Drance Numbers: What do we know about NHL goalies after fifty games?

Drance Numbers is the silly research wing of PITB. While Messrs. Wagner and Mooney blog nationally and solve mysteries, Drance Numbers will look into the minutiae of quantifiable NHL data and bore you with it every Friday. Today, Drance looks at what Cory Schneider’s first 50 games tell us about his future in the NHL.


Vancouver’s hockey fans are smitten with their young, red-headed backup goaltender, and why not? Cory Schneider seems like an excellent teammate, he air-guitars like a pro, and can seemingly bend space and time when making desperation saves.

While I’ve never played goalie, and don’t know a lot about the position from a technical perspective, folks I trust rave about Schneider’s flawless technique and consummate ability. When I spoke with Justin Goldman (The Goalie Guild) over the summer about Cory Schneider, for example, he unequivocally raved about the guy, saying “Cory is one of the most well-rounded young goaltenders in the NHL. Everything about his size, speed, angles, positioning, reactions, his visual attachment to the puck, I could go on and on listing all the elements that he’s good at.”

So could many Canucks fans (although perhaps not as intelligently).

Now I don’t want to re-hash the tit-for-tat of the recent, somewhat overblown “goaltending controversy” in this space, but the major reason the situation in net has taken on a life of its own isn’t based on anything subjective (Canucks fans just don’t like Luongo!), nor is it the result of a cacophonous sports news cycle in the city. Rather, it’s based on fact and results: Cory Schneider has been the better goaltender, by a country-mile, so far this season.

After last night’s 4-3 comeback win, Roberto Luongo is sporting an .897 sv% after 17 starts, having faced just over 400 shots. Cory Schneider, on the other hand, has stopped .934% of  the 391 shots he’s faced in 15 starts. A situation in which the ostensible “backup” outplays the starter to this extent, even over a relatively small sample, is bound to generate over-wrought sports radio chatter and polarize a fanbase.

What I’m interested in, however, is learning more about what Schneider’s hot-start to his career means for his long-term development and performance. To these eyes, he looks like he may be the next great goaltender. But that’s been said over the years about everyone from Andrew Raycroft and Steve Mason to Jeff Deslaurier. The very nature of goaltending results in inconsistent performance, and it’s rare for goaltenders to maintain a high save-percentage season after season.

Looking at the numbers, is there anything we can glean that might give us some insight into just how good this kid might be?

To begin with, what I’ve done is compare Schneider’s “first 50 games” to the first 50 games played by 11 other “high-end” rookie goaltenders since hockey resumed post-lockout.

By the numbers, Cory Schneider has had the second-best start to his NHL career of any goalie that has come into the league since the 2005-06 season (not counting Ryan Miller, who played 20 games before the 05-06 NHL season):

Goaltender Total SA Sv% QS% BU%
Tuuka Rask 1498 0.931 72% 6%
Cory Schneider* 1267 0.927 65% 14%
Henrik Lundqvist 1429 0.924 70% 6%
Jimmy Howard 1533 0.924 64% 12%
Pekka Rinne 1453 0.923 66% 12%
Corey Crawford 1395 0.918 62% 14%
Steve Mason 1371 0.918 62% 16%
Jaroslav Halak 1580 0.915 50% 12%
Mike Smith 1229 0.910 54% 22%
Jonathan Quick 1410 0.910 50% 16%
Ilya Bryzgalov 1298 0.909 56% 18%
Antii Niemi 1250 0.903 48% 22%

(*) Cory Schneider has earned 42 decisions in his career and played over 35 minutes in a regular season game 43 times, his sample is slightly smaller than the other goalies in this table.

It’s promising stuff, and probably not altogether that surprising. What I’m really interested in though, is projecting future outcomes. We already know how these guys played during their first 50. What comes next?

Using the same sample of post-lockout goaltenders listed above, I took a look at their performance over the “next 50 games” of their respective careers. This is how those games played out:

Goaltender Total SA SV% QS% BU%
Ilya Bryzgalov 1505 0.926 66% 8%
Antii Niemi 1444 0.926 70% 8%
Jaroslav Halak 1535 0.923 62% 12%
Tuuka Rask* 894 0.916 58% 15%
Pekka Rinne 1365 0.911 50% 12%
Jonathan Quick 1354 0.911 52% 18%
Jimmy Howard 1407 0.910 54% 16%
Mike Smith 1551 0.910 56% 16%
Henrik Lundqvist 1372 0.907 58% 24%
Corey Crawford* 944 0.904 43% 8%
Steve Mason 1373 0.896 48% 28%

*Crawford’s sample includes only 35 games, and Rask’s 32.

That changes between the two tables are pretty interesting. The first thing I notice is, of course, that the bottom of the “first fifty games” table has flipped, and Bryzgalov and Niemi (who possessed the lowest save percentages among goaltenders in our sample through fifty games) had the highest save percentages in games 50 through 100.

The progression/regression looks like this: Jonathan Quick and Mike Smith essentially maintained the same level of play in games 50 through 100 as they managed in their first fifty games. Bryzgalov, Niemi and Jaro Halak all improved enormously in their second set of 50 games from their first. The other six goaltenders (Rask, Lundqvist, Rinne, Crawford, Mason and Howard) all showed signs of pretty significant regression.

There are a number of reasons why so many goaltenders struggle in the equivalent of their sophmore year in the NHL. Partly it’s bad-luck, partly it’s the fact that teams have now “scouted” their tendencies and weaknesses more closely. Lots of the goalies who regressed over their latter 50 games have bounced back and become super-elite goaltenders (Lundqvist, Rinne); others like Steve Mason are unlikely to be trusted as an NHL starter again in their career.

While Cory Schneider will likely see his career save percentage drop over the next fifty games of his career, based on his NHL performance thus far, the sky’s the limit for him. Sure, he’s played on a quality defensive club and has been sheltered from the majority of tough games thanks to the presence of Roberto Luongo, but the numbers speak for themselves.

Inevitably, Schneider is going to have a bad-stretch of games where he lets in soft goals, suffers from some bad luck, really struggles and has to hear about it in the press, but that’s what being a goaltender at the NHL level is all about.

While the numbers are inconclusive and don’t shed as much light on Schneider’s developmental curve as I’d hoped, what’s certainly clear is that he’s had an incredible start to his NHL career. When “goalie controversy” chatter rears its head in Vancouver, too often it’s preoccupied with Luongo’s performance (“he’s a headcase!!!”), when it should probably be focused on the impressive performance of the team’s talented backup. He has exceeded all reasonable expectations thus far.

Tags: , ,


  1. Kyle
    December 9, 2011

    “Today, Drance looks at Mason Raymond’s reputation as a “perimeter player”.”

    Now, I’ll admit that I’m no expert in advanced statistics, but this article seems like a very roundabout way to explain MayRay’s (whitetails’?) likelihood of storming the crease.

    VA:F [1.9.16_1159]
    Rating: 0 (from 0 votes)
    • Daniel Wagner
      December 9, 2011

      Copy-paste is the devil.

      VN:F [1.9.16_1159]
      Rating: 0 (from 0 votes)
      • Harrison Mooney
        December 9, 2011

        Indeed. The great Satan. Also, half-assed copy editing on my part doesn’t help much either.

        VN:F [1.9.16_1159]
        Rating: 0 (from 0 votes)
  2. J21
    December 9, 2011

    The other thing with Schneider is that he has maintained a pretty consistent development curve at each level of hockey he has played at.

    Modest-to-unimpressive start –> promising improvement –> holy crap, Batman, this guy’s a star!

    VA:F [1.9.16_1159]
    Rating: 0 (from 0 votes)
  3. Josh Peters
    December 9, 2011

    While this is interesting, I would like to see a comparison against a few more “bust” goalies (that aren’t from your timeframe) to see if there are any recurring themes. I would personally pick at least Raycroft, Turek, and my personal favorite, Jim Carey. Although since you have included Steve Mason, this category is probably already accounted for. And I also wonder how being a backup for the first few years compares with being the #1 or #1A – Schnieder is a bit more comparable with Rask, I would say, than with Mason in this regard.

    But I love your work so far, and I really enjoy reading your posts. Thanks for putting in the time!

    VA:F [1.9.16_1159]
    Rating: 0 (from 0 votes)
    • Tom
      December 10, 2011

      I agree, you had my hopes up when you mentioned Mason and Raycroft that you were going to give an analysis on the potential for a disappointment in the future.

      I really like Luongo, and for this season and last Schneider has done well, but to what degree does Schneider getting ‘easier’ starts help his stats? Given, that doesn’t seem to be a factor this year. I am surprised with how sour people are on Luongo because the have all this faith in Schneider to be a better option than him, now.

      VA:F [1.9.16_1159]
      Rating: 0 (from 0 votes)
    • ArtemChubarov
      December 10, 2011

      Thanks for your comment, Josh. The reason i didn’t include goalies before the lockout is that aggregate save percentage has increased steadily since then, which, would skew the numbers.

      VA:F [1.9.16_1159]
      Rating: 0 (from 0 votes)
  4. Anonymous
    December 9, 2011

    Great post as always! I’m still blown away at Steve Mason’s career trajectory.

    When “goalie controversy” chatter rears its head in Vancouver, too often it’s preoccupied with Luongo’s performance (“he’s a headcase!!!”), when it should probably be focused on the impressive performance of the team’s talented backup.

    Extremely true imo, especially from most media, especially from outside BC.

    VA:F [1.9.16_1159]
    Rating: 0 (from 0 votes)