View Single Post
Old
08-15-2012, 11:17 AM
  #1
overpass
Registered User
 
Join Date: Jun 2007
Posts: 3,539
vCash: 500
On-ice Shooting percentage - luck or skill?

The answer to the question, is of course, both. But in what degree?

5 vs 5

I used behindthenet.ca's .xls files, with data from the 2007-08 season through the 2010-11 season.

EDIT: See post 22 for updated numbers - I used saved shots instead of total shots for these numbers.

Using the binomial approximation to a normal distribution, I calculated the z-score for each NHL player's on-ice shooting percentage over this time period, where z-score = (On-ice GF - On-Ice SOGF*LgSH%)/(Standard deviation calculated using binomial approximation)

Here are the players with the highest and lowest z-scores

Player Z-score 5-on-5 on-ice SH%
SIDNEYCROSBY 6.42 13.3%
HENRIKSEDIN 5.75 12.5%
DANIELSEDIN 5.22 12.3%
MARIANGABORIK 5.17 12.9%
BOBBYRYAN 5.15 12.7%
EVGENIMALKIN 5.12 12.4%
RYANGETZLAF 4.98 12.1%
ILYAKOVALCHUK 4.93 12.2%
COREYPERRY 4.59 11.8%
DANYHEATLEY 4.54 11.8%
JASONSPEZZA 4.52 12.0%
ALEXTANGUAY 4.36 12.1%
NATHANHORTON 4.29 11.8%
ALEXBURROWS 4.27 11.8%
JAROMEIGINLA 4.27 11.5%
RYANWHITNEY 4.21 12.1%
PAULSTASTNY 4.14 11.8%
MARTINST. LOUIS 4.12 11.5%
MIKERIBEIRO 4.09 11.7%
NICKLASBACKSTROM 4.07 11.4%
ALEXANDERSEMIN 3.97 11.7%
PAVELDATSYUK 3.86 11.3%
DARRYLBOYCE 3.80 17.5%
ALEXOVECHKIN 3.77 12.1%
JEFFSCHULTZ 3.76 11.4%
STEVENSTAMKOS 3.75 11.8%
NICLASHAVELID 3.53 12.4%
ALEXANDEROVECHKIN 3.48 11.7%
BRENDENMORROW 3.48 11.6%
JOETHORNTON 3.43 11.1%

Player Z-score 5-on-5 on-ice SH%
TRAVISMOEN -4.09 6.0%
SHAWNTHORNTON -3.91 5.8%
COLTONORR -3.89 4.7%
RADEKMARTINEK -3.79 6.3%
ADAMHALL -3.78 5.3%
BRENDANWITT -3.77 5.8%
DEREKMEECH -3.69 4.7%
CRAIGADAMS -3.66 5.7%
RYANHOLLWEG -3.63 2.6%
DONALDBRASHEAR -3.58 4.6%
STEPHANEVEILLEUX -3.49 6.0%
JEFFTAMBELLINI -3.45 5.9%
ANDYGREENE -3.43 6.8%
NATETHOMPSON -3.37 5.8%
RODPELLEY -3.33 4.8%
TODDMARCHANT -3.27 6.3%
SAMUELPAHLSSON -3.23 6.5%
RAITISIVANANS -3.19 4.5%
FREDRIKSJOSTROM -3.19 6.3%
TIMSTAPLETON -3.19 3.6%
DANIELWINNIK -3.12 6.7%
SCOTTGOMEZ -3.09 7.2%
ANTTIPIHLSTROM -3.06 4.0%
TOMWANDELL -3.06 5.5%
ANDREWMURRAY -3.05 5.8%
THOMASPOCK -2.90 4.3%
JAMIEMCGINN -2.90 5.7%
ANDREWPETERS -2.88 2.5%
MIKEWEAVER -2.86 7.0%
DALLASDRAKE -2.82 3.8%

Generally speaking we see skilled players outperforming the league average and "less skilled" NHL players underperforming the league average in on-ice shooting percentage.

Taking the standard deviation of the z-scores, I get 1.47. If random variation was the only factor, the standard deviation of the z-scores would be 1.00.

Following the process from this blog post:
Quote:
Step 1: Figure out how much one standard deviation is based on a binomial distribution. Vokoun faced 4249 shots, and the league average save percentage was .920, so one SD is sqrt (.92*.08/4249) = .0042.

Step 2: Figure out how much away you are from the mean. Vokoun’s save percentage was .931, and so was +.0108 from the mean.

Step 3: Figure out how many SD that is. .0108/.0042 = 2.59. That’s his z-score.

Step 4: Do it for all the goalies. (Thomas is 2.57, Luongo is 2.53… Holmqvist is -3.11, Raycroft is -2.34).

Step 5: Find the standard deviation of all the z-scores. In this case, for these 55 goalies, it’s 1.38.

Step 6: Rejoice if the number is substantially higher than 1.00. Happiness sets in at 1.10. You did good at 1.20. If you get 1.40, you’ve definitely found something.

Step 7: Figure out the average number of opportunities for each player. In this case, the average shots faced was 2665.

Step 8: Do this: 1 - 1/1.38^2 = 0.47. That’s your r or r-squared. (Longer story later. Just call it r for now.) That 1.38 was from Step 6.

Step 9: Do this: (1-r)/r * 2665. We get 2969. The 2665 is from Step 7.

That’s the key number. 2969. Let’s call it 3000. That’s how much you use to regress a goalie’s performance. You add 3000 shots of league average performance. So Vokoun’s 4249 shots at .931 save percentage gets added to 3000 shots at .920 save percentage for a best-estimate true talent level of .926. Holmqvist’s .900 with 1809 shots becomes .912. So, the observed difference between the two goalies (.031 saves per shot) becomes a true difference of .014.
Step 7: 768
Step 8: 0.54
Step 9: 657

So, given a player's on-ice shooting percentage from 2007-08 to 2010-11, the best predictor of his 2011-12 on-ice shooting percentage is his 2007-08 through 2010-11 shooting percentage plus 657 shots at a league average shooting percentage. Because Sidney Crosby was on the ice for 1867 shots at a 13.3 SH%, his predicted 2011-12% would be 12.2%.

Best and worst predicted 2011-12 5 vs 5 on-ice shooting percentages

Player Predicted 5vs5 on-ice SH%
SIDNEYCROSBY 12.2%
HENRIKSEDIN 11.7%
MARIANGABORIK 11.7%
BOBBYRYAN 11.6%
EVGENIMALKIN 11.5%
DANIELSEDIN 11.5%
ILYAKOVALCHUK 11.4%
RYANGETZLAF 11.4%
JASONSPEZZA 11.2%
ALEXTANGUAY 11.2%

Player Predicted 5vs5 on-ice SH%
COLTONORR 6.8%
TRAVISMOEN 6.9%
ADAMHALL 6.9%
SHAWNTHORNTON 6.9%
DEREKMEECH 7.0%
BRENDANWITT 7.0%
CRAIGADAMS 7.0%
DONALDBRASHEAR 7.0%
RADEKMARTINEK 7.1%
STEPHANEVEILLEUX 7.1%

I'll see if I can get the 2011-12 data to test the predictions.


Last edited by overpass: 08-16-2012 at 12:33 PM.
overpass is offline   Reply With Quote