I realize this isn't a new concern, its been voiced before, but it just bugs me.

http://www.pdga.com/tournament-results?TournID=9740&year=2010&include_ratings=1#Open

tournament i'm talking about ^^^

just voicing the same opinion that has been stated before, local tournament ratings suck and they are freaking squashed down compared to the bigger tournaments. I don't know what to do to fix it, but it just doesn't work out right. the haikey creek classic this weekend has some crappy ratings. I understand that it has to do with lower rated people shooting closer scores to the higher rated guys. I know its not technically an "error" as far as the calculation goes, I'm just saying its messed up. Devan Owens won the tournament by 10 strokes, he played phenomenal all weekend, almost every large A tier or NT he goes to he cashes well and shoots over 1030, he crushed everyone this weekend and his average was 1026.67, which I know is just freaking low for how well he shot those rounds. I have had a couple 1040 rounds and quite a few up over a 1000 and typically know how I am shooting within 10 points or so, based on how many 30 footers jump putts and bogies I have in a round, this was one of my best rounds Ive ever played in a sanctioned event and it was 1026 rated, which just feels at least 20 points too low (the other rounds feel low too, by 10-20 points). I know it was a 24 hole layout that tends to have lower point spread per stroke, but when you start talking about beating all of these solid players by 2, 3, 4, 5 strokes per round there should be some kind of amplifier on those extra strokes that you were able to keep off.

It just sucks for our city, we have a lot of talent coming up and I know ratings are just a number, but I just hate how they are higher with bigger 1000 rated fields. whatev

There's never been any indication the average rating of the field matters and it has been analyzed in depth due to posts like yours which pop up regularly. The bottom line is what SSA is generated. The unofficial Haikey SSA for the event of 66 is in line with previous SSAs for 24 holes there.
http://www.pdga.com/course-ratings-by-course?RatingCourseID=294
The primary variance in SSAs on the same layout is due to wind, not the ratings of the players. Courses that have events like NTs and local C-tiers have the same SSAs for the same layout which means the players get the same rating for the same score at either event.

I think it does matter when your dealing with smaller tournaments, in smaller PDGA areas, with not a lot of propagators. Especially when there are several low rated propagators who are locals. Here is a great example:

http://www.pdga.com/tournament-results?TournID=10583&year=2010&include_ratings=1#Open

I beat a 984 rated player by 6 strokes and a 979 rated player by 8 strokes, won the tournament, yet I averaged about 12 points BELOW my rating - that is crap any way you look at it

basically I get screwed because there were not enough high rated propagators to balance out all the low rated propagators (most of which were locals)

It's a fallacy. The ratings calculations work just as well with 50 props in the 850s as 50 props over 950. It does not matter. Everyone pulls out their favorite example without looking at all of the data. Yes, there are individual examples that "prove" either side of the argument. It's a statistical process with normal variances in both directions.

But the proof is in the overall data from more than a million rated rounds now. My favorite example on the other side is Darrell Nodland of North Dakota who has a higher rating than the next closest player entered in C-tiers up there by sometimes more than 50 points. And yet he has no trouble retaining his 1025-1030 rating over the years in events with few and much lower rated props.
http://www.pdga.com/tournament-results?TournID=6944&year=2010&include_ratings=1#Open
http://www.pdga.com/tournament-results?TournID=8364&year=2010&include_ratings=1#Open
There's the definition of destroying your competition.

I'm not professing to know all the ins and outs of the system like you do chuck, but I do know the feeling of a very well played round and when I've done it in A tiers, and even in other B tiers before it was 1040+ rated, my final round at this tournament felt no different, if not better than my 1040+ rounds, yet it was 1026. Just saying its off a little.

I don't really care that much honestly, its just that the rating is kind of the number everyone looks at... and its hard to beef your number up when you don't play a lot of A tiers and NTs (I have no proof other than experience enough to know my ratings are always higher in higher rated fields and larger fields).

It's true that on an individual basis, some players play better the higher the tier but just as many play worse. But the ratings awarded are consistent for the same score at all tier levels within normal stats variances. I just saw the first draft of this next ratings update and Nodland is at 1037, third in the World after scorching another C-tier with few propagators that have much lower ratings.

its less to do with "small C tiers" as it is to do with "small PDGA areas"

the 912 rated player who finished 4th in my example only has 12 rounds to his name (4 of them being at a tournament 6 hours away) and he has improved fast, so his rating is really lagging behind his true skill - I can not possibly beat him on this course by the expected 81 ratings points

when you get about 5 of these type players at a tournament like this, there aren't enough established PDGA members to balance it out - so when they all shoot their true skill level, the ratings only assume they are shooting what their rating says their skill level is, which then brings the average SSA down - this isn't just 1 example, this same tournament with these same players will produce crappy ratings every single time

the example I gave shows how a small PDGA event in MS, an area with very few PDGA members, and even less PDGA members with more than 15 rated rounds to their name, can have the ratings swayed quite significantly by these few low propagators - I know there is nothing you can really do, you have even noted this problem before with having propagators only have 8 rounds, but it doesn't change facts that this is still a problem right now and will always be a problem in these type areas with fast improving players with not enough rated rounds to catch up to their true skill level

personally, I like the ratings - I think overall, they do a good job, but consistency is always going to be a problem - if you rate the courses, the conditions change, if you use players as propagators, players change

the reason small tournaments catch the brunt of this blame, has nothing to do with the size of the tournament, but the skill level of the participants

take for instance Player A: an 850 rated player who is improving rapidly - he has a lot of room to improve - a year from now, his rating could get up as high as 950

also Player B: maybe he is a 900 rated player, who is also improving rapidly - he also still has a lot of room to improve - a year from now, his rating could also get up to 950 or continue to climb

then there is Player C: he is a 1000 rated player, who is also improving rapidly - he is rounding into the peak of his game - although he is still improving rapidly, he can only get so much better - even if he becomes the best in the world within a year, he can only improve 35-41 points, but he is much more likely to only be able to increase his rating 10-30 points in a years time

big tournaments have a lot more player Cs - which pretty much contains everybody 980 and up - there are a lot more established players in this ratings range who have reached their peak and the up and coming fast improving players, can only improve so much

however, smaller tournaments, especially in the smaller PDGA areas, have a lot more player As and Bs - there are a few established players in these ratings ranges, but are generally far outnumbered by these faster improving players, whose rating is lagging behind their true current skill level - these players will bring down the average SSA of a course, no matter how you look at it

I think I read where you said the ratings formula makes a slight adjustment to correct for this Chuck, but even then, it can only do so much - the ratings formula simply does not know how fast a player is improving and how many of these types of players there are at a given tournament, so even with a slight adjustment, the problem remains the same (just less obvious)

How does Nodland keep his rating so high?
That is crazy. Guy must be a natural.

It will not be a problem most of the time due to statistical variance, just some of the time. But conditions have to be perfect for the system to break down. If a course doesn't have previous SSA values, there are no numbers to compare a new tournament SSA against to see if the "fast improving prop" is impacting the event. But most of the time, when SSAs are compared between years or C-tiers versus A-tiers on the same layout, the variance is within normal variation of +/- 2.0 on SSA. Darrell has had the same issue in ND as in MS with a handful of fast improving props and has never seen the effect on his rating.

I have seen the effect that Mr. Nash describes, but it is often more a function of the composition of the pools than the tier of the event. For example, at the 2004 Brent Hambrick the pros played in the NT and the Ams played in the similarly named sort of simultaneous A-tier. We each played one round per day but the weather did not change much. Other than one little storm cloud that seemed to hover over just one pro player's head, it was beautiful all weekend. In the Advanced/Intermediate pool you had to shoot two or three better throws to get the same rating as the pros. The pros would have had many more players who had peaked or were slowly improving. We would have had many more players who were rapidly improving.

As Chuck has pointed out, at most tournaments the pools get combined and those differences get averaged out. For an example of that, see the 2009 Michigan State Championship. Geoff Bennett and I shot the same score at Flip City. Unofficially, he got a phat 1030 and I got hosed with a 970. Officially, our pools were averaged together and we both got something just under 1000.

But, at the tournaments where they are not averaged together, because maybe the Intermediates and below played short tees, or maybe they played a similarly named tournament that was submitted on a different spreadsheet, the pros get their pro pool bonus and the lower ams get hosed.

If you care whether your rating is artificially high, you can focus on playing those tournaments where you are not pooled with AM2s and below, or focus on playing with them if you want to bag. It seems to me that if you only played in the lower am pool at tournaments where there was no inter-pool averaging, you could bag your rating by 25 points.

Chuck, what do you say to the TDs that have, just for fun, inserted Climo or any other 1030+ player into their results and the ratings went up regardless of the score that was entered?

Basically the ratings went up just for Climo "showing" to the event.

Nonsense. If Climo's score entered is less than his rating, then it will not boost ratings when done officially.

This may be the case because you would still be predicting his score. Maybe the problem lies within the system itself and how it factors in things. I don't understand it completely but is it a correct assessment that you don't factor in anything for different conditions such as wind and rain because the propagators scores will average this out? If this is the case then there are likely too many unknown factors and not enough data to give an accurate rating for an individual tournament, especially a smaller one. Sure it may work out from a statistical standpoint over hundreds of rounds and several years on a particular layout but not tourny to tourny.

Another thing I've seen before and question is how it could be possible for a 51 to have a six point higher rating than a 50 that were played back to back on the same day, same course, same basket positions, and same weather conditions? Example can be provided if needed. thanks

I have seen the effect that Mr. Nash describes, but it is often more a function of the composition of the pools than the tier of the event. For example, at the 2004 Brent Hambrick the pros played in the NT and the Ams played in the similarly named sort of simultaneous A-tier. We each played one round per day but the weather did not change much. Other than one little storm cloud that seemed to hover over just one pro player's head, it was beautiful all weekend. In the Advanced/Intermediate pool you had to shoot two or three better throws to get the same rating as the pros. The pros would have had many more players who had peaked or were slowly improving. We would have had many more players who were rapidly improving.

As Chuck has pointed out, at most tournaments the pools get combined and those differences get averaged out. For an example of that, see the 2009 Michigan State Championship. Geoff Bennett and I shot the same score at Flip City. Unofficially, he got a phat 1030 and I got hosed with a 970. Officially, our pools were averaged together and we both got something just under 1000.

But, at the tournaments where they are not averaged together, because maybe the Intermediates and below played short tees, or maybe they played a similarly named tournament that was submitted on a different spreadsheet, the pros get their pro pool bonus and the lower ams get hosed.

If you care whether your rating is artificially high, you can focus on playing those tournaments where you are not pooled with AM2s and below, or focus on playing with them if you want to bag. It seems to me that if you only played in the lower am pool at tournaments where there was no inter-pool averaging, you could bag your rating by 25 points.

http://www.21softs.com/emoticons/images/rockon.gif