|
|
Post by willtalk on Oct 22, 2019 20:15:01 GMT -8
The SEC has 14 teams, the ACC 15, and the PAC 12 has, not unexpectedly, 12. If they get 8 and 7 teams in, that's 51.7% of the total; our 6 is, obviously, 50%. That's not much of a disparity.If our conference added 2 more teams like Gonzaga and Boise State, we'd get 8. If we added Montana and Idaho State, we'd still get 6. Plus, you have to objectively look at the bona fides of the 6 projected to not make it. Both Washington schools and Colorado are easy to eliminate. The other 3- Cal's been there regularly, Utah has added some talent, as has USC. If they build a resume over the course of the season, and still get left out, then we can have this conversation about unfairness. But if they do well in a fairly soft non-conference schedule (I don't know any of their schedules) with no big quality wins, then go 9 & 9 or 10 & 8 in conference, with maybe a win against an Arizona mixed in, do they deserve a slot? You gotta have the RPI, you gotta have the numbers. Let's review this again in March. The number of teams in a conference has never been a criteria used to decide how many bids they will get. There are plenty of conferences with a large number of teams who get only one or two representatives. The limit for the Pac 12 is supporsed to only would be set by the quality of the teams in the conference. I do feel that last season that is exactly not what happened. They decided on a limit and who went was not based on the teams strength of schedule and record but on which of the Pac 12 schools eliminated themselves by competing for there limited amount of slots. The number was decided before conference play even began. And that number certainly did not correspond to the teams ratings. Even though other conferences have more teams, they have fewer strong teams and more weak ones. If you look at the records last season, Tenn had more loses than any team that made the NCAA's. And many of those losses were to really bad teams. Yet they got in while more deserving teams from other conferences didn't. That included Arizona, which waltzed through the NIT. They were an example of a team who got cut off so that more teams from a less deserving but bigger conference could get more team in. This was validated by the success of the Pac 12 compared to the other conferences like the SEC and ACC. Their top couple of teams only did well. Most of those picked got got knocked out early. |
|
|
|
Post by baseba1111 on Oct 22, 2019 22:33:31 GMT -8
Lol... the "top couple of teams" from each conference are usually the teams that do well.
There is zero east coast bias. Excuses galore... the women's game is heavily top loaded (you could easily pick the best 16 and save 2 rounds... sort of like OSAA football playoffs) just look at who makes the Sweet 16 on a regular basis.
The RPI doesn't favor the population centers... population does. Larger population, more schools, more talent centralized in those schools.
The Pac12 gets its statistically equivalent share as pointed out above. And that's going by number of conference members which has never been a factor.
The results don't lie. Sending another Pac12 team or two to lose in the 1st round doesn't change the successes of the East/South (whatever boundary you want to use).
Larger population, more quality WBB teams drawing more talent. Pretty simple.
Funny... the bias crowd always rears their heads speaking of filling the brackets. Yet, which teams do they want OSU WBB to get games with? Outside the Pac12, Gonzaga, and ?? where are the most talented opponents? Yep... excitement reigns vs NC, Louisville, ND, FLSt, MissSt, etc. Funny...
|
|
|
|
Post by beavheart on Oct 23, 2019 9:33:03 GMT -8
Lol... the "top couple of teams" from each conference are usually the teams that do well. There is zero east coast bias. Excuses galore... the women's game is heavily top loaded (you could easily pick the best 16 and save 2 rounds... sort of like OSAA football playoffs) just look at who makes the Sweet 16 on a regular basis. The RPI doesn't favor the population centers... population does. Larger population, more schools, more talent centralized in those schools. The Pac12 gets its statistically equivalent share as pointed out above. And that's going by number of conference members which has never been a factor. The results don't lie. Sending another Pac12 team or two to lose in the 1st round doesn't change the successes of the East/South (whatever boundary you want to use). Larger population, more quality WBB teams drawing more talent. Pretty simple. Funny... the bias crowd always rears their heads speaking of filling the brackets. Yet, which teams do they want OSU WBB to get games with? Outside the Pac12, Gonzaga, and ?? where are the most talented opponents? Yep... excitement reigns vs NC, Louisville, ND, FLSt, MissSt, etc. Funny... For such a smart guy, I'm not sure why this is lost on you? The RPI ABSOLUTELY rewards the population centers for exactly the reason you just stated. When "your" SEC or ACC school is surrounded by a bunch of lower conference programs that live somewhere in the middle of the RPI because they are "good mid-majors", but that you can count on beating 9 out of 10 times then your RPI rating gets a big boost. When you are OSU and the next best program outside of the Pac12 anywhere close by is Gonzaga, (an 8 hour drive away), and then there are crickets after that in terms of a quality opponents within shouting distance then you have a very difficult time padding your RPI rating. This dichotomy is magnified at the conference level because lower level teams in other conferences have more opportunities to play and win games against "quality" opponents. I thought this was pretty universally understood. This happens in all sports. Again, the RPI does the Pac12 NO favors. The RPI basically just mirrors the population centers. The RPI was created in part to remove regional bias in the rankings. It does a terrible job in that regard. The SEC had as bad of a year in WBB last year as they have had in a long time. They still sent half of their conference to the post season. In the Pac12 we have to have a banner year, like last year, to send half our conference to the post season. If the conference doesn't have a good year, we are lucky to get 3 or 4 teams in. You might be just fine with that discrepancy, but I'm not. |
|
|
|
Post by baseba1111 on Oct 23, 2019 10:10:10 GMT -8
Lol... the "top couple of teams" from each conference are usually the teams that do well. There is zero east coast bias. Excuses galore... the women's game is heavily top loaded (you could easily pick the best 16 and save 2 rounds... sort of like OSAA football playoffs) just look at who makes the Sweet 16 on a regular basis. The RPI doesn't favor the population centers... population does. Larger population, more schools, more talent centralized in those schools. The Pac12 gets its statistically equivalent share as pointed out above. And that's going by number of conference members which has never been a factor. The results don't lie. Sending another Pac12 team or two to lose in the 1st round doesn't change the successes of the East/South (whatever boundary you want to use). Larger population, more quality WBB teams drawing more talent. Pretty simple. Funny... the bias crowd always rears their heads speaking of filling the brackets. Yet, which teams do they want OSU WBB to get games with? Outside the Pac12, Gonzaga, and ?? where are the most talented opponents? Yep... excitement reigns vs NC, Louisville, ND, FLSt, MissSt, etc. Funny... For such a smart guy, I'm not sure why this is lost on you? The RPI ABSOLUTELY rewards the population centers for exactly the reason you just stated. When "your" SEC or ACC school is surrounded by a bunch of lower conference programs that live somewhere in the middle of the RPI because they are "good mid-majors", but that you can count on beating 9 out of 10 times then your RPI rating gets a big boost. When you are OSU and the next best program outside of the Pac12 anywhere close by is Gonzaga, (an 8 hour drive away), and then there are crickets after that in terms of a quality opponents within shouting distance then you have a very difficult time padding your RPI rating. This dichotomy is magnified at the conference level because lower level teams in other conferences have more opportunities to play and win games against "quality" opponents. I thought this was pretty universally understood. This happens in all sports. Again, the RPI does the Pac12 NO favors. The RPI basically just mirrors the population centers. The RPI was created in part to remove regional bias in the rankings. It does a terrible job in that regard. The SEC had as bad of a year in WBB last year as they have had in a long time. They still sent half of their conference to the post season. In the Pac12 we have to have a banner year, like last year, to send half our conference to the post season. If the conference doesn't have a good year, we are lucky to get 3 or 4 teams in. You might be just fine with that discrepancy, but I'm not. For such a smart guy you're lost and full of excuses... the Pac12, media, selection committee, RPI can't change the population centers. It's not rewarding or bias... it's how the country and the population was/is dispersed. Statistical crap like RPI is bad enough, what do you propose another statistical 'adjustment' to eliminate the imaginary bias??? LOL It only does a terrible job in the minds of those without a better solution. If you understand how RPI is calculated (like SOS and many other stats) the more teams you play with better resumes, the better off you are. On the the left coast much harder to do, as stated. It's not RPI's fault. If West Coast teams want to improve their RPI travel East. As stated... the hypocrisy of saying RPI favors East Coast teams yet that is where you go to play more power house programs. That is well known! LOL... SEC had (7) teams in... Seeded #1, #2, #4 (x2), #5, #6... one more team and same % of the conference as the Pac12. The SEC went 12-7, Pac12 12-6... SEC (5) and Pac12 (5) advanced to Rd of 32... SEC (3) and Pac12 (4) advanced to Rd of 16... SEC (2) and Pac12 (3) advanced to Rd of 8... SEC (1) and Pac12 (0) advanced to Rd of 4... SEC (1) and Pac12 (0) advanced to NC game. Effing terrible year!! So if that was a terrible year for the SEC the Pac12 was then worse and didn't deserve (6) seeds? 2017... SEC (8)... Pac12 (7)... both 15-7... SEC had both NC game participants 2016... SEC (9)... Pac12 (5)... 11-9/13-4... SEC had two in Final 4... this would be the SEC's worst showing with 2 major early upsets Anyway... bias... BS |
|
|
|
Post by nwhoopfan on Oct 23, 2019 11:10:26 GMT -8
What year are you looking at baseba1111? Last year's NC game was between Baylor and Notre Dame, other FF teams were Oregon and UConn. That would be zero SEC teams in the FF or the NC game. Stanford and Mississippi St. both made the Elite 8. SEC went 9-7 w/ 7 teams. Pac 12 went 14-6 w/ 6 teams. That's a significant difference. Pac 12 tied ACC for most combined victories, but ACC had 2 more teams participating.
|
|
|
|
Post by baseba1111 on Oct 23, 2019 14:42:29 GMT -8
What year are you looking at baseba1111? Last year's NC game was between Baylor and Notre Dame, other FF teams were Oregon and UConn. That would be zero SEC teams in the FF or the NC game. Stanford and Mississippi St. both made the Elite 8. SEC went 9-7 w/ 7 teams. Pac 12 went 14-6 w/ 6 teams. That's a significant difference. Pac 12 tied ACC for most combined victories, but ACC had 2 more teams participating. Three consecutive years vs one year of some upsets... '16-18. There is no population bias in the RPI formula... just because one conference is larger, one conference has had more consistent success, one conference has members playing consistently 'better' opponents is not a bias. The RPI formula doesn't look at where a school is located, what it's population center is... it's pretty basic math... RPI = (WP * 0.25) + (OWP * 0.50) + (OOWP * 0.25) where WP is Winning Percentage, OWP is Opponents' Winning Percentage and OOWP is Opponents' Opponents' Winning Percentage. The committee fills the bracket ist doesn't predict success/wins. Looking back on one year without looking at the huge success of a conference is foolish. Now if the argument is that RPI may be bias toward past success, ok. But, that is consistent from rankings on down the line. Past successes of teams and conferences build reputation and that obviously influences pollsters, media, statisticians, awards, etc. People can't have it both ways... do you honestly think that OSU is the 6th best team in the country to begin the season? But, historical success and SR's development of players/team have a direct correlation. Just as it will in a seeding room. The Pac12 will not build a reputation by getting 7+ teams in to a tournament. It will by consistently getting teams deep or maybe winning a Final 4 game or two. Close, but no cigar. And, since most of the mainstream media and hype machine is East the Pac12 is seen as no titles since 1992? Runners up 2008/10... but all Stanford. Bias in the media? Sure. But the media doesn't play a part in the RPI. All you have to do is look at Furd... the signature team of the West. TV seen as the preeminent coach of the West and top 5 of all time... and really until SR's recent success the only program with huge NCAA success... TV has 32 season with 30 NCAA appearances... 27 Sweet 16 or further. To some Furd still is the West. Again, stupid? bias? arrogance? But, none influence the RPI directly. As stated... pretty simple... want to improve your RPI and stature... play better NC teams. Or, go to the MBB NET ranking system... |
|
|
|
Post by nwhoopfan on Oct 23, 2019 15:47:04 GMT -8
Okay. Well agreed that until a Pac 12 team wins a Championship, that no titles since '92 thing can keep being held against them. I would say the Pac 12 stands alone in number of teams they've put in the Final 4 in fairly recent years--Stanford, Cal, Oregon St., Washington and Oregon; nearly half the conference members. Who else has done that? SEC's recent success is pretty much all South Carolina and Mississippi St., Tennessee is a pale shadow of their former selves and nobody else is really a national powerhouse (LSU used to be but has totally faded, Texas A&M's fluky run to a Championship was while they were in the Big 12). Similarly ACC is pretty much Notre Dame and Louisville (Syracuse did make a FF same year as OSU and UW; Duke used to be a power but also has totally faded). Some other solid teams but nobody else making a habit of deep runs. The Big 12 IS Baylor, without them they'd be nothing (Texas and Oklahoma have some history but nothing recent). Maryland has a few FF since joining the Big 10, but mostly that conference doesn't accomplish much in March/April. AAC will become irrelevant as soon as UConn leaves for the Big East.
Hmm, looking closer at SEC success in recent years--1 FF team in '15, 2 in '17 and 1 in '18. They had ZERO '09-'14, and also '16 and '19. That's really rather an unimpressive stretch for over a decade. That's a total of 4. Stanford alone has 6 in that time frame, and then add in 4 more w/ 1 each for 4 other Pac 12 schools. Prior to '09 Tennessee and LSU were both getting to the FF regularly but that was a different era and as mentioned both teams are nothing like they were back then.
|
|
|
|
Post by beavheart on Oct 23, 2019 17:06:40 GMT -8
For such a smart guy, I'm not sure why this is lost on you? The RPI ABSOLUTELY rewards the population centers for exactly the reason you just stated. When "your" SEC or ACC school is surrounded by a bunch of lower conference programs that live somewhere in the middle of the RPI because they are "good mid-majors", but that you can count on beating 9 out of 10 times then your RPI rating gets a big boost. When you are OSU and the next best program outside of the Pac12 anywhere close by is Gonzaga, (an 8 hour drive away), and then there are crickets after that in terms of a quality opponents within shouting distance then you have a very difficult time padding your RPI rating. This dichotomy is magnified at the conference level because lower level teams in other conferences have more opportunities to play and win games against "quality" opponents. I thought this was pretty universally understood. This happens in all sports. Again, the RPI does the Pac12 NO favors. The RPI basically just mirrors the population centers. The RPI was created in part to remove regional bias in the rankings. It does a terrible job in that regard. The SEC had as bad of a year in WBB last year as they have had in a long time. They still sent half of their conference to the post season. In the Pac12 we have to have a banner year, like last year, to send half our conference to the post season. If the conference doesn't have a good year, we are lucky to get 3 or 4 teams in. You might be just fine with that discrepancy, but I'm not. For such a smart guy you're lost and full of excuses... the Pac12, media, selection committee, RPI can't change the population centers. It's not rewarding or bias... it's how the country and the population was/is dispersed. Statistical crap like RPI is bad enough, what do you propose another statistical 'adjustment' to eliminate the imaginary bias??? LOL It only does a terrible job in the minds of those without a better solution. If you understand how RPI is calculated (like SOS and many other stats) the more teams you play with better resumes, the better off you are. On the the left coast much harder to do, as stated. It's not RPI's fault. If West Coast teams want to improve their RPI travel East. As stated... the hypocrisy of saying RPI favors East Coast teams yet that is where you go to play more power house programs. That is well known! LOL... SEC had (7) teams in... Seeded #1, #2, #4 (x2), #5, #6... one more team and same % of the conference as the Pac12. The SEC went 12-7, Pac12 12-6... SEC (5) and Pac12 (5) advanced to Rd of 32... SEC (3) and Pac12 (4) advanced to Rd of 16... SEC (2) and Pac12 (3) advanced to Rd of 8... SEC (1) and Pac12 (0) advanced to Rd of 4... SEC (1) and Pac12 (0) advanced to NC game. Effing terrible year!! So if that was a terrible year for the SEC the Pac12 was then worse and didn't deserve (6) seeds? 2017... SEC (8)... Pac12 (7)... both 15-7... SEC had both NC game participants 2016... SEC (9)... Pac12 (5)... 11-9/13-4... SEC had two in Final 4... this would be the SEC's worst showing with 2 major early upsets Anyway... bias... BS I don't have time to knock down all the straw man arguments you create. I'm not making excuses for anything. I am merely pointing out what I think is a flawed system that is systematically hurting our conference and our school. I don't have an RPI system that fairly balances out the distance problem. What pisses me off is that no one who should care or have the ability to do anything about it, does either. I'm not interested in going rounds about a bunch of points I'm not trying to make. I was talking about the season the SEC had had BEFORE the tournament had started, which wasn't great, and the fact they STILL got thier allotted 7 or 8 teams. In a good year it's more. THE POINT I AM MAKING is that there are biases built in to the process that are self fulfilling, and the RPI is one of them. If you don't agree with me, fine, but I can't for the life of me understand why. So, you are fine with the fact that the worst the SEC can do is put half their conference in the post season, while that is the best the Pac12 can do? I mean, why? All the records you want to break out don't mean much. The very thing I am talking about skews those records to begin with, so I'm not going to be swayed by the fact that SEC performed well enough. When OVER half their conference is GOING to make the post season, and they will probably be given a bunch of high seeds they better perform well. Lately their performance hasn't even been consistent with their "billing." Meanwhile, we will get to watch our Beavs travel all over the country again because our rating by the end of the year will only warrant a 6 or 7 seed. But hey, we should just gut it out, and win in spite of it all, right Baseba11? Maybe you're right, but if you're wrong... what's your point again? The Pac12 should just travel east more? Brilliant. Let's do that. |
|
|
|
Post by baseba1111 on Oct 23, 2019 17:53:25 GMT -8
For such a smart guy you're lost and full of excuses... the Pac12, media, selection committee, RPI can't change the population centers. It's not rewarding or bias... it's how the country and the population was/is dispersed. Statistical crap like RPI is bad enough, what do you propose another statistical 'adjustment' to eliminate the imaginary bias??? LOL It only does a terrible job in the minds of those without a better solution. If you understand how RPI is calculated (like SOS and many other stats) the more teams you play with better resumes, the better off you are. On the the left coast much harder to do, as stated. It's not RPI's fault. If West Coast teams want to improve their RPI travel East. As stated... the hypocrisy of saying RPI favors East Coast teams yet that is where you go to play more power house programs. That is well known! LOL... SEC had (7) teams in... Seeded #1, #2, #4 (x2), #5, #6... one more team and same % of the conference as the Pac12. The SEC went 12-7, Pac12 12-6... SEC (5) and Pac12 (5) advanced to Rd of 32... SEC (3) and Pac12 (4) advanced to Rd of 16... SEC (2) and Pac12 (3) advanced to Rd of 8... SEC (1) and Pac12 (0) advanced to Rd of 4... SEC (1) and Pac12 (0) advanced to NC game. Effing terrible year!! So if that was a terrible year for the SEC the Pac12 was then worse and didn't deserve (6) seeds? 2017... SEC (8)... Pac12 (7)... both 15-7... SEC had both NC game participants 2016... SEC (9)... Pac12 (5)... 11-9/13-4... SEC had two in Final 4... this would be the SEC's worst showing with 2 major early upsets Anyway... bias... BS I don't have time to knock down all the straw man arguments you create. I'm not making excuses for anything. I am merely pointing out what I think is a flawed system that is systematically hurting our conference and our school. I don't have an RPI system that fairly balances out the distance problem. What pisses me off is that no one who should care or have the ability to do anything about it, does either. I'm not interested in going rounds about a bunch of points I'm not trying to make. I was talking about the season the SEC had had BEFORE the tournament had started, which wasn't great, and the fact they STILL got thier allotted 7 or 8 teams. In a good year it's more. THE POINT I AM MAKING is that there are biases built in to the process that are self fulfilling, and the RPI is one of them. If you don't agree with me, fine, but I can't for the life of me understand why. So, you are fine with the fact that the worst the SEC can do is put half their conference in the post season, while that is the best the Pac12 can do? I mean, why? All the records you want to break out don't mean much. The very thing I am talking about skews those records to begin with, so I'm not going to be swayed by the fact that SEC performed well enough. When OVER half their conference is GOING to make the post season, and they will probably be given a bunch of high seeds they better perform well. Lately their performance hasn't even been consistent with their "billing." Meanwhile, we will get to watch our Beavs travel all over the country again because our rating by the end of the year will only warrant a 6 or 7 seed. But hey, we should just gut it out, and win in spite of it all, right Baseba11? Maybe you're right, but if you're wrong... what's your point again? The Pac12 should just travel east more? Brilliant. Let's do that. Blah blah blah... yet you claim your opinion as fact. See above the factual RPI calculation. No bias. Half isn't the norm and the number of teams from any conference has many determining factors. As stated at least twice, RPI is a simple mathematical process. Filling a bracket isn't. Good conferences get the benefit of the doubt, including the Pac12. |
|
|
|
Post by beavheart on Oct 24, 2019 9:18:59 GMT -8
I don't have time to knock down all the straw man arguments you create. I'm not making excuses for anything. I am merely pointing out what I think is a flawed system that is systematically hurting our conference and our school. I don't have an RPI system that fairly balances out the distance problem. What pisses me off is that no one who should care or have the ability to do anything about it, does either. I'm not interested in going rounds about a bunch of points I'm not trying to make. I was talking about the season the SEC had had BEFORE the tournament had started, which wasn't great, and the fact they STILL got thier allotted 7 or 8 teams. In a good year it's more. THE POINT I AM MAKING is that there are biases built in to the process that are self fulfilling, and the RPI is one of them. If you don't agree with me, fine, but I can't for the life of me understand why. So, you are fine with the fact that the worst the SEC can do is put half their conference in the post season, while that is the best the Pac12 can do? I mean, why? All the records you want to break out don't mean much. The very thing I am talking about skews those records to begin with, so I'm not going to be swayed by the fact that SEC performed well enough. When OVER half their conference is GOING to make the post season, and they will probably be given a bunch of high seeds they better perform well. Lately their performance hasn't even been consistent with their "billing." Meanwhile, we will get to watch our Beavs travel all over the country again because our rating by the end of the year will only warrant a 6 or 7 seed. But hey, we should just gut it out, and win in spite of it all, right Baseba11? Maybe you're right, but if you're wrong... what's your point again? The Pac12 should just travel east more? Brilliant. Let's do that. Blah blah blah... yet you claim your opinion as fact. See above the factual RPI calculation. No bias. Half isn't the norm and the number of teams from any conference has many determining factors. As stated at least twice, RPI is a simple mathematical process. Filling a bracket isn't. Good conferences get the benefit of the doubt, including the Pac12. Never claimed my opinion as fact. I would be happy to be wrong, but this is something I have noticed over years of watching our conference being dinged by the "system." Frankly, I don't know how you would think otherwise, but I'm happy to revisit this at the end of the season and see how it plays out again. Call me crazy, but I suspect that regardless of how well the Pac12 does this year, we will still be looking up at the SEC and ACC in terms of percentage of teams in the post season, and perhaps more importantly that in total they will be given better seeding. If I'm wrong, great! But I'll be honest...not a ton of faith over here that I will be wrong. When the same thing happens over, and over again, at some point it stops being coincidence. |
|
|
|
Post by bennyskid on Oct 24, 2019 12:16:59 GMT -8
That the RPI is biased is a mathematical fact. Specifically, it is biased against winning teams that play in a smaller pool. It is most evident in baseball, where the pool of west coast teams is much smaller than the pool of east coast teams and teams travel less. But it's significant in WBB and might still be in MBB.
I'll illustrate with an extreme example. Suppose OSU plays in a two-team pool and UConn plays in a four-team pool. Furthermore, lets assume that all the other teams are equal, but inferior to OSU and UConn. We'll assume that every team plays three games. Then the results are:
Western Pool
OSU 3-0
Western A 0-3
Eastern Pool
UConn 3-0
Eastern A 1-2
Eastern B 1-2
Eastern C 1-2
The RPI formula is (WinPct *.25) + (OppWinPct *.50) + (OppOppWinPct * .25)
For OSU, that would be (1.000 * 0.25) + (0.000 * 0.5) + (1.000 * 0.25) = 0.500
For UConn, that would be (1.000 * 0.25) + (0.333 * 0.5) + (0.555 * 0.25) = 0.555
The key is that when a good team is in a smaller pool, its victories depress the opponents' winning percentages more. The difference is surprisingly significant, particularly since tiny differences can make massive differences in the rankings. Last year, #24 OSU had an RPI of .6105, while #17 South Carolina had 0.6188, just 0.0083 different. The pool bias isn't the worst thing about RPI, but it absolutely exists.
|
|
|
|
Post by baseba1111 on Oct 24, 2019 13:53:28 GMT -8
That the RPI is biased is a mathematical fact. Specifically, it is biased against winning teams that play in a smaller pool. It is most evident in baseball, where the pool of west coast teams is much smaller than the pool of east coast teams and teams travel less. But it's significant in WBB and might still be in MBB. I'll illustrate with an extreme example. Suppose OSU plays in a two-team pool and UConn plays in a four-team pool. Furthermore, lets assume that all the other teams are equal, but inferior to OSU and UConn. We'll assume that every team plays three games. Then the results are: Western Pool OSU 3-0 Western A 0-3 Eastern Pool UConn 3-0 Eastern A 1-2 Eastern B 1-2 Eastern C 1-2 The RPI formula is (WinPct *.25) + (OppWinPct *.50) + (OppOppWinPct * .25) For OSU, that would be (1.000 * 0.25) + (0.000 * 0.5) + (1.000 * 0.25) = 0.500 For UConn, that would be (1.000 * 0.25) + (0.333 * 0.5) + (0.555 * 0.25) = 0.555 The key is that when a good team is in a smaller pool, its victories depress the opponents' winning percentages more. The difference is surprisingly significant, particularly since tiny differences can make massive differences in the rankings. Last year, #24 OSU had an RPI of .6105, while #17 South Carolina had 0.6188, just 0.0083 different. The pool bias isn't the worst thing about RPI, but it absolutely exists. EXCEPT... each team has the opportunity to play whomever they chose and the "pool" is exactly the same size. Each team until the postseason plays the exact same number of opponents... period. OSU's "pool" is not smaller. And, your "math" is correct but greatly skewed by the example you show. The "pools" are not such you can define in such a simplistic way as each opponent has their own set of opponents, each interacting with the formula. |
|
|
|
Post by bennyskid on Oct 24, 2019 14:22:17 GMT -8
Riiiiight. Pick a pacific-time-zone team at random and count how many central/eastern-time-zone teams it plays. The norm is three or four, out of 30+ game schedule. That isn't nearly enough to equalize the SoS component.
This really isn't something that you can argue, baseba1111. It isn't opinion, it's mathematics. Given the data and some programming skill, the precise amount of bias can be calculated. My back-of-the-envelope estimate is that it's between 0.0025 and 0.0075 points.
|
|
|
|
Post by lotrader on Oct 24, 2019 15:58:49 GMT -8
Riiiiight. Pick a pacific-time-zone team at random and count how many central/eastern-time-zone teams it plays. The norm is three or four, out of 30+ game schedule. That isn't nearly enough to equalize the SoS component.
This really isn't something that you can argue, baseba1111. It isn't opinion, it's mathematics. Given the data and some programming skill, the precise amount of bias can be calculated. My back-of-the-envelope estimate is that it's between 0.0025 and 0.0075 points.
You can always place a poster on "ignore" which is what I did some time ago. It makes reading this site much easier, as I don't have to wade through a bunch of nonsense. |
|
|
|
Post by beavs6 on Oct 24, 2019 16:15:10 GMT -8
Riiiiight. Pick a pacific-time-zone team at random and count how many central/eastern-time-zone teams it plays. The norm is three or four, out of 30+ game schedule. That isn't nearly enough to equalize the SoS component.
This really isn't something that you can argue, baseba1111. It isn't opinion, it's mathematics. Given the data and some programming skill, the precise amount of bias can be calculated. My back-of-the-envelope estimate is that it's between 0.0025 and 0.0075 points.
Oh 4 can argue it all day long. Don't underestimate him |
|
