In Leighton We Trust?

PHILADELPHIA - FEBRUARY 8:  Michael Leighton #49 of the Philadelphia Flyers looks on during a stoppage in play against the New Jersey Devils on February 8, 2010 at the Wachovia Center in Philadelphia, Pennsylvania.  (Photo by Len Redkoles/NHLI via Getty Images)

Despite all the rumours heading into the trade deadline, the Philadelphia Flyers did not acquire a goaltender.  That decision to stand pat surprised a lot of people (including me) given that the Flyers sit sixth in the East and have a strong enough roster to put together a playoff run.


With off-season acquisition Ray Emery on the shelf with season-ending surgery, the Flyers have an unlikely goaltending duo for a playoff team: Michael Leighton, a journeyman acquired off the waiver wire after a disastrous start with Carolina, and Brian Boucher, who started his career in Philadelphia but saw action in five other NHL cities before returning to the Flyers.


It takes some courage (recklessness?) on the part of Flyers G.M. Paul Holmgren to put his faith in Leighton, considering that his stats line in Carolina was an ugly 1-4-0 with a 4.29 GAA and .848 SV%.  He’s gone 13-3-1 since joining the Flyers, with a 2.33 GAA and .923 SV%.  How Leighton will perform the rest of the way and what to do with him has been a subject of much debate among Flyers fans, and one imagines inside the Flyers’ head office as well.


Is Leighton as good as he’s been lately?  No.  On the other hand, he’s probably not nearly as bad as he was in Carolina either.  Leighton’s approaching the 100-game mark in the NHL and he’s spent parts of seven seasons in the AHL, so we have some past history to reflect on.  Let’s start with his AHL track record, going back to 2004-05.


Season GP Record Save %
2004-05 41 20-16-3 0.921
2005-06 40 15-22-1 0.888
2006-07 21 10-6-3 0.920
2007-08 58 28-25-4 0.931


With the exception of an ugly 2005-06 with the Rochester Americans, Leighton’s been excellent at the AHL level.  In fact, he has been for his entire career; as a 20 year-old rookie with Norfolk, he managed a .920 save percentage in 52 games.  So that’s in his favour.  What about his NHL record?


Season Team GP Record Save % EV Save %
2002-09 Multiple 69 17-31-12 0.884 0.910
2009-10 Carolina 7 1-4-0 0.848 0.848
2009-10 Philadelphia 20 13-3-1 0.923 0.922


Leighton’s save percentage in all situations and at even-strength since joining the Flyers is superior to his prior career.  I’m not too worried about his even-strength save percentage, given that we don’t have a lengthy NHL track record to compare it to, and given that it’s only a modest jump on his career numbers.  I strongly suspect he’ll be able to maintain a number in the 0.915-0.920 range at even-strength.  However, I have no doubt that his overall save percentage is going to plummet.


The biggest reason is Leighton’s penalty kill save percentage of 0.927 – shockingly, a number better than his even-strength or his overall save percentage.  The NHL leader in short-handed save percentage (minimum 25 games played) is Ryan Miller, with a 0.918 save percentage; Leighton’s been better than him in that situation since joining the Flyers.


It won’t continue, for a variety of reasons.  Leighton’s never been a particularly good penalty-killing goalie in any season of his career, so there’s no reason to believe he has some kind of special ability here.  Also, while the Flyers goalies are all bunched together at even-strength (Emery has a 0.930 EVSV%, Boucher has a 0.923 EVSV% and Leighton is sitting at 0.921 EVSV%) both Emery and Boucher have seen their numbers drop because of lousy penalty-kill save percentages, so there’s no reason to believe that the Flyers are this magically good at penalty-killing.


There is, however, significant reason to believe that Leighton’s short-handed save percentage is almost entirely random.  Gabriel Desjardins posted a great article on that subject today which showed that a goalie was likely to regress 96.0% back to the mean: almost entirely, in other words.  His conclusion is as follows:


Even-Strength save percentage, which is measured over a large number of shots, is a significant predictor of future performance, though it must be heavily-regressed to the mean save percentage.  SH save percentage, on the other hand, is essentially random.  Your best guess of next year’s SH save percentage is the league average.  Indeed, last year’s even-strength save percentage is a much better predictor of next year’s SH save percentage than this year’s SH save percentage.


In other words, Leighton’s numbers on the penalty kill are going to drop off.  He’s performing at a super-human rate right now, and there’s no reason to expect it to continue and every reason to expect it not to.  He should still be okay; 0.922 is a respectable even-strength number, but expecting more than that from him would be a mistake.