Imagine a conference where...

[The Mad Hatter]

The Mad Hatter,

I'm really confused. First, you said:

certain schools are able to avoid playing true road games

I told you that true road games are good for a team's RPI, so you then said:

That may well be true, but it appears to me that the rise of the neutral site game has been the real story since that formula change.

I told you that BCS teams only play about 1 neutral site game anywhere near campus (where you said they would have a semi-home court advantage). I said that this 1 game is not that big of deal over the course of a whole season, so you said:

It might or might not be, but that is hardly the extent of BCS scheduling advantages, which have been discussed at length both here and in the national media, and I'm not really interested in rehashing at the moment.

Every time you make a point, I respond, you agree, and then you move on to another point. Now you are going to simply say "well, there are a bunch of other points I don't want to mention right now"? That's an odd place to leave the discussion. Needless to say, I'm less than convinced.

Ok, you want me to flesh out a few problems with scheduling. Here goes:

1. Let's talk about those close to home neutral site games. They certainly may not always create an advantage (see our game against LSU, although even there LSU got to play in front of a home crowd without getting the home loss penalty for losing to us). However, I think we would all agree that neutral site games, in general, involve higher level opponents rather than low competition buy-out games.

So, if BCS schools average playing one game a year in a home like neutral environment against quality competition, does this give them an advantage? In many cases yes. It allows them to schedule a game against top-50 or top-100 competition with home crowd advantage, but without RPI penalty for a loss.

For how many teams on the bubble would one more top-50 win be the difference between being in and being out? If a Mid had to go on the road or to a true neutral court to get that win, their chances of getting it are lower. Perhaps the RPI reward for getting the true road win is greater, but the home court neutral is a much more winnable game, while mitigating the RPI differential between a road win and home loss.

Furthermore, this effect is magnified on the conference level. If your entire conference has say 12 extra opportunities for quality wins in a home like environment, but is not penalized for a home loss if they loss any of them, it provides an advantage to the whole conference's RPI.

2. The other area where neutral site games have gained even more prominence is early season tournaments. At least these are generally truly neutral site games, and do provide quality opportunities to both mid-majors and BCS schools.

Many of these tournaments, however, are seeded to the advantage of BCS schools. Take this season's NIT Pre-Season Tip Off. Missouri State was in the field with a chance to go to New York to play on a neutral court against decent RPI teams. However, in order to get the neutral site games, they were required to play Tennessee on Tennessee's home court, with the winner advancing. There was no merit system as to why Tennessee should get to play at home to advance, but they did. The result was a 4 point loss for MSU, which easily could have been a win on a neutral court or at home. The set up of this tournament gives certain schools better chances of winning and advancing to other good RPI games than to others.

[Jamar Howard 4 President]

Thanks for the explanation. I'll need more than a couple minutes to digest... get back to you later.

[Kung Wu]

Good work Kung Wu, here are my thoughts...

You listed the swings that win ranks and SOS ranks take, but then you went back to looking at the ranks themselves when you did your formula.

After accounting for the % weight each area holds:

Win Rank varies by 0.095
SOS Rank varies by 0.091
0.095 / (0.095 + 0.091) = 51%

When looking at the top 68 teams, on average 51% of the variance is based on an individual team's winning %. That leaves 49% to SOS. Honestly, it is hard to say what the percentages should be ideally, but considering that the top 68 includes teams from Georgetown to Utah State, SOS needs to have a solid part in the formula. Obviously, something crazy like 90% would be too high, but if we get too low, teams like Coastal Carolina and Belmont are going to be ranked in the top 10.

You are completely mashing two different concepts. My premise is that RPI itself is flawed because too much weight is placed on what a team has no control over, other than scheduling. I then proved that on average 69% of the final weighting is based on SOS -- which a team has no control over, other than scheduling.

The variance of Win Rank and SOS Rank is a completely different animal (that I showed only for interest). It has nothing to do with the weight.

So now the next step is to see how variance furthers your argument that the elite teams are properly selected and seated, which I will do when I get more spreadsheet time.

So far this statement still holds for the top 68 RPI spots:

On average, a WHOPPING 69% of RPI is based SOLELY on the strength of schedule.

More to come on your premise that the variance of SOS rankings makes the heavy weighting of SOS irrelevant -- but not until I get more down time.

[ShockTalk]

The Mad Hatter,

I'm really confused. First, you said:

certain schools are able to avoid playing true road games

I told you that true road games are good for a team's RPI, so you then said:

That may well be true, but it appears to me that the rise of the neutral site game has been the real story since that formula change.

I told you that BCS teams only play about 1 neutral site game anywhere near campus (where you said they would have a semi-home court advantage). I said that this 1 game is not that big of deal over the course of a whole season, so you said:

It might or might not be, but that is hardly the extent of BCS scheduling advantages, which have been discussed at length both here and in the national media, and I'm not really interested in rehashing at the moment.

Every time you make a point, I respond, you agree, and then you move on to another point. Now you are going to simply say "well, there are a bunch of other points I don't want to mention right now"? That's an odd place to leave the discussion. Needless to say, I'm less than convinced.

Ok, you want me to flesh out a few problems with scheduling. Here goes:

1. Let's talk about those close to home neutral site games. They certainly may not always create an advantage (see our game against LSU, although even there LSU got to play in front of a home crowd without getting the home loss penalty for losing to us). However, I think we would all agree that neutral site games, in general, involve higher level opponents rather than low competition buy-out games.

So, if BCS schools average playing one game a year in a home like neutral environment against quality competition, does this give them an advantage? In many cases yes. It allows them to schedule a game against top-50 or top-100 competition with home crowd advantage, but without RPI penalty for a loss.

For how many teams on the bubble would one more top-50 win be the difference between being in and being out? If a Mid had to go on the road or to a true neutral court to get that win, their chances of getting it are lower. Perhaps the RPI reward for getting the true road win is greater, but the home court neutral is a much more winnable game, while mitigating the RPI differential between a road win and home loss.

Furthermore, this effect is magnified on the conference level. If your entire conference has say 12 extra opportunities for quality wins in a home like environment, but is not penalized for a home loss if they loss any of them, it provides an advantage to the whole conference's RPI.

2. The other area where neutral site games have gained even more prominence is early season tournaments. At least these are generally truly neutral site games, and do provide quality opportunities to both mid-majors and BCS schools.

Many of these tournaments, however, are seeded to the advantage of BCS schools. Take this season's NIT Pre-Season Tip Off. Missouri State was in the field with a chance to go to New York to play on a neutral court against decent RPI teams. However, in order to get the neutral site games, they were required to play Tennessee on Tennessee's home court, with the winner advancing. There was no merit system as to why Tennessee should get to play at home to advance, but they did. The result was a 4 point loss for MSU, which easily could have been a win on a neutral court or at home. The set up of this tournament gives certain schools better chances of winning and advancing to other good RPI games than to others.

:good: Nice work, TMH. For example, Syracuse had zero road games, but 3 at neutral sites -- New York City and 2 in Atlantic City, NJ. Just how far did a large fan base really have to travel?

[Jamar Howard 4 President]

Ok, you want me to flesh out a few problems with scheduling. Here goes:

1. Let's talk about those close to home neutral site games. They certainly may not always create an advantage (see our game against LSU, although even there LSU got to play in front of a home crowd without getting the home loss penalty for losing to us). However, I think we would all agree that neutral site games, in general, involve higher level opponents rather than low competition buy-out games.

So, if BCS schools average playing one game a year in a home like neutral environment against quality competition, does this give them an advantage? In many cases yes. It allows them to schedule a game against top-50 or top-100 competition with home crowd advantage, but without RPI penalty for a loss.

For how many teams on the bubble would one more top-50 win be the difference between being in and being out? If a Mid had to go on the road or to a true neutral court to get that win, their chances of getting it are lower. Perhaps the RPI reward for getting the true road win is greater, but the home court neutral is a much more winnable game, while mitigating the RPI differential between a road win and home loss.

Furthermore, this effect is magnified on the conference level. If your entire conference has say 12 extra opportunities for quality wins in a home like environment, but is not penalized for a home loss if they loss any of them, it provides an advantage to the whole conference's RPI.

2. The other area where neutral site games have gained even more prominence is early season tournaments. At least these are generally truly neutral site games, and do provide quality opportunities to both mid-majors and BCS schools.

Many of these tournaments, however, are seeded to the advantage of BCS schools. Take this season's NIT Pre-Season Tip Off. Missouri State was in the field with a chance to go to New York to play on a neutral court against decent RPI teams. However, in order to get the neutral site games, they were required to play Tennessee on Tennessee's home court, with the winner advancing. There was no merit system as to why Tennessee should get to play at home to advance, but they did. The result was a 4 point loss for MSU, which easily could have been a win on a neutral court or at home. The set up of this tournament gives certain schools better chances of winning and advancing to other good RPI games than to others.

I decided to check just how common these "semi-home" games are. I used Ken Pomeroy's site because he very clearly distinguishes home, semi-home, neutral, semi-road, or away. Guess what?!

Out of his top 45 teams, there were only 15 games played on "semi-home" courts. That's only 1 of 3 teams playing such a game... much less than we had previously assumed.

Of those 15 games, there were 5 instances of a BCS team getting a "semi-home" game over a mid-major and one instance (Gonzaga vs Illinois) of a mid-major actually getting a semi-home game over a BCS team.

If you let the Gonzaga game cancel out one of the 5, and you count BCS teams in Pomeroy's top 45, you are left with the top 34 BCS teams playing a net 4 semi-home games over mid-majors.

Neutral courts are never as much of an advantage as home courts, regardless of the crowd. In the case of the LSU game, their crowd was weak and the advantage was negligible. Even in the most hostile environment, a neutral court can never quite match a home court. The crowd may be an advantage, but both teams are still playing in an unfamiliar environment. Home teams normally win 65% of the time. Even the best neutral court advantage is probably only 60/40. This 60/40 advantage equates to about a 10% greater chance of winning. Personally, I think 60/40 is an exaggeration and only shows up in the most extreme neutral court cases, so I think the real effect is even less than this. Also, don’t forget that most of these semi-home games are between 2 BCS teams, so there isn't any BCS bias there. Every bit of advantage that the "semi-home" team gains is a disadvantage for a fellow BCS team playing a "semi-road" game.

Moral of the story, we are talking about an extremely small effect on the college basketball landscape as a whole. Roughly 1 in 8 BCS teams is getting to play 1 game on a neutral court where they have about a 10% increase in their chance of winning due to the "semi-home" factor. That averages out to 1 BCS team winning 1 extra game per year, all conferences included. I'm pretty sure this is a non-issue.

I'll look at your second point in a separate post.

[ShockTalk]

JH4P - Does Pomeroy's give a definition of semi-home vs neutral game?

[Jamar Howard 4 President]

:good: Nice work, TMH. For example, Syracuse had zero road games, but 3 at neutral sites -- New York City and 2 in Atlantic City, NJ. Just how far did a large fan base really have to travel?

Syracuse played all 3 games against BCS teams. That means that any advantage Syracuse received, the other BCS teams were playing at a disadvantage even though the schedule called it a neutral game. The location of these games did absolutely nothing to help BCS teams gain an advantage over mid-majors.

[Jamar Howard 4 President]

JH4P - Does Pomeroy's give a definition of semi-home vs neutral game?

I don't know his general rule. He called New York City a semi-home game for Syracuse. He called Atlantic City (a 5 hour drive) a neutral game.

[shoxlax]

Bottom line is this:

http://espn.go.com/blog/collegebasketballnation/post/_/id/12511/scheduling-still-difficult-for-mid-majors

I may be a good hitter but only three at bats is not a worthy representation to show my ability.

Give me 12-15 at bats, then we will talk.

[ShockTalk]

:good: Nice work, TMH. For example, Syracuse had zero road games, but 3 at neutral sites -- New York City and 2 in Atlantic City, NJ. Just how far did a large fan base really have to travel?

Syracuse played all 3 games against BCS teams. That means that any advantage Syracuse received, the other BCS teams were playing at a disadvantage even though the schedule called it a neutral game. The location of these games did absolutely nothing to help BCS teams gain an advantage over mid-majors.

My arguement is a little different than TMH. I think teams should be required to have a minimum of true road and neutral games in the OOC. A point system.....like having a minimum of 3.25 points - 1 point per true road game, .75 for a semi-road game, .5 for a neutral game, and .25 for a semi-home game. Failure to do so would have penalties to RPI and to qualification to Dance.

[Jamar Howard 4 President]

Fair enough. I would actually like to see the NCAA do something similar. It is a little silly that teams can just sit at home until January. The NCAA has plenty of power to do something about this. I wish they would.

However, let's just make sure we aren't justifying our desires with arguments that aren't true. So far, I think I've shot down 2 of them:

- BCS teams get an RPI boost by not having to play on the road in non-conf (False)
- BCS teams get an RPI boost by playing "neutral" games vs mid-majors at sites that give them a decent home court advantage (Extremely minimal)

[ShockTalk]

JH4P - Does Pomeroy's give a definition of semi-home vs neutral game?

I don't know his general rule. He called New York City a semi-home game for Syracuse. He called Atlantic City (a 5 hour drive) a neutral game.

That's why I was curious. Atlantic City is only an hour farther away from Syracuse, but the difference for either Michigan school to NY vs Atlantic City is an additional 30 minutes. Seems to be spliting hairs, given both games were outside sponsored, Jimmie V and Legends Classic.

[shoxlax]

:good: Nice work, TMH. For example, Syracuse had zero road games, but 3 at neutral sites -- New York City and 2 in Atlantic City, NJ. Just how far did a large fan base really have to travel?

Syracuse played all 3 games against BCS teams. That means that any advantage Syracuse received, the other BCS teams were playing at a disadvantage even though the schedule called it a neutral game. The location of these games did absolutely nothing to help BCS teams gain an advantage over mid-majors.

My arguement is a little different than TMH. I think teams should be required to have a minimum of true road and neutral games in the OOC. A point system.....like having a minimum of 3.25 points - 1 point per true road game, .75 for a semi-road game, .5 for a neutral game, and .25 for a semi-home game. Failure to do so would have penalties to RPI and to qualification to Dance.

I suggested this last year and got shot down with the notion that KU would play at UMKC, Texas would play at Prarie View, Duke would play at North Carolina A&T, and so on and so forth. Any counter to that one?

[ShockTalk]

:good: Nice work, TMH. For example, Syracuse had zero road games, but 3 at neutral sites -- New York City and 2 in Atlantic City, NJ. Just how far did a large fan base really have to travel?

Syracuse played all 3 games against BCS teams. That means that any advantage Syracuse received, the other BCS teams were playing at a disadvantage even though the schedule called it a neutral game. The location of these games did absolutely nothing to help BCS teams gain an advantage over mid-majors.

My arguement is a little different than TMH. I think teams should be required to have a minimum of true road and neutral games in the OOC. A point system.....like having a minimum of 3.25 points - 1 point per true road game, .75 for a semi-road game, .5 for a neutral game, and .25 for a semi-home game. Failure to do so would have penalties to RPI and to qualification to Dance.

I suggested this last year and got shot down with the notion that KU would play at UMKC, Texas would play at Prarie View, Duke would play at North Carolina A&T, and so on and so forth. Any counter to that one?

To be honest, I don't know exactly how the RPI works, but I would think who you play/beat on the road has an RPI impact and, certainly, it would look poorly on your resume, especially if a "road/neutral site" rule became a point of emphasis.

Obviously, this is already done. Louisville played no neutral site games and only one road game, against Western Kentucky. Cincinnati played two road games against Toledo and Miami of Ohio.

Keep in mind, the kind of requirement I mentioned still wouldn't keep Big 6 teams from playing other Big 6 teams and, in many cases, a loss on the road to a quality Big 6 team wouldn't look as bad as a loss to a slightly less quality mid-major. Nevertheless, to help cure the obvious problem of mid majors being able to increase their schedule quality would take more tweeking.

[Calshockfan]

It seems to me that strength of schedule being a major component eliminates any possibility of fairness. BCS teams control simply by refusing to play a reasonable number of road games against the mid-majors. They learned their lessons in 2006. In the event that a mid-major has a superior RPI it is simply ignored because of a lack of wins over BCS schools. I believe that the current system along with JH4P's analysis to be flawed. It can only be meaningful if scheduling is somehow equalized and the influence of the media is quieted. I believe that the NCAA and NIT tournament results thus far prove that point.