Jump to content
Sign in to follow this  
1032004

How is RPI determined?

Recommended Posts

https://drexeldragons.com/services/download_file.ashx?file_location=https://s3.amazonaws.com/drexeldragons.com/documents/2019/3/14/2019_Final_RPI_Ranking_Mar_13.pdf

 

I think this might help explain some of the controversial seeds like Derek White #1, but some of these are wacky even where it didn’t impact seeding - Ashnault #5?  Nolf #2? Marinelli #2?

Edited by 1032004

Share this post


Link to post
Share on other sites
2 hours ago, AlexSteenTOM said:

(Win% * 0.25) + (OppWin% * 0.50) + (OppOppWin% * 0.25)

I've never understood why the RPI is supposedly a measure of your strength of schedule.  If you wrestle a bunch of teams from the EWL, and they in turn wrestle guys from the EWL, all of their win %'s will be inflated vs teams from the B1G.  Just because you and your opponents have a high win %, doesn't mean they were top notch competition. No disrespect to the EWL, but they don't see the same caliber of wrestlers week in and week out as the B1G does, for example.  IIRC, the first year they switched to the new qualification system, the ACC almost doubled their allocations, yet the number of AA's remained virtually unchanged.  I seem to recall remarking at the time that the new system definitely skewed towards spreading more allocations to the non-traditional powers.  RPI was clearly a factor in subsidizing the qualifications of wrestlers from weaker conferences.  Note: That was before the resurgence of UNC, NCSt, VT, etc.

Share this post


Link to post
Share on other sites
1 hour ago, tigerfan said:

I've never understood why the RPI is supposedly a measure of your strength of schedule.  If you wrestle a bunch of teams from the EWL, and they in turn wrestle guys from the EWL, all of their win %'s will be inflated vs teams from the B1G.  Just because you and your opponents have a high win %, doesn't mean they were top notch competition. No disrespect to the EWL, but they don't see the same caliber of wrestlers week in and week out as the B1G does, for example.  IIRC, the first year they switched to the new qualification system, the ACC almost doubled their allocations, yet the number of AA's remained virtually unchanged.  I seem to recall remarking at the time that the new system definitely skewed towards spreading more allocations to the non-traditional powers.  RPI was clearly a factor in subsidizing the qualifications of wrestlers from weaker conferences.  Note: That was before the resurgence of UNC, NCSt, VT, etc.

The ACC was actually pretty strong at the time, to the point where they probably deserved roughly double the number of national qualifiers they were getting...but all of those guys were in the 13-20+ range, not the top 12 range, so they weren't getting those additional qualifiers under the old Onion System.

Also, remember, the RPI is only one criterion, of which each wrestler had to have two of.

Share this post


Link to post
Share on other sites
3 hours ago, tigerfan said:

I've never understood why the RPI is supposedly a measure of your strength of schedule.  If you wrestle a bunch of teams from the EWL, and they in turn wrestle guys from the EWL, all of their win %'s will be inflated vs teams from the B1G.  Just because you and your opponents have a high win %, doesn't mean they were top notch competition. No disrespect to the EWL, but they don't see the same caliber of wrestlers week in and week out as the B1G does, for example.  IIRC, the first year they switched to the new qualification system, the ACC almost doubled their allocations, yet the number of AA's remained virtually unchanged.  I seem to recall remarking at the time that the new system definitely skewed towards spreading more allocations to the non-traditional powers.  RPI was clearly a factor in subsidizing the qualifications of wrestlers from weaker conferences.  Note: That was before the resurgence of UNC, NCSt, VT, etc.

Also it doesn't reward wrestlers who compete in more matches and thus are more likely to have a statistically representative RPI. I seem to recall that wrestling requires a minimum # of matches to qualify for an RPI score, but it doesn't assign greater weight to those whose RPI is calculated based on a season of say, 25 matches rather than 12.

Share this post


Link to post
Share on other sites
18 hours ago, pamela said:

Also it doesn't reward wrestlers who compete in more matches and thus are more likely to have a statistically representative RPI. I seem to recall that wrestling requires a minimum # of matches to qualify for an RPI score, but it doesn't assign greater weight to those whose RPI is calculated based on a season of say, 25 matches rather than 12.

Well, the minimum number of matches is 17, so a wrestler with 12 matches wouldn't have an RPI at all.

Share this post


Link to post
Share on other sites

RPI is a terrible system.  50% is opponents W/L which means it thinks that it's harder to beat two middling wrestlers with say 12-10 records that it is to beat a guy with a 20-0 and a guy with a 3-20 record.  This really comes into play if the 20-0 guy beat both the middling wrestlers upping the opponents of opponents which is 25%.   RPI works better in a sport like basketball because almost no one goes undefeated, not in wrestling with standouts like Nolf and Nickal.  For wrestling, RPI should be 50% own record, 30% opponent and 20 % opponents opponents.

Share this post


Link to post
Share on other sites
22 hours ago, pamela said:

Also it doesn't reward wrestlers who compete in more matches and thus are more likely to have a statistically representative RPI. I seem to recall that wrestling requires a minimum # of matches to qualify for an RPI score, but it doesn't assign greater weight to those whose RPI is calculated based on a season of say, 25 matches rather than 12.

This is why nearly all the Lock Haven wrestlers have 40+ matches in a season. The chances of getting a solid RPI rank are worse than getting your win percentage up to qualify a spot. 

Share this post


Link to post
Share on other sites
4 hours ago, SetonHallPirate said:

Well, the minimum number of matches is 17, so a wrestler with 12 matches wouldn't have an RPI at all.

Thanks. Are wrestlers who have competed in more matches given stronger consideration when it comes to earning a good seed or even an at-large bid? It looks like there are some guys who either wrestled fewer than 17 matches or did just the bare minimum to qualify for RPI scoring (Micic, Tariq, Mueller, Carr, Milhof, Stoll, Weigel, Shak, for example). 

The top 4 seeds across all weights have wrestled an average about 25-26 matches this season while there are many lower-seeded guys who have portfolios of 28, 29 30+, 40+ matches. I get that injuries happen and it's unreasonable to expect today's wrestlers to log 40+ matches per season, but it would be nice to reward wrestlers who compete more - esp if RPI is just one part of the seeding calculation. 

Interestingly, the top seeds at 184 have wrestled an average of ~21 matches while the rest of the field has wrestled closer to 29-30. Seen in a certain light, it can give the impression that ducking and sitting out pays off. 

Share this post


Link to post
Share on other sites
1 minute ago, pamela said:

Thanks. Are wrestlers who have competed in more matches given stronger consideration when it comes to earning a good seed or even an at-large bid? It looks like there are some guys who either wrestled fewer than 17 matches or did just the bare minimum to qualify for RPI scoring (Micic, Tariq, Mueller, Carr, Milhof, Stoll, Weigel, Shak, for example). 

The top 4 seeds across all weights have wrestled an average about 25-26 matches this season while there are many lower-seeded guys who have portfolios of 28, 29 30+, 40+ matches. I get that injuries happen and it's unreasonable to expect today's wrestlers to log 40+ matches per season, but it would be nice to reward wrestlers who compete more - esp if RPI is just one part of the seeding calculation. 

Interestingly, the top seeds at 184 have wrestled an average of ~21 matches while the rest of the field has wrestled closer to 29-30. Seen in a certain light, it can give the impression that ducking and sitting out pays off. 

Only in terms of having more quality wins.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
Sign in to follow this  

×