On this week’s Legends vs. Computers, Gene Stallings and Bobby Ross asked some interesting questions of BCS computer pollster Richard Billingsley. Read the exchange below.
Stallings: Now, let me ask you a question just for my own benefit. When the computer poll ranks them, does the score have anything to do with the ranking? In other words, let’s say Oklahoma State beats somebody 20-7 or they beat somebody 70-7, does that affect the ranking?
Richard: No, Coach, it doesn’t and that’s an interesting point. The BCS removed the margin of victory beginning in the 2002 season. So, it really has no bearing on it at all which makes for some really interesting differences sometimes between the computer polls and the human polls.
Ross: Richard, I had one, too. Is it able to measure week to week injury situations?
Richard: No, it’s not and that’s a very strong contention of mine as to why it’s very appropriate to have a human poll in conjunction with a computer poll. Computer polls measure strength of schedule I believe more accurately than even possibly a human can because it goes to an ‘inth’ of a degree so to speak. But there’s no way that a computer can measure injuries or weather conditions or penalties at untimely points in the game and that’s why I think it’s very, very important to have a human input. People find it strange that I say I’ve never been in favor of a computer poll being the sole determining factor for a National Championship race because there are just too many intangibles that only a human can look at and make an accurate determination about.
Stallings: Richard, if you were going to pick the four top things that determine rank and I understand score is one of them and then the toughness of the schedule has to have something to do with it. But what are the other two things that have more bearing on the ranking than anything?
Richard: Well, I do think that just the straight up wins and losses are a factor. I don’t think it should be the determining factor because I don’t think all 5 and 2 teams are the same or all 4 and 3 teams are the same. But I also think the side of the game is important. I think it has some bearing on showing the strength of a team. I think if you’re playing on the road you deserve a little bit more credit and I also believe that specific venues should give you a little bit more of a bump in your ranking. Because I really do believe that playing at Tennessee or at Florida or at Oregon or at Alabama, I think that that is a little harder task than playing at Rice or at Kent State.
Stallings: That’s because of the teams not the place. You know, it’s a whole lot easier to play Rice anywhere than to play Alabama. So, what’s that got to do with it?
Richard: Well, I just think that it may be a little bit more difficult to play in front of 106,000 people screaming Rocky Top than it is playing – well, let’s give a better example perhaps – they’re both SEC venues, but I think it might be harder to play at Tennessee than it would be to play at Mississippi State or at Ole Miss.
Stallings: Well, you look at Tennessee they’re screaming that Rocky Top right now and they’re not beating anybody.
Richard: Well, it’s still hard to play there.
Ross: What I’m hearing you say then is the caliber of the opponent and the site are really kind of a combination of the two, right?
Richard: It is. I would say it is. And you know, Coach, it’s not just the venue itself as much as it is the capacity, I would think, that’s in the crowd. Because you can have very large stadiums, and I think Rice is a pretty good example of this, and Rice I mean they’ve had their day back in the ‘50’s they played really great football and a couple of years ago – 2 or 3 years ago – Rice had a pretty good football team, but they’ve got a 70,000 seat stadium. I lived in Houston for 25 years and went to every Rice home game.
Ross: I coached there.
Richard: But, you know, when you’ve got 15,000 people in a 70,000 seat stadium it just doesn’t have quite the impact playing there as it does, perhaps, playing at a sold out stadium in Tulsa.
Ross: When I coached there we’d look over on the other side at A&M and Arkansas and Texas and people like that would have far more many fans in our stadium than we would.
Richard: Yeah, some of the biggest crowds were probably the Arkansas, Texas, Texas A&M crowds who came in from out of town.
Mike Griffith (KnoxNews.com): So, Richard, am I to understand that the capacity at the game is part of your – I know all the computer formulas are different, but that’s something that factors into your computer formula?
Richard: It actually does, but you have to remember, guys, it’s such a very, very slight part of the formula. Every year at the end of the season I re-evaluate the NCAA attendance figures to see what the capacity of a stadium is in a particular season and make a determination at that point according to the capacity that they’re filling as to how many people are actually in the stands and are there to generate excitement and momentum for their home team. But now you have to remember this is a fraction of a point.
Stallings: I understand that can be a factor, but there’s something that’s got to weigh heavier than others. What are the heavy things that you put emphasis on?
Obviously score has got to be something, but what else? Where they play? Toughness of schedule? What are the top things? That’s my question.
Richard: The top factor, of course, is the strength of schedule and that weighs in probably 80-85% of a ratings goes into the strength of the opponent that a team is playing.
Stallings: Okay, that makes sense.
Richard: The other I would have to say would be the wins and losses. As I said earlier I don’t gauge it according to just wins and losses because all 5 and 2 teams are not the same.
Moderator: John Cooper mentioned on our call yesterday that even with Wisconsin with two losses, he still thinks they’re the best team in the Big 10. You’re the only one that has them rated. You’ve got them No. 20 in your poll. None of the other computer polls even have Wisconsin in their Top 25. There you’ve got that human factor looking at a team and he’s saying it’s the best team in the Big 10 and all the computers have totally dismissed them except yours.
Richard: Well, I think that Wisconsin and their record may be a pretty good example of what I’m talking about when you gauge teams if you’re looking strictly at wins and losses. Because when you look at a team like Wisconsin, Michigan State, teams that right now that have two losses, but you look at the conference they’re playing in and then you look at another group of teams like Louisiana Lafayette and Arkansas State and I don’t want to take anything away from their programs, but they have two losses. But they’re playing in the Sunbelt and I don’t think you can really make a comparison there and say that those teams are equal. I just don’t believe they’re equal.