember the Oilers having a greater than average injury rate in the previous season as well, thus the 3 year running average might not be that useful. It is possible to have 2 bad years in a row.

    Also – does your three year average include 2010-2011 or is it the previous 3 seasons. With such a small set (3 seasons), including the potentially ‘abnormal’ season in the comparison stats would greatly mute the comparison.

  • O.C.

    ====

    Were the Oilers been particularly unlucky when it came to injuries in 2010-11?

    ====

    HAVE the Oilers been particularly unlucky when it came to injuries in 2010-11?

  • Quicksilver ballet

    JW-I appreciate your attempt to put a more specific value on the number of man games lost, but I think that your formula leaves out the value of each individual game lost. If you’re trying to determine if the Oilers were ‘unlucky’ or possibly have deteriorating ice, or a less competant training staff etc, then fine. However, I would think that to truly determine unlucky, you have to correlate with the quality of players missing those games. Using the Pens as an example: even if the entire rest of the team stayed healthy last season, missing just Crosby and Malkin has an enormous impact on the team that is likely far greater than the rest of the team missing, say, a combined 250 games. Even if the rest of the team stayed 100% healthy, I would say that the Pens were very unlucky last season.

    Turning to the OIl: Hemsky, Horcoff, Whitney, Eberle, Hall and Gagner’s missed games hold far more value than that of JFJ, SMac, Jones, Foster, Smid etc.

    When several of the players who saw an increase in lost time were your top point per game players and those who saw a decrease in lost time were your pluggers, I think that you have to account for that. Yes, maybe the Oilers were in the same ballpark for their usual man games lost, but to determine what the impact of those man games, what if you took each of those players (minus goalies) and correlated their games lost total with their PPG. This should give you a more accurate value of the quality of the the games lost.

    Finally, as book!e pointed out, haven’t the OIl had an exceptionally high number of man games lost for the last 3 years, thus skewing the team’s numbers?

  • Everything will be better this year. Ryan Smyth is back.

    But to those who don’t think using the last 3 years is fair because we’ve had so many injuries over that time, at what point is bad luck with injuries actually the result of expecting too much from injury prone players?

    I am a huge Hemsky fan, but if he misses 20 games next year I shouldn’t be shocked and call it bad luck. Same goes for Horc. These guys are repeating a pattern of injury.

  • book¡e

    I think it is useful to note that JW is checking to see if the Oilers have been unlucky and thus he is trying to control for the possibility that the Oilers are simply injury prone. So, it is possible to have way more man games lost to injury, but not be unlucky. If your best dman is nicknamed Broken Ankle Whitney and your best forward is Glass Shoulder Ales, then you cannot blame ‘luck’ for your situation.

    With that said – the 3-year baseline is pretty short and two years of ‘bad luck’ could actually occur.

  • Particularly unlucky?

    Yes. Yes they were — not in number of games lost to injury but in who got hurt and when (which a math spin doesn’t speak to). That’s obvious. Good grief.

    Edmonton’s best offensive defenseman plays 35 games, leaving the overwhelmed Tom Gilbert as the next-best option in terms of generating offence. Whitney outscores Gilbert by a point despite playing 44 fewer games.

    Not one of the players who finished among the team’s top four in scoring — Eberle, Hemsky, Gagner and Hall– played even 70 games. You can’t take that much production out of the line-up of an offensively challenged and defensively suspect team and expect success.

    Strip a rookie-laden roster of the experience (and production) of Hemsky, Horcoff and Whitney, not one of whom played 50 games, and what do you get?

    Who played something approaching a full season, say 75 or more games? Paajarvi, Gilbert, Jones and Smid. That’s it. That’s all. Problem? Yes. Yes they were particularly unlucky.

    • book¡e

      Well, there is an aguement out there that Whitney, Hemsky, and Horcoff being injured has little to do with LUCK and instead should be seen as an expected result. Bad luck would be if they were injured more than average for themselves (as opposed to vs other NHL players who may be more durable).

      • OB1 Team Yakopov - F.S.T.N.F

        It’s hard to know for the average player how much is really “bad luck”, and how much is “injury prone”.

        For example, Gagner has missed 14 games each of the last 2 seasons, but I’m not really sure that’s the same thing as Khabibulin.

        But I would guess that it’s reasonable to project Bouwmeester (0 games missed in the last 4 seasons) to play more games than Whitney (89 games missed), or Smid (74 missed) next season. Doesn’t mean it will happen, but probably a reasonable guess.

    • Self preservation, while maintaining productiveness, is a skill most great players have, or learn, quickly.

      Lots of skilled players lose great careers to bad luck injuries. Lots also lose great careers to that simple little thing of not being able to avoid the big hits.

      It’s too late for Hemsky likely, but I really hope Hall and Eberle learn it. Nobody has a tough brain, and Hall’s balls out thing will get him hurt bad sooner or later. He has to be smarter and I am sure he can learn it.

      Eberle gets nailed trying to beat guys board side and gets hit hard, too much. He has to slip off those or go inside.

      If the Oilers didn’t carry the chronically injured guys the numbers wouldn’t be too bad. I don’t think many teams that have signed injury prone players have won that bet much lately.

  • O.C.

    I don’t buy this methodology. This doesn’t tell us if we had an injury affected results. This tells us the Oil have a bunch of people that have been hurt a lot recently. I’m pretty sure we already knew that. In fact the data seems so obvious the conclusion seems goal-seeked.

  • If the only purpose of the article is to dispute the use of the word “luck” in any sentence involving the words Oilers and injuries, then one would expect a league wide study over a number of years, say back to the latest major equipment improvement. And then compare those results to the ones the Oil experienced. Are injured players always injured or for the most part is this more like shooting percentage?

    And as Brownlee rightly points out timing and luck go hand-in-hand, so that has to be considered.

    Is Roli’s GM1 SCF injury the same “luck” as the injury to Chorney last year?