Comments for Forensics - Tuukka Rask at http://hockey.dobbersports.com , comment 1 to 2 out of 2 comments http://hockey.dobbersports.com Sat, 18 May 2013 11:41:36 +0100 FeedCreator 1.7.3 http://hockey.dobbersports.com/index.php/columnistsarticles-mainmenu-77/918-frozen-pool-forensics/4672-forensics-tuukka-rask#comment-21169 Those numbers also give a vote of confidence for Schneider... Who would you rather own in a salary cap league (keep players until their contract ends, then complicated rules on resigning) - austeane Fri, 08 Jun 2012 13:19:45 +0100 http://hockey.dobbersports.com/index.php/columnistsarticles-mainmenu-77/918-frozen-pool-forensics/4672-forensics-tuukka-rask#comment-21166 I like numbers, but let's be honest... Rask had a very small sample size in 23games. I'd have loved to see his past 3 years stats combined - even compared to just this year's goalie stats. (3 years of goalie stats = ideal). Evaluating small sample sizes is a statistical no-no, not something you should lean on. As an example, take a quick look at Rask's game log. http://espn.go.com/nhl/player/gamelog/_/id/3405/tuukka-rask A quick inspection suggests that Rask wasn't even facing top level competition. 23 games this year including WPG(3), CBJ(2), NYI(2), EDM(2), CAR(2), CGY(1), Look at the top 10 scoring NHL teams (Rask GP against):. 1. PIT(0) 2. BOS(-) 3. PHI(0) 4. OTT(0) 5. VAN(0) 6. CHI(0) 7. DET(1) 8. NAS(0) 9. TB(0) 10. TOR(1) So, Rask played exactly 2/23 games against the high-scoring teams in the NHL. Small sample size = potential for skew. Great number crunching. Next time grab a bigger sample (3yr) - Pengwin7 Fri, 08 Jun 2012 08:22:56 +0100