GhostofPotvin29
New member
Champ Kind said:Potvin29 said:Also, it was a short season. Over 82 games this past year's team likely regresses further to the mean than the season let them.
On what basis can you conclude they 'likely' regress? If I have a population of 82, I feel pretty darn confident that I can extrapolate the results of 48 case sample across the entire population.
Also, weren't you just arguing that Grabovski's 2009-2010 point totals should have been extrapolated across an 82 game schedule?
What? I'm getting confused what you're asking. Yes, Grabovski's 2009-2010 point totals should be shown based on a PPG basis to compare it to a normal season - why would you compare point totals from 58 games with 82 games? There's a whole lot of games missing to get points, it's unfair.
The idea of regression for a team is different - if a team has an unusually high SH% or unusually low SH% it is a good bet that it will either decrease or increase over time. If the Leafs have a SH% that is well above the norm (the highest in a season is typically ~9% for a team) then it is likely to fall - more likely to fall than it is to say that the Leafs teams are historically great snipers. It is just saying that if a team is too high or too low that it should correct itself - New Jersey had the opposite problem this season (partly also due to injuries). They couldn't score despite playing well - a trend unlikely to continue.
I've run into problems before so I will emphasize the phrases/words, "good bet", "unlikely" etc.