quite a few players have talked about ratings inflation lately. their perception is that lately the ratings have been higher than before for similar rounds. i've also felt the same way for the last five months or so.

i ran the ratings distribution chart against the 4-15-2006 ratings to see if he perception has any basis in fact. take a look at this chart (http://www.earthoffice.net/discgolf/ratingdistribution.htm) . it sure looks like tha ratings inflation is real, but not as pronounced as it feels.

since the 12-15-05 ratings:

* the percentage of rated players less than 900 has dropped from 52.2% to 48.5%

* the percentage of players rated higher than 910 grew from 47.8% to 51.5%

* the percentage of players rated higher than 960 grew from 12.3% to 13.3%

* the percentage of players rated higher than *940 grew from 24.8% to 27.5%

* the percentage of players rated between 910 and 960 grew from 35.5% to 38.2%


i'm not really sure of the conclusions from this, so i will leave it as an observation.

Part of the inflation perception was due to the online unofficial factor not being adjusted to match the official factor for 2006. Dave corrected that over a week ago so the unofficial ratings will now be lower. All player ratings were going to drop in 2005 with the inclusion of all but the bottom 2% of a player's rounds. To offset that drop, we boosted the SSA to roughly match the expected drop. It's possible we overboosted by a few points. In addition, the lower rated propagator versus higher rated propagator factor we introduced in the second half of 2005 boosted the ratings of lower rated players. The double weighting also boosted more lower rated than higher rated players. I believe more higher rated players than lower rated players renew which will slightly skew the numbers upward.

The average points in the rating system had eroded for several years because players on the upswing had to take points from existing propagators. The 2006 SSA adjustment factor is now lower than 2005 but still adds 0.36 to the SSA. That means a 1000 rated player shoots rounds rated at 1003.6 on average intead of the original 1000. Last year, 1000 rated players produced rounds rated 1008.1 on average. If the reduced factor for 2006 is still pumping too many points into the system, we'll ease it back a little more in 2007. However, I expect the current value will balance the overall system point loss derived from fast track newcomers.

quite a few players have talked about ratings inflation lately. their perception is that lately the ratings have been higher than before for similar rounds. i've also felt the same way for the last five months or so.


Possibly lost in his reply, Chuck confirmed this above.

Looks to me that players are getting better. :D

quite a few players have talked about ratings inflation lately. their perception is that lately the ratings have been higher than before for similar rounds. i've also felt the same way for the last five months or so.


Possibly lost in his reply, Chuck confirmed this above.

ummm, thanks for clarifying that rhett :cool:

yeah, they are. someday we'll be complaining that 1000 rated players are still playing AM2 :D

Ratings, much like baseballs, are completely JUICED.

Chuck, I think that post is the most obtuse post I have ever read. :p

Could you explain the change in the "unofficial online factor" that will make the unofficial ratings lower. Isn't a rating calculation pretty much a straight-forward calculation? What do y'all do to ratings between the unofficial state and the official state?

Any ways, my perception of the ratings is that they have gotten much better with the new changes, especially if I'm in a pool of ams. I used to be resigned to the fact that if my pool was dominated by ams, the pro ratings would always be "lower than they should be." I no longer feel that way. For the last year, I feel that ratings have been much more "accurate" then in the past.

Pat, if ams are having those impressions, it is certainly true due to the new method as Chuck explained. If pros around Texas are having those impressions, I suspect it is true because am ratings have gone up and their presence has increased pro ratings in our normal am dominated fields.

yeah i agree that the ratings have seemed better in the last 6-9 months, especailly with am heavy pools. i dont understand all the nuances, especially the arbitrary adjustment factors that seem designed to compensate for changes that are deliberately introduced... but i try not to worry about things i dont understand :)

The unofficial online ratings calculations were done several years ago by a volunteer, Jason Haas, who did a great job duplicating the specific round rating calcs that Roger does. However, Jason has moved on and we have improved the ratings calcs since that time. We don't have anyone specifically responsible for making sure the online unofficial calcs keep up with the tweaks we've done since Jason's work. Theo and Dave do what they can when they can to try and keep the unofficial calcs done the same as the official ones. I think we're up-to-date now with both official and unofficial round ratings being pretty close. They'll not always be perfect because the official process sometimes will exclude someone by the time they are done. Or, the most likely problem is the TD won't know how to or make the effort to map the course layouts online versus getting them correct in the TD report.

A far as 'splainin' the adjustments, I'll save that for another time. Each of them affect the ratings less than 10 points each but they do make the final product a little better and more fair since 2004.