NCAA Tournament selection committee leader explains why even easy wins matter
Bruce Rasmussen told CBS Sports what fans need to know about the updated selection process
Creighton athletic director Bruce Rasmussen is the chair of the NCAA Tournament selection committee this year. With the bracket only days away from being revealed, Rasmussen will be the person serving as the face, body and voice of the 10-person committee. You'll see and hear him a lot over the next week, so why not get to know him now?
And before everybody can scream about unjust actions in the bracket, it's best to hear Rasmussen's opinions in the buildup to Selection Sunday. I sat down for nearly a half hour with him in New York City over the weekend to discuss a number of topics of interest to college basketball fans. Basketball fans can't get enough information when it comes to understanding how teams are debated and ultimately appointed inside and outside the bracket. Rasmussen, who has been as media-friendly and accommodating in recent months as any prior committee chair, understands that.
"As a committee we haven't done a good enough job with being transparent about the process," Rasmussen said.
Rasmussen has a high opinion of the evolving methods of the selection committee, but he also knows there are still miles to go when it comes to getting the right information and at the same time being open to everyone's opinions and observations in that committee room.
"In the room, we haven't talked enough to each other about certain parts of the process," he said. "We've done a good job, but I wished we communicated more even still."
Rasmussen compared the importance of bringing information -- and listening to others' reasoning -- to a poem he learned from his blind grandfather about an ancient allegory: the blind men and the elephant. The committee will formally begin its assignment Wednesday morning.
Below is my Q&A with him, which has been lightly edited for clarity. Rasmussen talks about everything from the time crunch of seeding the bracket to the end of the RPI era to why the quadrant system is being overplayed by the media.
CBS Sports: You're here at Madison Square Garden watching games in person, but on Saturday there were 17 games in which both teams playing are either for sure in the NCAA Tournament or fighting to get in. How do you balance the positive of watching a game in person vs. the need to see some of those results on television or a screen device?
Bruce Rasmussen: Most of the games I watch I don't watch live. I tape them, that's an old term I guess, but I record them because I can watch a game in about an hour. Later in the year, as I've seen teams, first of all I'll look at the box and the play-by-play, and I'll see where key parts of the game were and then I'll go watch that. I can watch the key parts of a game in about a half-hour. Thank god for electronics.
I had scheduled to come in and watch -- when we set this up before Christmas, and the expectation was six or eight teams in the Big Ten would be candidates for the tournament. I think there's a value in seeing teams in person. You see the physicality and the length. You see how they respond to things that don't go their way, maybe a call. You see a little bit more. You get a feel for the flow of the game and why the game ends up being the way it is.
It gives you a point of reference, too. If I've seen Michigan State in person, when Michigan State's playing Purdue, I've got a benchmark. So physicality, size, speed and so forth. It isn't that you have to see a team in order to get a better feel for them, but if you've seen somebody that they play it gives you a benchmark.
CBS Sports: One thing in particular I'm curious about and I think a lot of people will be curious about: Do you look at the team sheets daily?
BR: Every Monday. Starting on Sunday night as games are over, my Sunday night and Monday I update my own metrics and my own system. So I spend from Sunday night, when all the games are over, because games impact others, I'll go through and really look at a lot of different metrics, look at my own system. To me, and it's probably a bad analogy, but I coached track for nine years. A mile run, you know every step is important. Halfway through the season, you've got lap two in. Your job at that time is not to predict how you think it's going to finish, you're trying to accurately predict where everybody is at that point. After two weeks ago, it was 75 percent done.
Every Monday I'm trying to say, "OK, if the season ended today." People think the last few games are more important than the first few games. They're not. But the difference is, because there's so many teams compressed together, it's like watching the end of mile race where there's 20 teams all right up next to each other, so one bad step or good step passes people or puts you behind people -- where you don't notice it in lap one. It's not like it's any is more important, it's just more obvious because it's down the stretch.
CBS Sports: Disagree with this if you'd like, but in a lot of years it seems like the bubble is weak. It's the bubble: it's supposed to be weak. But this year it does seem like it's relatively strong compared to recent seasons.
BR: My feeling is basketball is deeper than it's ever been. By that I mean the difference between the 30th best team and 60th best team is paper thin. And that's why you got all these teams coming down together and that one game may make a difference. It isn't like the last game is more important than others, it's just that there's so many so close teams that one good win or one bad loss can move them.
I'm doing my own analysis. I'm seeing -- not at the front end -- but from 30 to 55? One or two games can change a team dramatically. I think it'll be very critical for the committee to have conversations that goes beyond the metrics. We can all see what KenPom is, we can all see what KPI is, we can all see what RPI is. To have the conversation to say, "You know, why is this team different than this team?" The difference between observations and metrics: Why are there outliers one way or the other? Why does this team show real well in the metrics, but when you watch them you're not quite bought into it? Or the opposite, where the metrics aren't great but when you look at them you say, "Boy, this is a very good team." Because of what reasons?
CBS Sports: One criticism some have had in recent years is that losses don't seem to matter as much as they should, particularly in some instances where teams with a lot of losses could be justified either not being put into the field or being a seed line lower. Is that true?
BR: The committee has a lot of conversations in private about those last two quadrants. However, when we're out talking, we tend to talk about the positives and not the negatives. I look at it from two [ways]. One, I think one of the hardest things to do is win the games you're supposed to. I don't think we talk about that enough. It's not necessarily that they're great wins, but you win the games you're supposed to -- 18- to 21-year-olds, that's hard.
If you coach, you see that over a 30-game period of time, probably a handful of games you say, "Boy, I wish we played this way all the time." And a handful of games you go, "I don't know who put on our uniforms, but let's just get the hell out of town." I don't want our committee to put too much value on those couple games where it just wasn't working. Or in those couple games you say, I mean, Butler beat Villanova. They hit 13 out of their last 14 3s. That's an anomaly. That's an outlier. You're trying to describe the team by looking at the entire picture, and, yes, you want to give them credit for that great win and you want to make sure you're aware of that bad loss but also understand the nature of 18- to 21-year-olds and playing 30 games.
In shorthand, sometimes we talk too much about that great win or we talk too much about that bad loss. The reality is, let's have a better description of the team.
CBS Sports: What about potential bubble teams that could have multiple, three or four, Q3 or Q4 losses, like Penn State. Is it reasonable to expect detailed discussion, if not a bracket result, that acknowledges such performance?
BR: It's a discussion item. Why did they have those bad losses? It's why observation is important. Every committee member [is assigned to monitor] seven or eight conferences. So when they have a loss that doesn't make sense, "OK, tell me. Why did they lose? Did they miss shots and go 2 for 19 from 3. Did they have [foul trouble]? Or are they a schizophrenic team?" If we're not going to have those discussions about those games, then why do you have the committee watch games and why do you have a committee?
The conversation in the room goes a lot deeper, but when we get in front of the public we've got 30 seconds to explain. If a team has three or four bad losses, we want to know why. Did they have kids out? Kids get in foul trouble? Or is that a better indication of the team?
CBS Sports: Your run as chair comes at an interesting time of transition. There's the quadrant system, but that system is still using RPI as the native data point. This could be the last year RPI is the metric embedded in team sheets, though. Where do you fall philosophically in following the rules and protocol in place with RPI while balancing the modern metrics and advanced ways of analyzing teams through data?
BR: I've seen the evolution of the committee since I've been on it for five years. The RPI is a sorting tool. People think it's used more than it is in the room. The dialogue used to be "top 50, top 25, top 75" as if that's a silo. It's a shorthand discussion but the reality in the room was, if you beat a Virginia or Villanova, that's a different win than whoever's a 24 RPI. To explain that in a short period of time is almost impossible. The RPI, while it has its critics and with reason, as we get the analytics guys together they will tell you the RPI is more accurate than the public perception -- and yet it's got its faults.
The RPI is like a 5 iron in my golf bag. I've got it in my bag. When I use it, I hope it works well, but it's not my putter and it's not my driver. As we look at and as we become more familiar with maybe some metrics that are more accurate, we've got to have that blend with the committee, even though the RPI is still the official tool. It's a balance and it has made it a little more difficult for the committee, I think.
CBS Sports: A new worry is that the introduction of the quadrant system will lead to an overemphasis on Quad 1 wins and further diminish the impact of losses. Do you agree with this presumption? For example, a road win over the No. 1-ranked team registers as a Quad 1 win just the same as beating the 74th-ranked team on the road. Is that 1-75 gap too large?
BR: The perception is that if it's in the Quadrant 1 they're all looked at the same. They're not. It's just a sorting tool. The reality is we aren't doing anything this year than we have in the last few years, it's just that the public perception was top-25, top-50 and a win on the road -- we knew as a committee was more more important than a win at a home. A win neutral was more important than a win at home.
What I've done myself to break down the silo some is I've assigned a sliding point scale. So it's against 1 on the road is more valuable than against a 1 at home. There's no scientific backing to what I've done, but I've taken home, neutral and road and you look at them and I've assigned a value. So a 75 on the road has a value, but it's not the same as a 1 at home or a 15 at home. A win on the road against a team ranked 20th is worth more than against a team 30th.
I've done this the last two years: I've taken one column, which is the most points you can accumulate, OK? For how you do against a tournament-caliber team. That's a top-75 RPI. Why top-75? Because, typically, if your RPI is worse than 75, you're not getting in. There are a lot of them better than 75 that don't, but at least you're in the conversation. I assign a point value and I've got two columns: the team that's accumulated the most points, then I've got another that's the most points per opportunity. So you take a Gonzaga, that doesn't get enough opportunities, but what did they do with the opportunities they had?
(Editor's note: Last year the team that had the most points in column one in Rasmussen's system was North Carolina. The team that had the most points per opportunity was Gonzaga. They went on to meet in the national title game.)
Where you feel good about a team is where they're toward the top in both. But what happens is, sometimes you get tricked because, say you have a team from the ACC. They have 20 opportunities and four good wins. Out of 20! So they accumulate points, but points per opportunity isn't very good. That's where we have to have a discussion.
CBS Sports: That's interesting. So Middle Tennessee, if it didn't win the auto bid, is your system likely to help its case get closer inspection and longer discussion in the room?
BR: For me, I don't select teams based on my system, but I sort. I'll see teams with a lot of points but down in opportunity. What does that tell me? This team doesn't have a lot of points but they only had four opportunities. To me, I try to identify where you say, "OK, how do I describe this team?" For a Middle Tennessee or a Nevada or a Rhode Island, you could look at a number of those teams and see they've done a great job with the opportunities they've had. Also understand that it's a lot harder if you're in the ACC and you're playing murderer's row. The next five games are against tough teams as a opposed to: two weeks from now I have a great game. You've got to recognize those teams that have done a good job with the opportunities that have been presented. Gonzaga, Nevada, Rhode Island, Middle Tennessee I would put in that boat.
CBS Sports: The process of bracketing teams late on Sunday seems squeezed, sometimes leading to mis-seeding critiques. Sometimes it seems like conference tournament runs on Saturday and Sunday don't really impact seeding. John Calipari has complained about this. How much do these conference tournament semifinals and finals results affect seeding when squeezed right up against bracketing, and does the committee give itself enough time to bracket?
BR: There are 14 championship games on Saturday and five on Sunday. By then you're down to two teams in a conference. We have a lot of discussion on Saturday and Sunday about if this happens we go this way, if that happens we go a different way. Saturday and Sunday, depending upon the year, are very intense, very long. Last year was shorter because I think 21 conferences had their regular season champ win their conference tournament. The [top 16] didn't really change much last year from January to February to March, so we didn't have the upsets that really throw you off.
I think that there's a lot more discussion about what-if than people think. And so do I wish we had another day? Yeah I wish had another day. By the time we get to say Saturday night, we've got a lot of the teams that are finished in their slots. I understand the thought. I don't think it's as big of an issue as people think, but two years ago we had 12 brackets on Sunday at noon. Last year, two. I would be lying to tell you [two years ago] that we gave that as much time as it needed. It's impossible.
The podcast closes with a conversation about Penny Hardaway at Memphis
Our advanced computer model simulated Friday's Texas Tech vs. Purdue Sweet 16 game 10,000...
The road to the Final Four has been nearly paved, with 16 teams vying for the four precious...
Three teams were delayed on their way to the Sweet 16 on Wednesday due to bus issues
The bracket is a beautiful catastrophe -- so who comes after Villanova, Duke and Kansas?
Our advanced computer model simulated Friday's Syracuse vs. Duke Sweet 16 game 10,000 time...