• Goalies And Horrible Four Game Stretches

    by Tyler Dellow • October 14, 2013 • Hockey • 11 Comments

    There are two schools of thought with respect to Devan Dubnyk at the moment. The first is pretty simple: Something Is Horribly Wrong. This school of thought has variants – some think Dubnyk has never been a good goalie (for our purposes, “good” means “good” in the sense of being a league average starter), some think he’s a good goalie who is broken at the moment but can be repaired, some think he was a a good goalie who was destroyed by the change in the equipment rules this year. The other school of thought is that this is just the sort of thing that all goalies go through from time to time and that it tends to resolve itself as mysteriously as it appeared – there’s probably not a lot of difference between adherents to this line of thought and adherents of the Something Is Horribly Wrong That Can Be Repaired line of thought, but the former tend to be a lot calmer while the latter are shouting a fair bit.

    If you’ve read very much that I’ve written about goalies over the years, you won’t be surprised to learn that I fall into the second school. I’ve talked about what I perceive as being the holes in the argument that Dubnyk’s not a good goalie before – he played for a poor team and put together three above average seasons in save percentage between 2010-11 and 2012-13. His save percentage for that period is towards the top of the mushy middle of all starting goaltenders. Dubnyk had a pretty solid pre-season too – if he was having some issue adjusting to the new equipment, wouldn’t it have shown up then? Yes, he’s facing a mix of AHL/NHL shooters but he really didn’t see to have to much in the way of difficulties in the pre-season.

    Do stretches like this happen for a lot of goalies? I wanted to test this out but it’s somewhat tricky to do. I ended up gathering goalies who met the following criteria: a) had faced at least 3000 shots since 2005 at the time of their struggles, b) saw at least 90 shots in a four game stretch and c) had a save percentage at least 80 points worse than their average since 2005 to that point during those four games.

    I then gathered four pieces of information about those players:

    a) what they did during the bad four game stretch;

    b) what they did in the next four games they appeared in;

    c) what they had done in the NHL from 2005 through their bad four games; and

    d) what their season looked like without the bad four games.

    All of that information is in this table (larger form here):

    Screen Shot 2013-10-14 at 5.28.54 PM

    The first thing that jumps out at me is the average age of the guys who have a stretch at least 80 points below their established norm. It’s an awfully old crew, with 21/27 of them being 30 or older. A lot of these guys were old goalies who just couldn’t play anymore who were getting by on their name.

    The average save percentage of these fellows over the bad four games is a hyper-abysmal 0.819. Dubnyk laughs at these fellows – his save percentage is at a relatively lofty .829 so far. But then look at what happens in the next four games – these guys who had been getting shelled are suddenly just fine.

    As you can see if you slide your eyes over to the next column, which is their save percentage up to and including the bad stretch, they’re actually better than they have been historically. Now, I suspect that this is a randomness thing and I doubt that a horrible stretch actually predicts a better than average stretch but you can’t really say that a horrible stretch predicts another horrible stretch.

    The last column is the most interesting. In that column, I’ve calculated the save percentage for each goalie in the season with the bad four game stretch. It’s basically essentially the same as their collective save percentage leading up to that horrible four game stretch. Are there some guys who were substantially worse? Yes, although many of them were a) old goalies and b) at the end of their careers. Everyone 30 or younger ended up with numbers for the season that ended up in the same ballpark as their career numbers.

    There’s a tendency, when a goalie is struggling, to assume that he must be doing something wrong. The Goalie Guild has made a career of proclaiming this (and then, if necessary, deleting the instances where he was wrong). I watched the tying goal against Toronto again (and again and again and again) and the commentary was kind of funny. Craig Simpson mentioned that Dubnyk was deep in his net. Of course he was – the Leafs had a guy in the slot and if Dubnyk had challenged Lupul aggressively, Lupul would have thrown the puck over to him and if he’d scored, the line would have been “Dubnyk’s being too aggressive because he’s struggling.” When a puck goes in, there’s usually a way to blame the goalie.

    If you go back and look, actually look, at the goalies who’ve suffered a stretch like Dubnyk, it sure looks like it’s just one of those things that happens sometimes. Technical fixes, mental strain, whatever it is, it just seems to fix itself for these goalies as a group. If that’s the case, then it doesn’t make a lot of sense to be that worried about Dubnyk. Lord knows I wanted a hot start for these Oilers and I desperately wanted them beat the Maple Leafs badly but looking at the actual data, I can’t find any reason to think, based on four games, that we should expect Dubnyk to be anything but a .913 goalie the rest of the way (and I’d bet on him beating that). Hopefully there’s no panic trade or signing and in a few weeks, this is all a footnote.

    Email Tyler Dellow at mc79hockey@gmail.com

    About Tyler Dellow

    11 Responses to Goalies And Horrible Four Game Stretches

    1. October 14, 2013 at

      Spent a lot of the summer poring over goalie data. The short version: all NHL goalies have a lot of bad games. It sucks, but it’s nothing to overreact to.

    2. Fractastic
      October 14, 2013 at

      I hope you are right and Dubnyk turns this thing around, or Labarbra runs with the ball. I just want Edm to win right now.

    3. Lewis Grant
      October 14, 2013 at

      Agreed.

      The collective panic over Oilers goaltending ignores the fact that Dubnyk has demonstrated that he’s a reasonable goaltender (as your posts have shown). While it’s possible that there are equipment and/or confidence issues, those things tend to right themselves after awhile.

      Hopefully MacT is less jumpy than the fanbase. I’m not sure he is. Especially with the Caps game disaster in the making.

    4. jdude
      October 14, 2013 at

      trade him

    5. Dan
      October 14, 2013 at

      His numbers and stats say that he is a good goalie.. But my concern is that he seems to let in key goals, he never seems to make the big saves when we need him to.. The players also do not have the confidence in him like they should.. Winning teams have goalies that stand on their heads when the game is on the line, I have never seen that outstanding stand on your head performance from Dubnyk. The last time Oilers have had that was Cujo, some could say Roli from time to time..

    6. Bank Shot
      October 15, 2013 at

      A) You keep mentioning that Dubnyk played for a poor team so it was good news that his save percentage is as good as it was. But I believe I have seen you argue in other instances that “a shot is a shot” and bad teams don’t give up inherently better scoring chances then good teams, just more of them. If you admit that play style has an effect on save percentage, then it makes the comparison of goaltenders save percentage league wide basically pointless.

      B) What do you make of the fact that this has only happened to 5 goalies under the age of 30. And 4 of those goalies are not starters? Does this say something about the company Dubnyk is in?

      Why hasn’t it happened to guys like Price, Fleury, or Ward who have played perhaps double the games of most of these goalies in the same time frame?

      Mostly every goalie that has been in the league for a substantial amount of time has an off year. Dubnyk hasn’t. If he has his then he could drop to .906% or a .908% goaltender. Would you be looking at him the same way if he didn’t hit the percentage lottery last season posting a .920? That might be the aberration thus far.

      • Tyler Dellow
        October 15, 2013 at

        a) I tend to think that a really good or really bad defensive team can change things by + or – 5 or 10 points of save percentage. I don’t think it’s enough to make a bad goalie look like a star or anything but enough nudge things up or down a little bit. In that context, I think it’s worth noting that Dubnyk was on a poor team – he’s certainly not been a system goalie.

        b) Price has been close a couple of times – he just missed some of the cutoffs. He let in 15 goals on 80 shots over 4 games at one point last year (.813) which would have made the list but for him getting pulled quickly and MTL not giving up a ton of shots.

        Ward’s got a ton of these, with difficulties in catching the cutoffs. 2011-12, for example, he had a 17 G in 104 shots sequence; his career save percentage is lower than DD’s by virtue of having played in the NHL’s livest puck era immediately post lockout II, which makes it harder for him to hit hte cutoffs.

        Fleury’s come close a few times too. He’s been at four games 75 points below his career norm; again, he benefits a bit by getting bombed during the lively puck era, just because of my methodology. These guys all have their brutal stretches though.

        I don’t think Dubnyk’s a .920 true talent guy. My estimate is more like .915 or so. That said, that’s nice and league average for a starter.

    7. DD
      October 15, 2013 at

      Was there any point of throwing a jab at Goalie Guild? I think your article has enough substance without it.

    8. Jon
      October 15, 2013 at

      I think edmonton’s problem would be that league average goaltending will not get them into the playoffs.

    9. Pingback: BTO – Bachman’s Time Overdue? - The Oilers Rig

    10. Pingback: Islanders News: Matt Donovan, Josh Bailey Rebound; Offense from the D; Boss and Trots | All New York Islanders

    Leave a Reply

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