Revamped personal rating system

[runnaman]

I have revisited my rating system and decided to have it be more like the PDGA's. So for this one, all you have to do is enter the information in the green cells.

Obviously, it works better with the more people you enter, but it's fairly accurate on the tests I have done.

Let me know what you think.

Thanks

[runnaman]

Updated:

I added another step which calculated the difference in what a player shoots compared to the average of the field.

I found that this produced slightly better ratings.

I also left in some numbers from a practice tournament I was doing so you can kind of see how things work.


Again, if you have any questions or comments, feel free to express them here.

I want to know what people think.

[Kenny B]

I'm curious to know how you came up with the 2.1 number from the following:

Calculated points per stroke=(1000/(D19*2.1))
where D19 was the average score

[runnaman]

that, and the 2 in =F19+(C22/2) are the only random numbers, but they're in because they're the numbers that worked.

[Kenny B]

Originally Posted by runnaman

that, and the 2 in =F19+(C22/2) are the only random numbers, but they're in because they're the numbers that worked.

I see. Thanks for pointing out the other random number. I had not seen that one before.

It appears in the equation "Projected rating for a score of 51=F19+(C22/2)" that you divided the ratings points per stroke in half in order to add that half stroke to the average score. This makes some sense and is becuase the average score was about 50.5, which no one can actually shoot. But what happens when I repopulate the scores and player ratings for a different round and the average score is 50.0? I think at that point, the equation "rating for score of 51=F19+(C22/2)" will have an error by adding only half of cell C22 (rating points per stroke).

For example, I can change Gagne's score to 53. Doing so changes the average score to 51 exactly. The average rating is still 1018.25. So a score of 51 should = a rating of 1018. However, your spreadsheet is still adding that half stroke to the projected rating for 51 and giving a rating of 1024.

[runnaman]

Edit: Now I understand what you're saying.

The reason I add half a stroke is because it assumes that if 1 person were to play a round, they would shoot half a stroke better than there rating. It's my attempt to account for people getting better.

Hope that makes sense.

I understand what you're saying though and I'll work on that. If you have any ideas, let me know.

[runnaman]

It appears I may be looking at this all wrong. The difficulty in calculating scores isn't really figuring out what a 1000 rated round is, it's all about finding how many points each stroke is worth.

I tweaked it a little more and re-uploaded.

You must use 15 scores though or else you get a DIV/#0 error. (you could just delete that cell, but it's more accurate with the more players anyway).

[runnaman]

Uploaded a new update. I had to change a formula because the points per stroke cell wasn't correlating with the players ratings.

[runnaman]

Once again, there is another update available. This allows you to use as many or few scores as you like and calculates ratings base on the half score to make it more accurate.

(Thank you Kenny B for pointing that out)

Download is available in the first post.