The Pavelec Perplexity

 

 

I don’t necessarily have a bone to pick with Ondrej Pavelec, because at the start of the season if you told me that the Winnipeg Jets were going to make the playoffs, I would have told you that it would be “thanks to” and not “in spite of” the young Jets goaltender.

Pavelec has played some amazing hockey games this season, however, he’s also put up some stinkers. Remarkably inconsistent, but terrifyingly fun to watch when he’s at his best, Pavelec is the kind of goalie who could win you a playoff series as an underdog.

However I think that people tend to get carried away with some saves that he’s made. Specifically, this one against Philadelphia:

What’s great about this video is that it’s not only a spectacular, out of nowhere save by Pavelec, but it isn’t like it’s desparation. He still pushes himself off with his skate and makes a stretching goalie motion rather than diving across the goal mouth. It’s also the only time a save has been made in a Winnipeg/Philadelphia game this season.

But there’s still a bit of a concern with Pavelec. A great save or a crucial save at an important time can fuzz our opinion on the quality of a goaltender, making us forget about what really matters: does the goalie make saves? If the goalie does make saves, does the goalie make more saves than a replacement-level goalie?

Before we go too far into this, let’s remember that Pavelec is just 24 and really shouldn’t be counted on. Yes, Marc-André Fleury won a Stanley Cup at Age 24, but he had a .912 save percentage that year and played behind one of the best two-way forward groups assembled post-lockout. Jonathan Quick was a .905 at 24 and has become a top goalie. Brian Elliott was a .910 at 24 and was an All-Star this season. So you don’t need to be too worried about Pavelec’s play right now (he’s a .911) affecting his future, because there’s still a long career ahead.

Does Pavelec make better saves than a replacement goalie? I went through every team and found goalies that weren’t expected to be one of the best two goalies on their teams. Essentially, there are 20 goalies who have seen NHL action this season that were, literally, replacements.

Tallying them all up, I found that the replacements have a save percentage of .909, slightly below Pavelec’s .911 (he’s ahead by three goals against) and Chris Mason’s .910.

On a game-by-game basis, Hockey Prospectus runs with a statistic called “quality starts”. A quality start is a game where a goalie stops either 91.3% of the shots faced, or allows two or fewer goals and saves 88.5%. A good goalie will maintain a quality start rate of over 55%. But in the 69 games started by replacement goaltenders this season, they’ve had an astonishing rate of quality starts: 58%.

Joey MacDonald, Leland Irving, Al Montoya and Ben Scrivens have all contributed to this, while only Dallas’ Andrew Raycroft is a replacement goalie who has started more than twice and has a QS rate of less than 50%.

Pavelec, however, also has a 58%. So we can’t really distinguish him from a replacement-level goalie that way. The opposite of quality start, the “blow up” created by Thomas Drance of the Nations’ own Canucks Army, is a game where a goalie stops 84.9% or fewer of his shots, or allows 5 goals while seeing 39 or fewer shots.

“The Replacements” as we’ll call them, have a high BU% of 14.5% (a real good goalie will be between 10 and 12%) but Pavelec’s is much higher at 16%. That’s an extra start where Pavelec’s play cost his team a chance at winning than a replacement-level goalie would.

Clearly, there is a very good crop of replacement goalies this year, whether by waiver wire pick-up or minor-league call-up for a team with an injured goalie. They’ve been a little prone, though, costing their team every seventh game. So has Pavelec.

So while Pavelec may make some big saves every now and again, or get on the highlight reel with a wave of “Pavelectricity”, his play this year has not been good enough to push Winnipeg ahead of where they could be with a shrewed waiver wire pickup.

Here’s the full chart;