Disc Golf Center
  #31  
Old 02-02-2014, 01:23 AM
TalbotTrojan's Avatar
TalbotTrojan TalbotTrojan is offline
* Ace Member *
 
Join Date: Oct 2008
Location: Highlands Ranch, CO
Years Playing: 5.6
Courses Played: 86
Throwing Style: RHBH
Posts: 3,013
Quote:
Originally Posted by Discette View Post
To go back to the original question/statement. The OP likes the frequency of ratings updates and thinks they are sufficient.




In this area, there are a lot of amateur players that are very concerned with ratings and the frequency of ratings updates. Some players are upset that the first ratings update of 2014 won't be until March 18.

As a general rule, rapidly improving players, and those that place a lot of value on player ratings, are the ones that think the ratings should be updated more often. These players are more inclined to think there are flaws with the system.

Established and declining players seem more satisfied with the fairness/accuracy of the ratings and the frequency of updates.

My rating does not affect my enjoyment of the game. (Says the established player with the declining rating.)
Hmm, I think you make a good point at least in terms of frequency. I tend to like the frequency where it is at and happen to be improving. I have clearly questioned the accuracy of rating across regions. But that is really the only complaint I ahve about the system, and that may be a mere consequence with little taht can be done to correct it. In general I think it is a good system and I am not asking for it to be redone or changed in anyway. Like you said, it is just a number and does not change the way I play the game. I simply enjoy having a rating as a way to measure improvement over time.

By the way, thanks for all that you have done in So Cal. It had an impact on my enjoyment of the game and playing in tournaments.
Reply With Quote
  #32  
Old 02-02-2014, 01:25 AM
TalbotTrojan's Avatar
TalbotTrojan TalbotTrojan is offline
* Ace Member *
 
Join Date: Oct 2008
Location: Highlands Ranch, CO
Years Playing: 5.6
Courses Played: 86
Throwing Style: RHBH
Posts: 3,013
Quote:
Originally Posted by krupicka View Post
As a TD I like more frequent updates to keep things fair. I think this year's update schedule is pretty good. For me though, it would be great if the time between the deadline for TD reports and the ratings update could shrink a bit. Last year there were a couple of updates with 3 week lead times in prime season. For established players this doesn't matter too much. But for new players without ratings, it often extends the time that they are PDGA members without a rating.

Our policies have always been that PDGA members can play in whatever division their rating qualifies them for. If they have no rating, then they are qualified for any division. (If they are not a member, we can use other criteria for limiting their division choices). That long lag matters more in that case.

I guess I'm fine with monthly updates, but let's get the deadlines down to a week before the update rather than the 2-3 we have now. To me, streamlining that part of the process would be a good goal for the PDGA.
I know Mr. Sweeton has a ton on his plate and the amount of work he has to do for an update is significant. I think you make a valid point as to fairness at tournaments and newer players taking advantage of the situation. I wonder if there is another solution to this problem?
Reply With Quote
  #33  
Old 02-02-2014, 08:25 AM
JC17393 JC17393 is offline
Double Eagle Member
 
Join Date: Aug 2012
Years Playing: 16.3
Courses Played: 112
Throwing Style: LHBH
Posts: 1,672
Quote:
Originally Posted by TalbotTrojan View Post
You make a valid arguement. What you are saying is very true.

In this instance it is not the only possible solution. Based on the traveling I have done before and after my move, there still appears to be an imbalance in ratings from one region to another.
Instead of just alluding to these imbalances, provide an example. What region/event did you travel to where you perceived an imbalance that couldn't be easily explained as a region where local custom doesn't observe PDGA ratings suggested criteria for their divisions.
Reply With Quote
  #34  
Old 02-03-2014, 07:24 AM
krupicka krupicka is offline
Birdie Member
 
Join Date: Sep 2008
Location: Naperville, IL
Years Playing: 12.8
Courses Played: 55
Posts: 400
Quote:
Originally Posted by TalbotTrojan View Post
I know Mr. Sweeton has a ton on his plate and the amount of work he has to do for an update is significant. I think you make a valid point as to fairness at tournaments and newer players taking advantage of the situation. I wonder if there is another solution to this problem?
I don't disagree that Big Dog has plenty to do. My point is that the process should be looked at to see if it can be improved. There is a lot of manual work that goes into producing the ratings update and automating more of that should improve the turn around and lessen his load.
Reply With Quote
  #35  
Old 02-03-2014, 08:02 AM
JC17393 JC17393 is offline
Double Eagle Member
 
Join Date: Aug 2012
Years Playing: 16.3
Courses Played: 112
Throwing Style: LHBH
Posts: 1,672
Quote:
Originally Posted by krupicka View Post
I don't disagree that Big Dog has plenty to do. My point is that the process should be looked at to see if it can be improved. There is a lot of manual work that goes into producing the ratings update and automating more of that should improve the turn around and lessen his load.
I agree that more automating could probably make things move more quickly. But the root cause of most of the manual work is the number of errors made by TDs in filling out the reports (or even submitting at all). No amount of automation is going to fix that.
Reply With Quote
  #36  
Old 02-03-2014, 11:09 AM
Karl Karl is offline
Birdie Member
 
Join Date: Jan 2009
Posts: 450
Quote:
Originally Posted by Nemmers View Post
From the PDGA Ratings FAQ:
Being overly-concerned with "ratings" made me dependent on that number. I became obsessed with it.
...
Have you found yourself obsessed with PDGA ratings? If so, why? Does it affect your enjoyment of the game?
Nemmers,

Coming from a chess background (I did too - back in the 70's-90's, last posted at 2032) I can see that, but what you have to be careful of is "playing for the number in and of itself". If you have a tournament and you don't "at least do your present rating" you'll be bummed. I know a couple players in NJ that are exactly this! They have WAY more potential (than they're showing) but "let the number get to them". Don't succumb!

Besides, the concept of 'certain horses for certain courses' is too prevalent in dg to really put a definitive yea or nay on any one person beating another person - as one may just 'fit' that course (and win) even though they're lesser rated at the time.
It would be like a VERY positional chess player playing in a theme tournament where you HAVE to play the King's Gambit Accepted...probably not all peaches and cream for said player!

Trust your instincts.

Karl
Reply With Quote
  #37  
Old 02-03-2014, 11:31 AM
mullethead326 mullethead326 is offline
Par Member
 
Join Date: Jun 2011
Years Playing: 1.6
Courses Played: 83
Posts: 128
One example of how ratings could be improved is for players in their first year or two of tournaments--the guys who rapidly jump from rec to intermediate to advanced, some to open. Very few disc golfers begin playing tournaments as fully-developed throwers; the first year of playing is initially frustrating, then a series of small breakthroughs and epiphanies, occasional plateaus, and in general rapid improvements. Given enough tournaments (not sure what constitutes an adequate sample size), these improving players often exhibit what I call a "ratings lag", in which it takes some time for rounds early in one's career to drop off (by time or 2.5σ) or be neutralized by rounds that more accurately reflect a player's current ability.

http://www.pdga.com/player/58320/details

This is me. Therefore, this person has no problem being used as an example. I have included unofficial results from Throw Down The Mountain and the IDGC Ice Bowl as well.

I started playing in September 2012, and my first PDGA tournament was NCDGC in 2013. Now, I think the NCDGC results are a little spurious, because only about 60% of the field are propagators, some with only the minimum number of rounds, and many rapidly improving. After that tournament, my rating was 978, but I think I played up and was really more like 960-965.

These are the averages (no double-counting, no accounting for number of holes if >13) of my round ratings, from a starting point to the present:
Since April 2013: 995.5 (48 rds)
Since June 2013: 1000.9 (36 rds)
Since August 2013: 1005 (24 rds)
Since October 2013: 1023.8 (11 rds)

Even with double-weighting recent tournaments, I don't think tournaments from last April are really that representative of how I'm playing now. Since October is probably cherry-picking a little bit, as I've played better than I expected at three of those tournaments and 11 isn't much of a sample size. But even so, my rating is 998 right now (about 1000-1001 if ratings updated today), and there's a whole bunch of rounds from early in my history that will still be counted for another 4-6 months.

There's a residual effect from ratings lag, as well. When I, as well as other players exhibiting the same effect (I can think of 4-5 others in the Atlanta area who are legitimately playing 10 or more points above their ratings), enter a tournament, it impacts how round ratings for everyone shake out--not much, maybe 3-4 points maximum.

Now, if Chuck or a PDGA ratings guru is reading this, I can understand why he would be hesitant to implement any tweaks to this system, because there's way more to the algorithm than some dudes who have a lot of time to practice their new hobby--counterexamples would be folks who just got a new job or family and don't have time to practice, or guys playing through an injury. Those players want their older rounds to dilute their more recent performances--and double-counting the most recent 25% of rounds accounts for current performance in some respects. But anyway, there's a tactile example of how the ratings system falls through for a particular type of player. It should work out as my play stabilizes within the next six months or so.
Reply With Quote
  #38  
Old 02-03-2014, 12:29 PM
Cgkdisc Cgkdisc is offline
.:Hall of Fame Member:.
 
Join Date: Jul 2009
Location: Twin Cities
Years Playing: 25.1
Courses Played: 569
Throwing Style: RHBH
Posts: 7,282
It may sound sacrilegious but the lags in the ratings system are deliberate choices to support the reasons the system was created. Even if we had instantaneous ratings updates after everyone's rounds, those values would not be any better for the purposes of the system. The primary reason for the system (improving on the old local bump rules where they had them) is to get amateur players in their appropriate division or at least not playing below it. Just like most sports, teams/players play within a division for a whole season even if it becomes clear a player or team has moved a level ahead of everyone during that season.

Likewise, none of us should be upset when a "natural" quickly improves during our tournament season and clearly has surpassed the level he or she is playing but is part of a series and stays at that level, at least for the season. Many of those players play up anyway including those who shouldn't be playing up at least from a competitive chance standpoint. So the lags in the current system help delay to some extent the bagging cries when someone has a few more events in Intermediate to complete their series when they are clearly playing at an Advanced level.

Statistical systems are based on having enough data to increase accuracy. The reality is that the average PDGA player only posts a little over one rated round per month on average. I'm not in favor of the current once a month update for that reason. But player pressure has prompted the PDGA to increase the number of updates per year to once per month. The increase in number of events has also driven the need to do updates more frequently to keep the number of events in each update more manageable.

The 25% double weighting of most recent rounds may truly impact a small percentage of fast improving players but it sounds like a logical thing to do to boost their ratings faster. For most players, there's enough chatter (up and down movement) in their round ratings that it's potluck whether your doubled rounds average higher or lower than the average of your older rounds at any given update.

Less frequent updates would reduce that chatter. That's essentially what we do with ratings for World Rankings (no double weighting) and doing updates less frequently than regular ratings updates. And at yearend, we don't value more recent events higher than the oldest ones in that year, since for pros, their whole season (one year) matters to determine pecking orders.
Reply With Quote
 
  

  #39  
Old 02-03-2014, 12:35 PM
mullethead326 mullethead326 is offline
Par Member
 
Join Date: Jun 2011
Years Playing: 1.6
Courses Played: 83
Posts: 128
Thanks for the explanation, Chuck. That all makes sense.
Reply With Quote
Reply

Bookmarks

Thread Tools Search this Thread
Search this Thread:

Advanced Search

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -4. The time now is 07:18 AM.


Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2014, vBulletin Solutions, Inc.