PDGA rating calculations

[Darth Anovin]

I am wondering if anyone can provide me with a link or an explanation as to exactly how ratings are determined. I've searched this site and PDGA's site but the only thing I could find was an explanation on the way they determine what rounds are used. But nothing beyond that. Is it just simply an average of the rated rounds used?!? Thanks in advance.

 

[jeverett]

I am wondering if anyone can provide me with a link or an explanation as to exactly how ratings are determined. I've searched this site and PDGA's site but the only thing I could find was an explanation on the way they determine what rounds are used. But nothing beyond that. Is it just simply an average of the rated rounds used?!? Thanks in advance.

Are you meaning player ratings? Or the ratings generated from rounds themselves? Assuming you mean player ratings:

Your player rating is the weighted (e.g. an 27-hole round is worth 1.5x as much as a 18-hole round) average of all sanctioned rounds you have played for (at least) a calendar year. If you have fewer than 8 total rounds in a calendar year (calculated based on the date of your most recent round), the rating system will go farther back (up to two years) until it finds at least 8 total rounds.

Your most-recent 25% of rounds are double-weighted, and if any of your rounds are outside 100(?) points or one(?) standard-deviation of your rating, they are not included in your rating.

Is that what you were wanting to know?

 

[mashnut]

I believe rounds have to be 2.5 standard deviations below your rating to be thrown out.

 

[ejvogie]

Paging Dr. Kennedy, paging Dr. Kennedy, please answer the white courtesy phone.

 

[jeverett]

I believe rounds have to be 2.5 standard deviations below your rating to be thrown out.

Thanks, Mashnut.. I couldn't remember the exact criteria for excluding rounds from player ratings. As an aside, the DGCR player ratings don't drop any rounds.

 

[Horsman]

Take all your rounds and average them together. Count your 25% most recent rounds twice. Drop any number from that if it is 2.5 standard deviations away from the mean or 100 point below the mean. Recalculate. Should get you pretty darn close. Also only rounds from the last year get counted towards your rating unless you have less then 8 I believe so drop older rounds if need be.

 

[Darth Anovin]

Thanks for all the input. But if I only have two rated rounds in the past 1.5 years, will it go back for just my previous 6 rounds? Here's the reason I ask: I rated 894. I played my first event yesterday in over a year and a half. My first round I shot a 950 rated round. I was stoked right?!? Well, I tanked my second round and shot an 820 rated round. So I'm just trying to get an idea of what it will be at after this event is included.

 

[Horsman]

Thanks for all the input. But if I only have two rated rounds in the past 1.5 years, will it go back for just my previous 6 rounds? Here's the reason I ask: I rated 894. I played my first event yesterday in over a year and a half. My first round I shot a 950 rated round. I was stoked right?!? Well, I tanked my second round and shot an 820 rated round. So I'm just trying to get an idea of what it will be at after this event is included.

They will go back up to 24 months to get up to 8 (I think) rounds to count towards your rating so yes they should be included.

 

[jeverett]

Thanks for all the input. But if I only have two rated rounds in the past 1.5 years, will it go back for just my previous 6 rounds? Here's the reason I ask: I rated 894. I played my first event yesterday in over a year and a half. My first round I shot a 950 rated round. I was stoked right?!? Well, I tanked my second round and shot an 820 rated round. So I'm just trying to get an idea of what it will be at after this event is included.

Hi Darth,

The number of rounds it will find is pretty easy to determine: look at the date of your most recent recorded round. All rounds up to a year before that date will be included, definitely. If that's less than 8 total rounds, the system will go back even farther, up to two calendar years total, until it finds at least 8 total rounds (or hits the two year mark and stops).

 

[Darth Anovin]

But will it stop at 8 total rounds?

 

[jeverett]

But will it stop at 8 total rounds?

Hi Darth,

It will stop at one calendar year, *if* that year includes at least 8 total rounds. That could be a lot more than 8 rounds.

If that test fails, it will stop at either 8 total rounds, or two calendar years worth of rounds, whichever comes first. Does that help?

 

[JSpohr-WCDG1]

Just in case you guys are trying to figure out your next ratings I use my buddy's (Max Nichols) ratings calculator on his website. You just enter your round ratings and the date of the tourney and it figures it for you. Check it out http://www.mydgjourney.com/ratingCalc.html

 

[Darth Anovin]

Hi Darth,

It will stop at one calendar year, *if* that year includes at least 8 total rounds. That could be a lot more than 8 rounds.

If that test fails, it will stop at either 8 total rounds, or two calendar years worth of rounds, whichever comes first. Does that help?

Yes. Thank you. Exactly what I was looking for

 

[Darth Anovin]

Just in case you guys are trying to figure out your next ratings I use my buddy's (Max Nichols) ratings calculator on his website. You just enter your round ratings and the date of the tourney and it figures it for you. Check it out http://www.mydgjourney.com/ratingCalc.html

Your buddy's website is pretty kickass btw.

 

[Allurex]

I believe rounds have to be 2.5 standard deviations below your rating to be thrown out.

I've been wondering lately, is it 2.5 Standard Deviations below your current rating? Or that many below your rating at the time of the round?

Like if I was 930 rated in June, and shot a terrible 874 at that time, but now I'm 951 rated. Does that round need to be 2.5 Standard Deviations below my old or current rating?

 

[mashnut]

I've been wondering lately, is it 2.5 Standard Deviations below your current rating? Or that many below your rating at the time of the round?

Like if I was 930 rated in June, and shot a terrible 874 at that time, but now I'm 951 rated. Does that round need to be 2.5 Standard Deviations below my old or current rating?

I'm fairly certain it's current rating, and as someone mentioned above they run that calculation multiple times. You could have a bad round dropped which would raise your average and possibly make other rounds eligible to be dropped.

 

[Darth Anovin]

Ok, so after re-reading this thread, I have another question. What is considered a standard deviation? If I'm rated 894, would an 820 qualify?

 

[mashnut]

Ok, so after re-reading this thread, I have another question. What is considered a standard deviation? If I'm rated 894, would an 820 qualify?

Standard deviation is a statistical measure of how much data tends to deviate from an average. It depends on both your average and on how much variance there is in your other rounds so there's no way of saying whether that round would qualify without knowing all the other rounds included in your rating.

 

[BogeyNoMore]

IIR statistics correctly, say two players have the same/similar average ratings over the same # of rated rounds. One player has ratings that run the gamut from 820 - 950, while the other has ratings that are all pretty closely packed around 890.

The player whose round ratings are more consistent would have a smaller std deviation than the player whose scores are all over the place. Hence, the threshold for one of their rounds being dropped is tighter than the threshold would be for the less consistent player. So, the more consistent your rated rounds are, the easier it is to drop a bad round.

Essentially, rounds are more easily identifiable as statistically aberrant when the data is closely packed together. Spreading round ratings over a greater range means there's greater deviation, so it's tougher to say a low score is truly a statistical outlier.

I hope that helps for those unfamiliar with how std deviation works.

 

[mashnut]

IIR statistics correctly, say two players have the same/similar average ratings over the same # of rated rounds. One player has ratings that run the gamut from 820 - 950, while the other has ratings that are all pretty closely packed around 890.

The player whose round ratings are more consistent would have a smaller std deviation than the player whose scores are all over the place. Hence, the threshold for one of their rounds being dropped is tighter than the threshold would be for the less consistent player. So, the more consistent your rated rounds are, the easier it is to drop a bad round.

Essentially, rounds are more easily identifiable as statistically aberrant when the data is closely packed together. Spreading round ratings over a greater range means there's greater deviation, so it's tougher to say a low score is truly a statistical outlier.

I hope that helps for those unfamiliar with how std deviation works.

Yup, that's exactly how it works.