Player ratings are just a representation of your past. They can't tell you how you will play tomorrow. My rating is 971 and I just beat a 999 rated player in a playoff. According to ratings that shouldn't have happened.

basically what Chris said. I'm def. playing 10 or so points better than my rating right now, but for the rounds that were rated, my rating is close to how I played.

If ratings were updated, say once a month (i know that it is a ton of work to do) it would help accuracy.

Another big thing that a lot of people don't realize is most of the ratings faults can be blamed on lazy TD's. Let's say that you play a tournament in October and man you just chit the bed. Nothing's going right, you shoot like 20-30 points below your average. Then you play a tournament in Feb and just kill it. You shoot 20-30 points above your rating.

Now, that Feb. tournament should go in your ratings probably in the early summer report, and the Oct. should be in the year end report. But the TD from the oct. tourney doesn't turn his stuff in to the PDGA till Feb, the same time as the other TD from the Feb. tournament does. No, this tournament that you played 4 months before will be rated later than it should have and it will cancel out your tournament from Feb. making it seem like you have no improvment b/c your rating stayed the same.

Just something to think about....

...My rating is 971 and I just beat a 999 rated player in a playoff.


I KEEL YOU!!!!!!

very much lag in the updates....they just tell me that a couple months ago, i was playing such a way. I think that ratings work for the every once in a while tourney player; but not us. More updates!!!!!

My rating says 932... there are tournaments from 2004 still on there that will be gone once I play an event or 2 before the may update. If you take the average of my events since june, I am closer to being over 950 than I am 930...

I think with any system the pdga comes up with, its still going to rely on a lot of past performances and will not accurately show rapid improvements in some players. I will say though that with the new system they have the ratings will be "closer" to being accurate and your play from tournament to tournament will have a greater effect on your ratings.

For me a rating is just a number, its something that I can look at to see where the pdga has me amongst my competitors, but it is not written in stone. Beyond that I could care less where my rating is, I just want to win going into every tournament I play.

Vince, I didn't say anything about it being your home course and the place where you learned to play disc golf. My point was about ratings.

My rating is slightly lower than my play but that that happens with most improving players.

My rating is slightly lower than my play but that that happens with most improving players.



Too bad you aren't one of those improving players.

Mine seems accurate (935), still waiting for that bump up from a 11/05/05 turnie though

Take a rating for what it is...A Statistic. Mark Twain said, "There are lies, there are **** lies and then there are statistics." Your rating should not govern how well you play or how well you think you should play. Play to the best of your abilities ALL THE TIME. Like any human you will have good and bad days. Just hope all your bad days are not during rated rounds! And remember "Practice doesn't make perfect...perfect practice makes perfect.

Another big thing that a lot of people don't realize is most of the ratings faults can be blamed on lazy <font color="blue">(or slow)</font> TD's...



<font color="blue">Wanted to requote and add to that--it is overlooked too often and blamed on the HQ.</font>

Ratings are for lack of a better term your AVERAGE ability compared to others AVERAGE ability based upon what you have done in the past . It has the possibility to suggest what you are going to do on that given day , BUT it is not a steadfast rule if you are higher rated you will win .
If a player who has a 999 rating plays ok and a player with a 971 rating plays great , YEP you are going to beat him .
I absolutly love the player rating idea and all its variants ( handicapping and various other ideas that can come from that )
The only down side is the lag in ratings update change but that will improve with time I feel .

In conclusion the rating system is a great tool to use for grouping , handicapping , and the like ( much better then some arbitrary format of I play this for this reason ) , but to use it as a stead fast this 980 player should beat this 955 player every time , NO that is not good . You still must play the game .
Do I feel my ratings are accurate , they might be alittle low ( since I have been moving up in the disc golf world ) and have been medium low ( cause I did not play many PDGAs in the past ) but all in all yea it is accurate enough .

No, I do not think PRs are accurate. Mine is usually lagging pretty badly. Especially when there are a couple of tournies not reported on time like this update for me. I was playing 968(what I'm rated now) golf back in Aug '05 at AM worlds when I was only rated 959. Another instance where they are not accurate is when a player has less than 25-30 rounds. I dont think you can get an accurate rating unless you have at least this many rounds on a few different courses. Some people's rating is inflated because of only playing a few tournaments on their home courses and not traveling. For the most part, I dont even pay attention to the ratings because I know there are people rated lower than me that are better than me, and there are people rated higher that I can beat consistantly.

how many times do I have to say this.
Ratings are CRAP

I think my individual round rating are very close to accurate. After every round my buddy and I have a bet on how close we can guess the round ratings, based off of top level consistent golfers at the same course/round.

For instance the best round of my life came in a tournament where I creamed everyone in my division and had the 3rd/4th best round of everyone including open players. I said it had to be rated over 1000, it ended up being 1004 and my highest rated round yet.

Overall I think my rating is behind, but that is because I have a lot of old events and bad rounds in there. Come May, most should be gone, and the update before Worlds should clear all bad rounds out except for the occasional 1 round blips. I realize that everyone will have bad rounds, but for the most part, my average and good rounds are dominate and I feel that I am realistically playing at a 950-955 level because I am improving faster then the updates can keep up with.

But it is just a number, the only thing that matters is how you do on the course compared to everyone else.

My rating is accurate . . . . just play 30+ pro tournaments a year and then you don't have to worry about having them updated all the time, just play consistant :p

I was playing 968(what I'm rated now) golf back in Aug '05 at AM worlds when I was only rated 959.



How can you say that this is not accurate??!! This is off only 1 stroke per round. How much closer does it need to be for you to consider it accurate?

How can you say that this is not accurate??!! This is off only 1 stroke per round. How much closer does it need to be for you to consider it accurate?



My point was

Mine is usually lagging pretty badly.



That one stroke per round turned out to be 8.5 strokes over worlds. And before that, my rating was WAY behind. I am only one case though Dave. Other peoples ratings are way off because of the aforementioned reasons:


Another instance where they are not accurate is when a player has less than 25-30 rounds.




Some people's rating is inflated because of only playing a few tournaments on their home courses
and not traveling.

...My rating is 971 and I just beat a 999 rated player in a playoff.


I KEEL YOU!!!!!!



Hey !!!!! isn't that Scott Stokley on your avitar ?

Take a rating for what it is...A Statistic. Mark Twain said, "There are lies, there are **** lies and then there are statistics." Your rating should not govern how well you play or how well you think you should play. Play to the best of your abilities ALL THE TIME. Like any human you will have good and bad days. Just hope all your bad days are not during rated rounds! And remember "Practice doesn't make perfect...perfect practice makes perfect.



I'm getting all misty eyed...

I think my Rating is fairly accurate but it won't be on the 14th as I have been playing 1000+ golf for the past 5 events but they aren't going in. Oh yeah but the one from 05' that I played poorly at will go in. A end of year rating update is useless unless the Touring Pro cards are based off of it but they aren't so why not just act as if the Feb update is a regular? Just include everything that gets turned in prior to 2/1. /msgboard/images/graemlins/smirk.gif

My opinion is that the PDGA should PAY someone for doing this work and making sure it is updated at least monthly.
Chuck and crew are doing a great job for how much they are getting paid ( which I think is nothing).
Like most things in life you get what you pay for, in this case we are getting much more but it's obvious we want more, as in, more frequent updates and ones that are more accurate.
This person in charge of Ratings could also shake down delinguent TD's for overdue results, making sure the ratings are accurate.
:o

I think the updates every three months is ok but that is not what is really happening. They have almost 5 months worth of events going into the April update and that is TOO long. They just need to do a March, June, Sept, and december update(Dec would be used for Tour cards). The March to June is only two months but has a lot of rounds entered. If you don like that you could do March, July, sept/oct, and december updates.


5 MONTHS IS TOOOOOOO LONG.

The mid-April update will have less than three months of tournaments from the lightest period in the year. June will have two more months. Then, September three months which is the heavy update with Worlds and Mid-Nats. December is three lighter months and Feb is the annual wrap-up.

what is the point of the Feb is my question? I mean the only possible reason would be for tour cards right? They use the Dec ratings though. :confused:

There have been about 75 events reported since the December update, a little over 10% of events. It wraps up the year giving someone a final rating just like final points. For most members in the country, this rating will be their first one that means anything because they won't be playing their first event until April. Sandalman might be able to check this and see how many different members play an event before April 1 when 25% of the year is over.

All of TX, OK, CA, FL, NC, SC, GA, WV, and a few more.

MI, HI, OH, AL, Australia, AZ, AR, KS, VA, WA, IA, NM, WI, KY, and TN all by April.

If it's only 1000 members, that's only 10% of the PDGA in three months. But we'll see how it comes up. Maybe we'll be surprised. Just because 10 Minnesota members escape to the Memorial in March doesn't mean the other 240 Minnesotans play a sanctioned event by April which is normally our start.

So the %10 minority doesn't matter? Also the three biggest states have events ALL year and also have the MOST players so why wouldn't it be more than %10?

They do matter and that's why we moved up the ratings from May to April :)

ratings are incredibly accurate. (sorry, mr ware.) if you take a large field and look at where the ratings predicted they would finish, then look at the actual results you will be amazed at how right-on they are. i've done it with several large events and found that if i was willing to accept a +/-10% in terms of placement in the field, the ratings were like 80% correct in predicting the finish. that is the definition of accurate.

That is true for all but the newer/younger players who are getting better. Zboaz for ezample.

Pat, are you able to tally how many different members played an event between Jan 1 and Mar 31 of 2005?

That is true for all but the newer/younger players who are getting better. Zboaz for ezample.



I feel that round ratings are really accurate. But read the line below.

Pat, are you able to tally how many different members played an event between Jan 1 and Mar 31 of 2005?



From the 05' schedule it looks like A LOT.

Pat, are you able to tally how many different members played an event between Jan 1 and Mar 31 of 2005?

thats all ya want? a count of how many different players competed in a sanctioned event during a specified date range? yeah, thats simple... lemme know,

That and maybe the total number of rounds played in that period. I think Roger indicated we rate around 130-140 thousand rounds in a year.

number of distinct players who played sanctioned event between 1-1-05 and 3-30-05 = 3,359

Out of what like 10,000 current members? Sounds like a lot to me.

here ya go:

Reg'd Players: 3359

Round 1: 3345
Round 2: 3282
Round 3: 2028
Round 4: 1032

Total Rounds (rated) during the period: 9687

999 rounds were not counted in totals

heres a quick take at 2005 rounds:

<table border="1"><tr><td>Round 1</td><td>30083
</td></tr><tr><td>Round 2</td><td>29658
</td></tr><tr><td>Round 3</td><td>17701
</td></tr><tr><td>Round 4</td><td>10157
</td></tr><tr><td>Round 5</td><td>1009
</td></tr><tr><td>Round 6</td><td>699
</td></tr><tr><td>Round 7</td><td>599
</td></tr><tr><td>Round 8</td><td>194
</td></tr><tr><td>Round 9</td><td>0
</td></tr><tr><td>Round 10</td><td>0
</td></tr><tr><td>Semis</td><td>637
</td></tr><tr><td>Finals</td><td>1234
</td></tr><tr><td>Total</td><td>91971
</td></tr><tr><td></tr></td></table>

thats a quick calc based on my current understanding of the table structure for rated rounds.

Thanks, Pat. Looks like less than 15% of the rounds in 2005 got played in the first three months. Since that's 25% of a year, it looks like we don't need to move it up any earlier than a mid-April ratings update if we're going to spread them out over the year. Maybe 2500 of those 3359 are or will be members, so that means about 25% in action during the first 3 months, which is pretty good.

3358 of the 3359 players had pdga nums. all of the pdganum = null (nonmembers) would have rolled up into a single "player" since i selected for distinct pdganums

Gotcha

Still for those people who play all year long it is unfair to have to wait so long for an update. :confused:

I also wouldn't be suprised if the Oct-Dec period is even slower yet they get a speedy update.

The other update periods are all bigger than mid-April because for mid-Dec we're still getting tardy events coming in from May thru Aug. And mid-Dec also gets all of September thru mid-Nov which has a lot of activity. I think even the final cleanup update coming in mid-Feb will have as many rounds as the first three months of the year because all of the events in the first three months that Pat tallied either can't or won't get reported by the first week of April for us to process them.

My rating is too high. Probably 20 points too high. New England has zero 1000 rated players. Before I went to the USDGC, my rating was 960. Now, after the USDGC and two other events up here in New England, my rating is 980. Trust me, I did not practice that much. It seems that there is still a bump for playing in tourneys with lots of good players.

What New England needs is for the Frizzaks to stay in Carolina until they break the 1000 barrier, then move back to New England so we can kick their butts some more and get our ratings above 1000.

Then they can leave town. Again. With their tails tucked. :D

I have felt for the most part that Player ratings amongst most improving players, are within +-20 points or 2 strokes a round. So for the most part I would think they are accurate enough

It seems that there is still a bump for playing in tourneys with lots of good players.



While it's true, the effect isn't as big as some seem to think. If two pools of players play the same course at the same time and one pool averages 100 points higher (say 975 versus 875), the higher pool will generate an SSA about 1 shot higher. That gives those players ratings about 10 points higher. We've now included this 1 throw per 100 rating points adjustment into the formula end of 2005. So New England and Oregon might start to see a boost which are places where this effect seemed to be noticed.

i'll second that. the "better-pool" bias is mostly gone. the calculation tweaks that started last year eliminated a large part of that.

I feel that round ratings are really accurate.




I will also add that the player ratings ARE accurate, they just lag behind imporving players.

Case in point http://www.pdga.com/tournament/tournament_results.php?TournID=5723&year=2005&includeRatings=1#Open My average round rating for the event was 956. While I did "leave a lot out there" it was no more then the typical advanced player. It was also 35 degrees and 25mph winds.

So I change my opinon. Player ratings are very accurate of your level of play if you have reached a plateau in skills. Granted this is nothing new or profound. I am just recanting my prior opinion on player ratings.

this is true by tautology for any measurement of a trend.

to solve it, we could use only the last eight rounds or something like that but can you imagine the hades that would break loose over that change? :D

this is true by tautology for any measurement of a trend.

to solve it, we could use only the last eight rounds or something like that but can you imagine the hades that would break loose over that change? :D



I'd like to see what would happen if a calc was run with the SD only being calculated based on double rated rounds. This would, I believe, help to cause player ratings to catch up more quickly to improving players, while consistant players would stay nice and level. This would really only benefit on a short term calc, so perhaps having 2 calculations listed, one that is the current calc method, and is shown as the official rating, and another that updates based on unofficial results and updates using the alternate method and is labeled Current Trend or something else like that.

improving players would be over rated. declining players would be under rated. i've thought that a cool thing to do would be to publish each players SD along with their rating. almost always the player with the lowest SD has the most "accurate" rating.

improving players would be over rated. declining players would be under rated. i've thought that a cool thing to do would be to publish each players SD along with their rating. almost always the player with the lowest SD has the most "accurate" rating.



You mean, consistant shooters have a more consistant average? ;-) Yeah, I understand that flaw in this possible approach, that's why I thought calling it 'current trend' or similar so you can get an idea of how your recent sets of rounds average out without having to do all the math.

That is a good idea in my opinion. If the SD was based on the most recent rounds, that would take into the current consistency of players. I would argue that the consistency of an advanced player is greater then an Intermediate player, and also that the consistency of a pro is better then an Advanced.

Rounds that I had a year ago really do not affect how I am playing now. While they will always be a part of my history, they are not a real measure of my recent (Dec 2005, update) performance.

I think keeping the SD the most current is a terrific idea.

The average rate for people with increasing ratings is not as much as those who are improving think it is from a statistical standpoint. People tend to remember their good rounds as proof of improvement and forget their bad ones as anomalies. One thing we've thought about is actually calculating the trend line for a player that has a minimum number of rounds in their past 12 months, say 16. Then, the weighting factor for the top 25% would be increased more the steeper the slope, perhaps varying from 1.5 to 4 times.

I just did a little look at my ratings in more detail to see how closely my current rating is.

Out of tournaments from 2004 still when I was not improving fast and not practicing (2 tournaments) my year end rating was 901 and both tournaments averaged below that at 884.

In 2005, early on in the season, first tournament was 927 average, the next two I played with injurys and averaged about 911.

My year end rating is 932, my remaining tournament averages from May to August are in order as follows:
934,962,947,973,907,945,953,953

The 907 was a week where I played with an elbow injury as well as dehydration. If you throw that one tournament out and average the others, it would be 952 average, or 20 points above my current rating. This probably means absolutely nothing to anyone, even I could probably care less. But I always felt ratings were + or - 20 point range at any given point and my last year or so supports that in my case.

The truth is, I am really bored and had nothing better to do with my time :)

I am over rated :D

My SD would be bigger than a lot of other players. I have sub 900 rounds and over 1000 rated round :confused:

The average rate for people with increasing ratings is not as much as those who are improving think it is from a statistical standpoint. People tend to remember their good rounds as proof of improvement and forget their bad ones as anomalies. One thing we've thought about is actually calculating the trend line for a player that has a minimum number of rounds in their past 12 months, say 16. Then, the weighting factor for the top 25% would be increased more the steeper the slope, perhaps varying from 1.5 to 4 times.



Ok, I will keep my mouth shut.....

"Do you feel your rating is accurate?"


No, I am not that good.

I have an idea, how about if we do away with all this controversial crap and say that your rating is based ONLY on your last rated round. We will call this the:

"Stop worrying about your rating, you self-absorbed disc golfers" Player Rating

According to the new SWAYRYSADGPR system I am a 903 rated golfer.

See, wasn't that nice, no discussion, no controversy, easy as apply pie!


please notice that I am joking and love all you guys, despite your mirrored view. :)

I think they're accurate. And they make me appreciate how great the 1000-rated players are. A few times I shot 1000-rated rounds and, man, that good ol' majic was there where the disc -- get this -- was going where it was supposed to.

45 Elements! 52 Pyramids Gold! Those, my friend, are 1020-rated rounds under ANY conditions. :D

From Chuck Kennedy on the 'Final 2005 Yearend Ratings Posted' thread:
A surprise entry in the top 10 is Ben Gaddis rated 1029, a new master from SC. He threw down a -14 in November at the Upstate Classic that rated 1086!



A guy throws a 40 on a course where the SSA is 47.5 and gets a 1086 rated round. That, along with scores from just one other tournament, gets him a 1029 rating, making him the 7th-best disc golfer in the world.

nation, I like that.
My SWAYRYSADGPR rating should be around 830.

That, along with scores from just one other tournament, gets him a 1029 rating, making him the 7th-best disc golfer in the world.




He's the player with the 7th best rating, but who's not yet a propagator. That's quite a bit different. The World Rankings will be coming soon based on this final update. People will see a legit ranking of the top players that incorporates results where they actually face each other in competition. Ben won't be on that list yet.

I can't wait till someone comes up with a real handicapp system...the rating system we have now is terrible.

Any rating system is just a mirror with a memory. Some don't like what's reflected back...

Not exactly true!! Especially when the mirror is altered to reflect distorted images, or numbers in the case of the ratings!! If it was merely a mirror , why is it changed so **** much!! Why is it that when there is an error it is not corrected on the update in error , rather on the next update?? Thus meaning there will indefinately be in-accurate data on past ratings!!

Why do we regularly change rules in disc golf or any sport? To make them better. Our life would be much easier if we didn't change it. But when we discover issues to improve the structure, then we move to do it.

As far as retro corrections, that's starting this year with the May correction update for the mid-April regular ratings update. If there are any corrections, we won't call it a May update but continue using the mid-April date because no new events will be in it. From a PDGA competition standpoint, once an update is past, there's no reason to expend the extra effort to rerun and correct the previous update, just correct the actual data which we do. All that counts are the current ratings.

I think the ratings system is accurate based off the rounds put in but I have been shooting 1000 golf lately and still go down 1 point. Why? Because they don't enter any rounds in 06 so the only event that was entered in my ratong was a bad one. :confused: :D:D

Again the answer is for the pdga to hire someone( like Chuck) to do the job instead of it being done by a volunteer

Roger and I are now getting paid an honorarium and are under contract to do processing like we have been doing it thru 2008. We have two other helpers who can help when needed who get paid per event for processing reports at crunch time.

Thanks for accepting the job Chuck, keep up the good work. Have you been keeping any stats on what the average last place cash's rated rnds are? Or winning rated rnds? I bet most golfers would find that information useful, especially when choosing a field to compete in.

Whenever we've checked A-tiers, you have to shoot around 1000 rating to cash. For lower tiers, I'm not sure what the cashing level would tell us because the lower tiers you go, the more varied the mix of ratings and average ratings for the divisions becomes due to regional variations. So it becomes more apples to oranges for comparisons. You probably realize how tough it is to cash in most divisions in NC compared some other areas. At least for A-tiers and even more so for NTs and Majors, the pool of players and their ratings are more similar and consistent.

Would figuring B & C tiers for each state be too time consuming? That is where the most attendance is...

The information is available by looking at event reports online. Pros typically look at the payouts for the previous year(s) for events they might attend. The round ratings and course ratings are all there and how far down they paid. We don't have payout info for Ams online but you can count down almost halfway to figure out what the last mercher probably shot. Actual stats like this for events in your area or wherever you want to travel are better indicators than any data consolidation we would do.