Do The Thrashers Have Large Talons?

Tuesday, December 23, 2008

Luck and the NHL Standings

Luck happens.

In the NHL sometimes the puck hits on the inside portion of the post and it goes into the net and sometimes the puck hits the outside portion and it deflects wide. The adage is that the "bounces" even out in the long run. But is this true? What part of hockey is "luck" and what part is "skill" (a "skill" being a tendency that is repeated over time while "luck" is a tendency that does not repeat i.e. is random over time)

Tyler Dellow had an outstanding post that shed some light on distinguishing team "luck" from team "skill" in the NHL standings. He showed that any club that has a higher (or lower) than average Save Percentage or Shot Percentage at Even Strength can not repeat that pattern later in the season.

What does that mean in plain English? Teams can get hot shooting the puck in the short run but in the long run your ES offense is a product of getting shots on net--you can not count on having an above average ST% all season. The same is true for you team goaltending--a net minder might get hot for a while but basically if you're giving up a ton of shots eventually it is going to kick your GAA in the teeth--that hot net minder will return to Earth.

The bottom line is that shot ratios matter. Good teams out-shot their opponents at ES and bad teams don't--you can only avoid gravity for a short while. A team's SF/SA ratio is a strong predictor of their final ESGF/ESGA ratio.

In his post, Tyler put up a list of the luckiest to least lucky as of November 30th. I've updated what he did so that it is up to date through 12-22.

How is this information useful to you the fan? Well the teams at the top of this list are more likely to see a fall in the standings while the teams at the bottom of the list are due a few more positive bounces the rest of the way--assuming these teams ES Shots For/Shots Against ratio remain steady the rest of the season.

Luckiest NHL Clubs at Event Strength
1.14 BOS (don't hand them the Cup just yet)
1.12 PIT (they're good, but not quite that good)
1.11 DAL (yikes, look out below)
1.10 NJD
1.09 ATL (looking more like a lock for the Taveras/Hedman lottery)
1.08 VAN
1.08 CHI
1.05 ANA

Teams with little "luck" bias so far.
1.03 MTL
1.02 CBJ
1.01 PHI
1.01 NSH
1.01 WSH
1.00 TOR
1.00 SJS
0.99 FLA
0.99 STL
0.99 EDM
0.98 PHX
0.98 CGY
0.97 BUF

The most "unlucky" clubs so far:
0.96 COL
0.94 MIN
0.94 CAR
0.93 DET (scary thought)
0.92 LAK
0.91 TBL
0.89 NYR (better than they look)
0.88 OTT (will better luck be enough to get them back into the hunt?)
0.84 NYI

4 Comments:

  • Do these numbers take into account the quality of shots in regards to shooting percentage?

    By Blogger Richard Pollock, at 5:57 PM  

  • no shot quality adjustments, just raw numbers.

    My hunch is that there is a correlation between teams that give up a lot more shots against and teams that give high quality shots against--bad teams probably allow both types.

    By Blogger The Falconer, at 7:24 PM  

  • I agree.

    I would be interested to see shot quality adjustments, especially in terms of the better teams. Maybe a team like the Bruins would be less apt to fall of its current pace if the shot quality was distinctly better than other teams in its class?

    By Blogger Richard Pollock, at 7:38 PM  

  • I don't know if quality of shot would be a factor. Many poor shots are actually goods shots since they often create rebound opportunities for teammates.

    By Anonymous Anonymous, at 4:27 PM  

Post a Comment

<< Home


 
Who links to my website?
View My Stats