Why RPI is stupid

[chef dixon]

Baseball:

William and Mary has only played 3 games against Ole Miss, going 1-2, yet their RPI is 35. Ole Miss is 5-2 with a series win over them, a series win over RPI 13 Wright State, and a win over RPI 12 Arkansas State. So after crunching the numbers, Ole Miss' RPI rounds out at 51.

I know it takes time for it to mean anything but the formula seems broken to me.

[smootness]

Why RPI is meaningless this early

FIFY

There are flaws in RPI, but using it to show that this early in the year is worthless b/c the formula makes no sense until you get much later in the year.

I get what you're saying, that a scenario like the one you highlighted shouldn't ever make sense, and I can't tell you why it looks that way, but it'll look fine by the end of the year.

[messageboardsuperhero]

Of course RPI is stupid two weeks into the season. Do you really think W&M is going to stay at 35 with a losing record?

RPI doesn't become valid until a few weeks into conference play- you shouldn't even be looking at it or trying to take anything from it at all until then.

[messageboardsuperhero]

The RPI formula/system isn't perfect, but it bugs me when people who clearly know nothing about RPI try to make stupid criticisms of it.

ETA: Not trying to call you out specifically, I was just generalizing.

[engie]

RPI is not valid in baseball until week 5. Almost wish Warren Nolan didn't put it out there until it locks in. Even then, it'll be week 7-8 before we start to settle into RPI spots. Really after that -- given that the northeast isn't even playing baseball yet and they have a trickle down effect even on our RPI...

Baseball RPI is absolutely a valid method of ranking teams. The part that's messed up and needs fixing is the home/road weighting is way off for baseball's realities. Road win = 1.3 win, home win =0.7 win. It's nowhere near twice as hard to win on the road in baseball as it is at home. I think that weighting factor should be about 1.15 and 0.85, and that's really generous if you think about it. The even better way is how I believe Boyd incorporates what he calls "park factors" into his ISR formula. Which is similar to RPI, but is weighted home and away by the actual difficulties of playing/winning in those places vs a mean. Much more precise and advanced really....

[bulldogcountry1]

From wiki: "RPI formula will value each road victory as 1.3 instead of 1.0. Each home win will be valued at 0.7 instead of 1.0. Conversely, each home loss will count 1.3 against a team’s RPI and each road loss will count 0.7 against a team’s RPI. Neutral-site games will retain the same value of 1.0"

This is where I have an issue playing too many SWAC games at home. If we must play those games, we should play them at Trustmark as the visitor to get the most value from it. I don't know how the neutral site rules work, but it seems like you could justify it being away games for if we played Alcorn and JSU there.

[Tbonewannabe]

Some parks give a huge advantage. Was it Clemson that has an incline on their warning track? Some regional we played had one and we had a couple of misplayed catches at the wall due to it.

[Johnson85]

The value of placing weight on the RPI is that it encourages teams to play tough schedules and to be willing to play away games. The downside is that if you are really relying on it to determine which teams are best, it relies way too much on SOS and being willing to play away games.

And of course as you acknowledged, it's worthless until it's been in place a while.

[HancockCountyDog]

Do we play Arkansas State this year - I think they took a series from a solid Louisville club. Im happy for Raffo.

[Smitty]

Boyd says we have the 30th toughest Pre-Season strength of schedule.. Not too shabby.

[smootness]

The value of placing weight on the RPI is that it encourages teams to play tough schedules and to be willing to play away games. The downside is that if you are really relying on it to determine which teams are best, it relies way too much on SOS and being willing to play away games.

And of course as you acknowledged, it's worthless until it's been in place a while.

I do agree with this to some degree. I think it rewards teams for challenging themselves more than it rewards teams for simply being good. At the same time, it's hard to come up with an objective measurement that truly rates teams based on 'how good' they are that's any better. You have to factor in schedule, road/home, etc. All in all, I think RPI is a pretty good measure, though I wish they would de-emphasize it just a little in actually selecting and seeding teams. It should be used as a tool, not as an end.

[smootness]

Boyd says we have the 30th toughest Pre-Season strength of schedule.. Not too shabby.

Our SEC schedule will be tougher than some, for sure, though UGA looks like they're going to be a good bit better than anybody thought, and we don't play them.

The West is really strong, and we don't play UGA or Kentucky.

As many have said, if we just win enough games OOC, or OOC schedule won't hurt us at all. We just can't lose more than 4-5 of them, and we really can't afford to lose any to the type of teams we lost to last year.

[MedDawg]

Baseball:

William and Mary has only played 3 games against Ole Miss, going 1-2, yet their RPI is 35. Ole Miss is 5-2 with a series win over them, a series win over RPI 13 Wright State, and a win over RPI 12 Arkansas State. So after crunching the numbers, Ole Miss' RPI rounds out at 51.

I know it takes time for it to mean anything but the formula seems broken to me.

The thing that bugs me about RPI is how beating a #250+ team will hurt an RPI way too much more than beating a #150 team, when both are just as beatable by a top 20 team. There should be some kind of diminishing penalty or max RPI once past a certain point, like all teams >150 count the same.

And an away win is worth almost twice as much as a home win? That's stupid. No way there should be that much of a difference.

These two things can artificially force a team from a #8 RPI down to a #12 and cost them a hosting spot.

If we are going to use the RPI system to our advantage (and not our detriment) then we will have to stop playing the in-state SWAC teams, no matter how much it may help them to play us, and replace them with away games.

[smootness]

But it's not as though 4-0 State was deserving of a #1 spot. I mean, basically the only reason we were there is because we played two teams and 4 games in the first weekend instead of one 3-games series. It had nothing to do with quality.

[messageboardsuperhero]

The thing that bugs me about RPI is how beating a #250+ team will hurt an RPI way too much more than beating a #150 team, when both are just as beatable by a top 20 team. There should be some kind of diminishing penalty or max RPI once past a certain point, like all teams >150 count the same.

And an away win is worth almost twice as much as a home win? That's stupid. No way there should be that much of a difference.

These two things can artificially force a team from a #8 RPI down to a #12 and cost them a hosting spot.

If we are going to use the RPI system to our advantage (and not our detriment) then we will have to stop playing the in-state SWAC teams, no matter how much it may help them to play us, and replace them with away games.

This is why there have been multiple threads about us needing to do better a job scheduling to boost our RPI. We should drop the SWAC games and replace them with decent mid-majors, and we should play a road non-conference series every single year somewhere warm when it's cold here.

[chef dixon]

Yea I have a problem with RPI rewarding teams for basically getting their ass handed to them as long as the RPI system deems their opponents highly rated in the same RPI system.

Texas in basketball is 41 RPI and 1-10 v RPI top 50.

[messageboardsuperhero]

I do agree with this to some degree. I think it rewards teams for challenging themselves more than it rewards teams for simply being good. At the same time, it's hard to come up with an objective measurement that truly rates teams based on 'how good' they are that's any better. You have to factor in schedule, road/home, etc. All in all, I think RPI is a pretty good measure, though I wish they would de-emphasize it just a little in actually selecting and seeding teams. It should be used as a tool, not as an end.

I tend to agree. RPI is currently used as the end-all-be-all by the committee when determining hosts, and I think they go a little overboard with how they emphasize it.

That said, I have yet to see any other system be as accurate with ranking teams... The system gets it right 95% of the time.

[messageboardsuperhero]

We could sit around all day bitching about how wrong the RPI weights are, how the committee emphasizes it too much, how weird the formula is, etc... Or we can buck up, use what we know about the formula, and exploit the system by scheduling in an advantageous way to help our own RPI ranking.

I choose option #2.

[bulldogcountry1]

We could sit around all day bitching about how wrong the RPI weights are, how the committee emphasizes it too much, how weird the formula is, etc... Or we can buck up, use what we know about the formula, and exploit the system by scheduling in an advantageous way to help our own RPI ranking.

I choose option #2.


I agree. While it's not always easy to know how good an opponent will be over a year out, the formula is there in black and white. There are obvious dos and don'ts that don't require the services of Steven Hawking to decipher. Don't play SWAC teams at home, and play some decent teams on the road. Cohen was smart enough to adjust his style to take advantage of BBCOR bats, so he should be able to adjust his scheduling to RPI.

I go to a majority of home games, and I would have no problem sacrificing a FEB weekend so we could play elsewhere and get a good RPI boost.

[smootness]

I think our scheduling will be fine going forward; I just don't think Cohen knew exactly what he had this year. I mean, he was willing to schedule the Houston tournament; he just backed out when he wasn't sure we could handle it.

But he's making these decisions specifically to put us in the best position in regard to postseason play. Otherwise, why back out of the Houston tournament?

I have no reason to think we won't start scheduling differently in the near future.

[messageboardsuperhero]

I think our scheduling will be fine going forward; I just don't think Cohen knew exactly what he had this year. I mean, he was willing to schedule the Houston tournament; he just backed out when he wasn't sure we could handle it.

But he's making these decisions specifically to put us in the best position in regard to postseason play. Otherwise, why back out of the Houston tournament?

I have no reason to think we won't start scheduling differently in the near future.

Thankfully we already have started scheduling differently for the future- assuming we don't back out of anything else.

It's just frustrating seeing how this year's schedule played out. It's not ALL our fault, but you could see the mistake of backing out of the Minute Maid Classic from a mile away.

[bulldogcountry1]

This would be my basic schulding philosophy:

- No control over SEC schedule. Moot.

- No SWAC teams, unless I could play them at Trustmark and have it be "away". Then, play no more than 2.

- Keep USM and UM games at Trustmark due to popularity and because they will always be decent for RPI.

- Play a true away series the second or third weekend of the season. This way, you get max potential RPI points. Playing in a tournament is nice, but some or all will be considered neutral site games that aren't as valuable. If my numbers are correct, winning 2 of 3 away = 1.9 net points. Winning 2 of 3 neutral = 1.0 net points

[State82]

I like the idea of playing the SWAC teams at Trustmark Park. Maybe we could do a weekend early in the season and get them out of the way. 3 games tops and that's it. All on a neutral field to get that little bit of RPI benefit.

[smootness]

Yeah, I say just stop playing SWAC teams altogether. Top teams just rarely schedule games like that.

I would schedule one series against a team you think is going to be good, a top-25 type team. And the rest, I'd schedule teams like Samford, Wright State, etc. Quality teams you should beat but that won't hurt you even if you don't win them all.

[maroonmania]

Baseball:

William and Mary has only played 3 games against Ole Miss, going 1-2, yet their RPI is 35. Ole Miss is 5-2 with a series win over them, a series win over RPI 13 Wright State, and a win over RPI 12 Arkansas State. So after crunching the numbers, Ole Miss' RPI rounds out at 51.

I know it takes time for it to mean anything but the formula seems broken to me.

gets LESS stupid as the year goes on because the longer you go the less the initial conditions (or initial rankings) matter. The problem is that like ANY formula with initial conditions the initial conditions are never fully factored out. And yes, to have W&M, Wright St. and Ark. St. currently sitting as 3 of the Top 35 RPI teams in all of college baseball right now does already show that the formula is very flawed because the initial assumptions are very flawed. Now, as stated the longer you go adding games the better the results but I don't see why their shouldn't be more human interaction in setting the initial RPI rankings. It helps significantly more to beat bad teams that have played tougher schedules than beating bad teams that have played weak schedules even though they may both be equally bad.

[smootness]

The RPI and formulas like it will always be inherently flawed because it uses its own formula inside its formula.

What I mean by that is, the RPI determines the strength of your schedule...by using your opponents' RPI. It's obviously flawed and circular. That said, by the end of the year, it does end up being a pretty good tool.

[dawgs]

It's not flawed, it's just not going to work correctly until you have a large enough sample size. It's not designed to give the correct results for small sample sizes.

[smootness]

It's not flawed, it's just not going to work correctly until you have a large enough sample size. It's not designed to give the correct results for small sample sizes.

It's still a flawed system because part of its formula is derived from results gathered from the same formula. How good is team A? Well, to determine that, we need to know how good teams B, C, and D are. Well, how good are those teams? That's determined partially by how good team A is. So how good is team A?

Like I said, it's inherently circular. It generally is a good tool because teams play enough variety of teams that it ends up being usable. But that flaw always exists.

It's like solving for x in an equation that relies on inputting a value of y, but the value of y is only determined by first solving for x.

[messageboardsuperhero]

gets LESS stupid as the year goes on because the longer you go the less the initial conditions (or initial rankings) matter. The problem is that like ANY formula with initial conditions the initial conditions are never fully factored out. And yes, to have W&M, Wright St. and Ark. St. currently sitting as 3 of the Top 35 RPI teams in all of college baseball right now does already show that the formula is very flawed because the initial assumptions are very flawed. Now, as stated the longer you go adding games the better the results but I don't see why their shouldn't be more human interaction in setting the initial RPI rankings. It helps significantly more to beat bad teams that have played tougher schedules than beating bad teams that have played weak schedules even though they may both be equally bad.

What are you talking about? There are no initial rankings or assumptions in RPI. It's based on your winning percentage, your opponents winning percentage, and your opponent's opponents winning percentage. It also takes into account where the game was played... And it takes every team playing about 30 or so games until the percentages work themselves out and you get something usable. I have no idea what gave you the idea that there was someone somewhere making initial rankings for the RPI system- RPI only deals with winning percentages of actual games and whether you played games home or on the road.

ETA: aGAIN- I'm sorry, it just bugs me when people who don't know how RPI works make stupid criticisms about it.

[dawgs]

It's still a flawed system because part of its formula is derived from results gathered from the same formula. How good is team A? Well, to determine that, we need to know how good teams B, C, and D are. Well, how good are those teams? That's determined partially by how good team A is. So how good is team A?

Like I said, it's inherently circular. It generally is a good tool because teams play enough variety of teams that it ends up being usable. But that flaw always exists.

It's like solving for x in an equation that relies on inputting a value of y, but the value of y is only determined by first solving for x.

Are you sure it literally uses the RPI of teams, or does it use a lot/all of the numbers/data used to derive a team's RPI as part of the formula to derive your own SoS which factors into your RPI?

The RPI doesn't have a preconceived notion of what a team's RPI is when it calculates the first week of RPI. It uses things like W/L, home/away, margin of victory (sometimes at least, opponents' W/L, home/away, etc., and opponents' opponents' W/L, home/away, etc. to calculate it. The problem with a small sample size is that W/Ls are minimized while home/away seems to be inflated. As the sample size grows, those numbers balance out and things move towards more conventional thought.

It may not be a perfect measurement, but I think your criticism is flawed because in week 1, no one has an RPI, so therefore the formula wouldn't even work if you literally used team A's RPI to calculate team B's RPI.

[chef dixon]

Are you sure it literally uses the RPI of teams, or does it use a lot/all of the numbers/data used to derive a team's RPI as part of the formula to derive your own SoS which factors into your RPI?

The RPI doesn't have a preconceived notion of what a team's RPI is when it calculates the first week of RPI. It uses things like W/L, home/away, margin of victory (sometimes at least, opponents' W/L, home/away, etc., and opponents' opponents' W/L, home/away, etc. to calculate it. The problem with a small sample size is that W/Ls are minimized while home/away seems to be inflated. As the sample size grows, those numbers balance out and things move towards more conventional thought.

It may not be a perfect measurement, but I think your criticism is flawed because in week 1, no one has an RPI, so therefore the formula wouldn't even work if you literally used team A's RPI to calculate team B's RPI.

They like to use metrics such as record v. RPI 1-25 etc. so I would assume that that is taken into consideration when determining SOS. I could be wrong. The whole thing is fluid though so its impossible to nail down.

[BLC]

Baseball RPI, at its core, is flawed because of the way the home/road formula was derived. They (whoever they are) looked at historical won/loss records for college baseball and applied that to come up with .7 vs. 1.3 formula adjustments. It's not that clean. They did not take into account the fact that for half the season (pre-conference), you have a weak team vs. strong team bias. MSU's schedule is a great illustrator of this fact. As someone else eluded, a better adjustment to use is probably around .9 to 1.1. That assumes 55% of games should be won by the home team in games where the opponents are essentially equal.

[smootness]

Are you sure it literally uses the RPI of teams, or does it use a lot/all of the numbers/data used to derive a team's RPI as part of the formula to derive your own SoS which factors into your RPI?

The RPI doesn't have a preconceived notion of what a team's RPI is when it calculates the first week of RPI. It uses things like W/L, home/away, margin of victory (sometimes at least, opponents' W/L, home/away, etc., and opponents' opponents' W/L, home/away, etc. to calculate it. The problem with a small sample size is that W/Ls are minimized while home/away seems to be inflated. As the sample size grows, those numbers balance out and things move towards more conventional thought.

It may not be a perfect measurement, but I think your criticism is flawed because in week 1, no one has an RPI, so therefore the formula wouldn't even work if you literally used team A's RPI to calculate team B's RPI.

You're right that it doesn't input a team's RPI into its formula. But my point is that if State plays Auburn, for example, in order to determine how good State is, you have to determine how good Auburn is. But to determine how good Auburn is, you have to determine how good State is.

Again, the more opponents you play, the less circular it becomes, but that aspect is always involved.

The opponents' winning % of your opponent will include you after you play them. Say, for instance, we beat Kentucky tonight in basketball. That will obviously help our RPI, but it actually won't help our SOS as much as if we had lost to Kentucky.

[Todd4State]

I may be misinterpreting this, but we don't need to play mid majors like Samford and Western Kentucky on the road. We should schedule like LSU. They barely leave the state of Louisiana. Now, if y'all are talking about scheduling a power five team like Oregon or Georgia a Tech on the road- that's fine.

Biloxi may help us out in this regard. It's another neutral site and they want college baseball there badly. We can play at least two there. The biggest thing I would do differently now is schedule a home/home with Memphis, South Alabama, and Tulane every year. Those are three pretty good and fairly local teams which would likely have fairly large pro-MSU crowds. I like the idea of playing Jackson State in Pearl, but that would make it a home game for them and they can be a pain in the ass to deal with. Incidents like the one last year are why we don't play them. I like the idea of playing Arkansas State, but that probably isn't happening as long as Raffo is there. It would be awkward more than anything and potentially open up old wounds.

[smootness]

I may be misinterpreting this, but we don't need to play mid majors like Samford and Western Kentucky on the road. We should schedule like LSU. They barely leave the state of Louisiana. Now, if y'all are talking about scheduling a power five team like Oregon or Georgia a Tech on the road- that's fine.

Yes, this is what I'm in favor of. I want to play a large majority of OOC games in Starkville, but it would probably be good to play one series on the road, and it makes sense to do that against the best team you face.

[sleepy dawg]

RPI is simply the best basic formula anyone's come up with. It's simple and it does a really good job of ordering teams based on a predefined formula that everyone can see. It is only as good as the data put in though, which (as others have said) is why it is absolutely pointless to consider it's value at this point. The more connections there are between teams, the more accurate it becomes.

There isn't a perfect formula, because no team plays exactly the same way every time. And what makes 1 team better than another team is really just a matter of opinion. RPI allows you to take the data you do have and turn into a rating system, and I have yet to see another system that is this easy to implement be as accurate as RPI when given enough result sets.

[maroonmania]

What are you talking about? There are no initial rankings or assumptions in RPI. It's based on your winning percentage, your opponents winning percentage, and your opponent's opponents winning percentage. It also takes into account where the game was played... And it takes every team playing about 30 or so games until the percentages work themselves out and you get something usable. I have no idea what gave you the idea that there was someone somewhere making initial rankings for the RPI system- RPI only deals with winning percentages of actual games and whether you played games home or on the road.

ETA: aGAIN- I'm sorry, it just bugs me when people who don't know how RPI works make stupid criticisms about it.

Well if from the very start its only based on winning percentages then how can a team like Ark. St. be 9th in the RPI right off the bat with a 4-3 record having only played teams with 4-2, 5-3 and 5-2 records? The 4-2 opponent only has one opponent with a 3-4 record, the 5-3 opponent has opponents with the following records (3-4,4-3,2-5,4-3,0-7,3-4) and the 5-2 opponent (OM) has opponents with the following records (1-2,4-2,4-3). I know what the RPI formula is suppose to be based off of but something doesn't jive even early on.

[messageboardsuperhero]

Well if from the very start its only based on winning percentages then how can a team like Ark. St. be 9th in the RPI right off the bat with a 4-3 record having only played teams with 4-2, 5-3 and 5-2 records? The 4-2 opponent only has one opponent with a 3-4 record, the 5-3 opponent has opponents with the following records (3-4,4-3,2-5,4-3,0-7,3-4) and the 5-2 opponent (OM) has opponents with the following records (1-2,4-2,4-3). I know what the RPI formula is suppose to be based off of but something doesn't jive even early on.

It's not meant to jive at all until over a month into the season- don't try to make sense of it right now. There are no initial rankings or inputs other than the actual games themselves, just trust me on that. Here some excerpts from the baseball RPI write-up from the NCAA:

An institution’s RPI ranking consists of three factors that are weighted as follows:
1. Division I Winning Percentage -- 25 percent of the RPI
2. Opponents’ Winning Percentage -- 50 percent of the RPI
3. Opponents’ Opponents’ Winning Percentage -- 25 percent of the
RPI

The RPI starts the season with every team exactly equal. Only games played against Division I teams are included in the compilation. Non-Division I games or contests against provisional Division I teams are not included in calculating winning percentage but teams do receive a deduction for every non-Division I or provisional Division I opponent played, regardless of the result or site, exceeding four per team.

[Todd4State]

It's not meant to jive at all until over a month into the season- don't try to make sense of it right now. There are no initial rankings or inputs other than the actual games themselves, just trust me on that. Here some excerpts from the baseball RPI write-up from the NCAA:

An institution’s RPI ranking consists of three factors that are weighted as follows:
1. Division I Winning Percentage -- 25 percent of the RPI
2. Opponents’ Winning Percentage -- 50 percent of the RPI
3. Opponents’ Opponents’ Winning Percentage -- 25 percent of the
RPI

The RPI starts the season with every team exactly equal. Only games played against Division I teams are included in the compilation. Non-Division I games or contests against provisional Division I teams are not included in calculating winning percentage but teams do receive a deduction for every non-Division I or provisional Division I opponent played, regardless of the result or site, exceeding four per team.

What does "exceeding four per team mean? Do they give you a deduction if you play four non D-I teams or do you get a deduction for playing any non D-I team?

IF I am reading that correctly we might be better off playing Delta State than Alcorn or any other SWAC team as long as we play them less than four times.

[smootness]

What does "exceeding four per team mean? Do they give you a deduction if you play four non D-I teams or do you get a deduction for playing any non D-I team?

IF I am reading that correctly we might be better off playing Delta State than Alcorn or any other SWAC team as long as we play them less than four times.

The way they've worded it, they're saying you're not penalized until you schedule more than 4 non-D1 teams.

The problem, though, is that while they don't count against your schedule, they also don't count as wins. And that is very important in the RPI. Your own record is extremely important.

It's better to record 3-4 wins against a team that will hurt your SOS some than to just not play 3-4 games.

[Todd4State]

The way they've worded it, they're saying you're not penalized until you schedule more than 4 non-D1 teams.

The problem, though, is that while they don't count against your schedule, they also don't count as wins. And that is very important in the RPI. Your own record is extremely important.

It's better to record 3-4 wins against a team that will hurt your SOS some than to just not play 3-4 games.

That's the way I'm reading it too. And that brings up another debate. To me, it would be better to play "54" games and cut out RPI killers Valley and Alcorn. Because if we played Delta State say twice- once in Starkville and once in Pearl, it would be like they never happened as far as RPI goes. It's essentially seen the same as our rainout with Valley.

The benefit to playing the SWAC schools is getting a lot of guys innings and at bats for future development. We could accomplish that against DSU and not take the RPI and SOS hit while facing a much better overall team.

And as far as the other SWAC games, we'll just replace them with people like Samford.

[smootness]

That's the way I'm reading it too. And that brings up another debate. To me, it would be better to play "54" games and cut out RPI killers Valley and Alcorn. Because if we played Delta State say twice- once in Starkville and once in Pearl, it would be like they never happened as far as RPI goes. It's essentially seen the same as our rainout with Valley.

The benefit to playing the SWAC schools is getting a lot of guys innings and at bats for future development. We could accomplish that against DSU and not take the RPI and SOS hit while facing a much better overall team.

And as far as the other SWAC games, we'll just replace them with people like Samford.

I'm not sure what would be worse given the current formula - not playing a handful of games or winning them against teams that hurt your SOS.

It does bring up an interesting question regarding our cancellation of Valley if it's better to go ahead and register a win than not to play at all.

[SouthMsDawg]

Yea I have a problem with RPI rewarding teams for basically getting their ass handed to them as long as the RPI system deems their opponents highly rated in the same RPI system.

Texas in basketball is 41 RPI and 1-10 v RPI top 50.

and have a 6-9 conference record.

[dawgs]

Yea I have a problem with RPI rewarding teams for basically getting their ass handed to them as long as the RPI system deems their opponents highly rated in the same RPI system.

Texas in basketball is 41 RPI and 1-10 v RPI top 50.

But that's 10 of Texas's 11 Ls. So sounds like they are a true borderline top 50 team. Rarely beat anyone ranked in the top 50 and almost always beat anyone ranked 50+.

[Smitty]

But that's 10 of Texas's 11 Ls. So sounds like they are a true borderline top 50 team. Rarely beat anyone ranked in the top 50 and almost always beat anyone ranked 50+.

So, 2012 MSU football then.

[chef dixon]

But that's 10 of Texas's 11 Ls. So sounds like they are a true borderline top 50 team. Rarely beat anyone ranked in the top 50 and almost always beat anyone ranked 50+.

That is a good point.

Wow Ole Miss basketball RPI dropped from 34 to 46 with their loss last night. That's pretty drastic, especially for losing to a team that RPI thinks is currently 29th. Its going to get hairy for them if they lose in Baton Rouge this weekend.