Fred Taylor
The Cyclone
- Sep 20, 2011
- 3,174
- 31
Question about adjusted stats
- Thread starter Fred Taylor
- Start date
Micklebot
Moderator
- Apr 27, 2010
- 53,842
- 31,052
The era adjustment component is done separately for goals and assists, so if there were more assists per game that year, it would have a similar impact on his pts totals. Assists are adjusted to 1.67 per goal. Depending on how they treat rounding, the difference of including SO goals would have either lost him 1 goal, or none at all. Since his adjusted numbers are only different for assists, we can assume that 2005/06 had more assists than usual, as SO goals even if counted do not have an assist, meaning it would artificially deflate assists per goal.
Hope this helps even if it doesn't directly answer your question.
To answer your question, they most certainly do include shootout goals. EDIT: Maybe I'm wrong but the below point still stands.
Another feature of their adjusted stats that I am not a fan of is that the site removes all goals a player contributed to when calculating the league scoring levels, which disproportionally benefits players in smaller leagues. For example, if a player in the O6 era contributed to half his teams goals, HR's calculated GPG average used in their adjusted points formula for that season could be close to 10 percent lower than it actually was. In modern times however, a player with a similar performance would only have the scoring average drop 1-2%.
TheDevilMadeMe
Registered User
If they are adjusting goals and assists separately, they probably are not including shootout goals since winning in a shootout, only counts as a "goal" for the purpose of team stats, not individual player stats.
Fred Taylor
The Cyclone
- Sep 20, 2011
- 3,174
- 31
TheDevilMadeMe
Registered User
Yes, that's an issue, and so is their "roster size adjustment," which adds another level of pollution to the data.
I just did a simple adjustment of some of Jagr's seasons, and the adjusted totals when using http://www.dropyourgloves.com/Stat/LeagueGoals.aspx , which is known to include shootout goals post-lockout, match up very well with Jagr's seasons, both before and after 2005. When I used quanthockey's values for post lockout, they seemed to be slightly (~2 points) higher than HR's adjusted stat totals after 2005.
The best way to figure this out is to replicate their calculations using the method shown on their site (http://www.hockey-reference.com/about/adjusted_stats.html)by both including and not including SO goals and see which one matches HR's final adjusted point values.
The best way to figure this out is to replicate their calculations using the method shown on their site (http://www.hockey-reference.com/about/adjusted_stats.html)by both including and not including SO goals and see which one matches HR's final adjusted point values.
