62933917

The Edmonton Oilers were almost assured of a spot outside of the playoffs prior to the puck being dropped for Game One in 2010-11.  A losing season is difficult for a lot of different reasons, and one of the difficult things about it is trying to judge the coaching staff: if draft picks and player development are the chief goal, then a losing season does not necessarily indicate a poor job coaching.

Still, that doesn’t prevent people  (including yours truly) from attempting to evaluate the coaches.  One of the most frequently asked questions is this: why has Tom Renney opted to sit the 24-year old first round pick with the 0.916 save percentage in favour of a 38-year old with a 0.891 save percentage on most nights?

The answer I’ve seen most commonly posited by those who follow the Oilers is simple: Tom Renney is trying to protect Devan Dubnyk.  The theory goes that Nikolai Khabibulin is what he is, and won’t suffer from getting shelled every night, while Devan Dubnyk might lose his confidence if he has to backstop a losing team.too often.

My belief is that if the coaching staff is making a concerted effort to shield Dubnyk in this manner, we should see a similar effort on their part to control who Dubnyk plays against.  If the limited playing time is a developmental choice meant to bolster Dubnyk’s confidence, it makes sense that the coaches would also try and play Dubnyk against beatable opponents.

There are two parts to checking the schedule: while Khabibulin has been healthy, and while he was injured.  We’ll start with the latter period, from November 18 to December 1.  The Oilers employed Dubnyk and 36-year old third-stringer Martin Gerber.  Dubnyk started five games over that stretch, and Gerber two.  We’ll look at the home/road split for each and their opponents’ points percentage as of the dates played:

  • Dubnyk: 2 home, 3 road, 0.595%
  • Gerber: 1 home, 1 road, 0.511%

Over this stretch, Dubnyk easily got the worst of the schedule, not only seeing more road than home games but also tougher teams.  The difference in the goaltender’s opponents winning percentages doesn’t seem like a lot until one projects it over an 82-game season: it’s a difference between an 84- and 98-point season.  Last year, that would have been the difference between 12th and 22nd in the league.

That may have been an aberration though: perhaps Tom Renney was still gunning for wins, and wanted to play Dubnyk against superior teams in the quest for points (though that wouldn’t explain why he started Gerber at all, let alone against the easiest opponents).  So let’s look at the difference between Khabibulin and Dubnyk during the time when the former was healthy:

  • Khabibulin: 17 home, 14 road, 0.588%
  • Dubnyk: 5 home, 7 road, 0.555%

This is a more interesting split.  Initially, we see that Khabibulin had a more favourable home/road split, but also played tougher teams than Dubnyk.  The difference in opponent strength is not extreme, however; over an 82-game season Dubnyk’s opponents would have an average of 91 points, while Khabibulin’s would have had an average of 96.  Last season, that would have been the difference between 12th and 14th in the league.

Meanwhile, the home road split makes quite a difference.  Last season in the NHL, home teams had a record of 690-389-151, while road teams went 540-540-150.  To put that in terms of points percentage, it’s the difference between 0.622 and 0.500 (or put another way, the difference between an 82- and 102-point team).  We can factor that into the calculation to give an adjusted strength of opponent:

  • Khabibulin: 0.582 expected opposition points percentage (96 point season)
  • Dubnyk: 0.566 expected opposition points percentage (93 point season)

In conclusion, Nikolai Khabibulin has faced very slightly tougher sledding than Devan Dubnyk, even when we allow for his greater percentage of home games, when both were healthy.  This may show a tendency on the part of the coaching staff to shield Dubnyk, but the difference is so minimal that I’d argue they either aren’t doing it intentionally or they’re very, very bad at it.

And frankly, that leaves me baffled as to what Renney & Co. are doing, starting the old goalie with vastly inferior numbers over the young goalie who looks like a star in limited minutes.  It doesn’t appear to make sense from a development angle.

Comments (3)

  1. I think your stats dont really prove anything except what you feel would be protecting Dubnyk. Renney is going on much different information such as how they are playing in practice, how the team as a whole seems to be playing, how he feels the opposition is playing, how the crowd is likely to affect the player. etc.

    Stats cant prove everything especially a persons intentions.

  2. VMR:

    No, but they do show that Dubnyk has played a schedule more or less as tough as Khabibulin. Hence my conclusion:

    “This may show a tendency on the part of the coaching staff to shield Dubnyk, but the difference is so minimal that I’d argue they either aren’t doing it intentionally or they’re very, very bad at it.”

    If they are trying to shelter Dubnyk, the evidence suggests they’re bad at it.

  3. Being a Rangers fan I can tell you that Tom Renney is actually awful at developing young players. In fact he doesn’t really have a plan for much of anything, other than his love of playing the fourth line. His best season came with what was actually a pretty veteran team. He tends to bench people at random, especially if they start scoring.

Leave a Reply

Your email address will not be published. Required fields are marked *