NHL's special team goaltending statistics - don't add up?

[Doctor No]

I've been wanting to do more with situational special teams goaltending save percentages. The first step, of course, is getting the data into my database.

However, looking here:
http://www.nhl.com/ice/playerstats....=G&country=&status=&viewName=specialTeamSaves

It shows Sean Burke (Carolina) with 446 even-strength shots faced, 123 power play shots faced, and 31 shorthanded shots faced (600 total).

However, Burke didn't face 600 shots that year with Carolina - he faced 655 (even NHL.com agrees on that).

So where did the missing 55 shots go?

If the league's totals don't balance, I'm going to have a hard time taking the results seriously.

Click to read more...

 

[Mantha Poodoo]

So I checked Trevor Kidd on the same team on the same year, and same thing. The numbers don't add up. Same for Ozzie. Brodeur was even funnier... his total saves were uh, higher than his total saves.

So I decided to check a number of goalies in the 98-99 seasons and later years. So far, this oddity seems to exist just for the 97-98 season on the NHL site. I guess the best thing to do would simply be to avoid using 97-98 if possible.

 

[hatterson]

My best guess (and it's exactly that) are they the numbers only count 5 on 5 (ES), 5 on 4 (PP), and 4 on 5 (PK), and other situations that fall under those categories (4 on 4, 3 on 3, 5 on 3, 6 on 5, etc.) aren't included on that screen.

Although it's interesting that the data is missing 55 shots but zero goals. Team wide it's missing 118 shots, but also zero goals. That seems rather unlikely if my above statement is true.

 

[Mantha Poodoo]

My best guess (and it's exactly that) are they the numbers only count 5 on 5 (ES), 5 on 4 (PP), and 4 on 5 (PK), and other situations that fall under those categories (4 on 4, 3 on 3, 5 on 3, 6 on 5, etc.) aren't included on that screen.

Although it's interesting that the data is missing 55 shots but zero goals. Team wide it's missing 118 shots, but also zero goals. That seems rather unlikely if my above statement is true.

I think the 97-98 stats are just hosed. I don't see any of this inconsistent data in any other year.

 

[hatterson]

I think the 97-98 stats are just hosed. I don't see any of this inconsistent data in any other year.

Yea that's entirely possible. Now that you mention it, I think I may have noticed something about that before when running data for other projects. Given that it's the first year that NHL.com has that data it's not entire shocking that they'd have a couple holes.

 

[Doctor No]

Thanks, guys! Figures that I'd have started with the oldest year.

If I'd started with this year and worked backwards, like a normal person, I might have made it a long time before paying attention.

 

[Mantha Poodoo]

Thanks, guys! Figures that I'd have started with the oldest year.

No problem! I've always liked following your goalie work.

If I'd started with this year and worked backwards, like a normal person, I might have made it a long time before paying attention.

Well, there are advantages to being abnormal.

 

[Doctor No]

No problem! I've always liked following your goalie work.

Thanks! Hopefully I can get some predictive power (lift) someday.

Well, there are advantages to being abnormal.

I've been relying on that for more than forty years.

 

[BM67]

I noticed that Luongo's split in 2006-07 didn't include a SH penalty shot, that was included in his overall totals. Doubt that's the answer here though. Burke only has 39 saves in the two games he didn't allow a goal with Carolina, so it's not just a couple of shutouts missing either.

Not goalie related, but Chris Pronger's TOI in his game logs shows two games with 0:00 played and one with 50:00 played in 1997-98. Al MacInnis also shows two 0:00 played games.