SSW: With so much luck, when are stats useful?

With all the talk about PDO, and given the time of year I figured it would be appropriate to show the final results from my work on traditional stats (G, Ast, Pts, PIM) reliability.  As has been discussed a lot lately PDO numbers tend to stabilize for teams around 1000.  This begs the question, what stats can we use for individual players? When are they reliable?  Here's a nice chart after the jump for you to ponder.  (ref early SSW work for methods herehere, and here)

Basically the second column on the left is the number of games a player must play before we can say with reasonable confidence that the stat in the row (traditional counting stat)  is significant (ie. probably accurate)

For Forwards;

 Stat   Games when alpha > 0.707   95% Confidence Interval   alpha at 120 games 
 Goals 40 27-51 0.860
 Assists 25 18-34 0.911
 Points 14 10-19 0.938
 +/-  n/a n/a 0.447
 PIM 36 29-47 0.907






For Defensmen

 Stat   Games when alpha > 0.707   95% Confidence Interval   alpha at 120 games 
 Goals 77 60-114 0.777
 Assists 38 31-47 0.891
 Points 31 21-36 0.918
 +/-  n/a n/a 0.449
 PIM 69 52-107 0.786






Alpha here is cronbach's alpha which is generated from a statistical formula that calculates the reliability of a test. It's basically a fanchy-schmancy correlation coefficient.  If you want you could just take it as the r at the games specified. Ref early work on how these numbers can be used to calculate regression statistics.

My intention was to compute these for advanced stats as well, but it will be a lot more time consuming.  I may or may not do it, time permitting.  If you really want to see those numbers I could explain how, and let you crunch the numbers.

Contact me for the full spreadsheet

If this FanPost is written by someone other than one of the blog's editors, the opinions expressed in it do not necessarily reflect those of this blog or SB Nation.