Dullard's Final 2003 College Football Rankings

[Tom]

"let's see, you asked if a team that went 10-3 won HALF their games and you wonder about MY response?? "

I was being sarcastic, seeing as you have them ranked as the 4th best team which seems rediculous to me. But since it's based on statistics, not the real world, I suppose it's just a reflection of the computer programmer's design choices.

 

[royaldank]

Thanks for sharing this season Dullard. Hope you do so again next year. It's been fun.

A lot of people agree with you - that only your games against the top X teams should count. That is a valid argument and I cannot rebutt it with a good argument. My only argument is a philosophical questipn: why bother playing those games if they don't count? If you do throw away all the games against easier teams, then I too agree that point margin isn't too important. But if you do consider the whole season, all ~14 games, then point margin MUST be used. Beating Army by 1 point is not the same as beating Michigan State by 14 points. You would have thrown both of those games out. I say if you can only beat the last place team by 1 point, then you deserve to be harmed in the ratings. If you choose to play easy teams, then you must blow them away or you aren't a good team.

I'd love a playoff system. USC vs LSU would be a great matchup. If you ask me, LSU would win by a field goal.[/quote]

I see where you are coming from. I guess it is a matter of differing theories. I tend to think any team can have a bad game. Beating Army by 1 is a bad day, sure enough. However, I don't think overall it matters all that much. As long as you aren't beating teams by 1 point every game, then I don't think it should hurt you. Heck, Ohio State is living proof that you can beat the crappiest of teams by 1 point and still be ranked incredibly high. Seriously, though, I think after a point margin of 7-10 is established, the bottom games don't really matter. They are played because there is potenial for upset. But, as long as they are wins they shouldn't be weighted too much.

We are arguing over something that could be solved by 1 more game. I think we'll both agree to that.

I think you did answer my next question, though. Do you think your rankings are accurate? Just wondering if you are defending them from a "this is true and how it is" stance or "this is what my system has and here's why." I figured it was the second one since the first is no better than the current crappy human polls. Props to sticking to your formula and not tweaking it to your liking (ie. getting LSU to finish first).

So, you started thinking about any adjustments you're going to make? Are you happy with USC finishing first?

 

[PlatinumGold]

Originally posted by: Dead Parrot Sketch
"let's see, you asked if a team that went 10-3 won HALF their games and you wonder about MY response?? "

I was being sarcastic, seeing as you have them ranked as the 4th best team which seems rediculous to me. But since it's based on statistics, not the real world, I suppose it's just a reflection of the computer programmer's design choices.

🙂

as these are DULLARDS rankings and i'm not dullard, i'm not sure what your point is. 🙂

 

[Tom]

Sorry, I misread your post as being from Dullard, anyway I was being sarcastic about Georgia's winning half their games. Just seemed like every time I heard about Georgia, they had lost again ! Kind of like VA Tech ! :evil:

 

[Red Dawn]

Originally posted by: dullard

Originally posted by: royaldank
I guess we agree to disagree. I don't see anything outside of the top 30 or so teams being very important, especially if you've beaten 5 top 30 teams. They were supposed to win those games, and both teams won those games. Point margin isn't all it's cracked up to be and one of the reasons why the human polls are so slanted. I still say that viewing the entire season, LSU had a better season. They beat GA twice whose #4. They beat #3. Granted, USC didn't get to play them, but I don't think that matters. USC can say they tried to schedule good teams, but in the end, they didn't play as good of a schedule and came up second. Most people just can't see how bad the human polls are when ranking teams or understand how slanted and biased they are towards public opinion. I'd have to say the BCS got the right teams and we had a heck of a game. The BCS was put in place as a tie-breaker between 3 tied teams. It was decided that rather than take a popularity contest, we'd run a bunch of mathematical equations and take that instead of whomever ESPN likes that week. Oh well. That is the wonder of college football.

Thanks for sharing this season Dullard. Hope you do so again next year. It's been fun.

A lot of people agree with you - that only your games against the top X teams should count. That is a valid argument and I cannot rebutt it with a good argument. My only argument is a philosophical questipn: why bother playing those games if they don't count? If you do throw away all the games against easier teams, then I too agree that point margin isn't too important. But if you do consider the whole season, all ~14 games, then point margin MUST be used. Beating Army by 1 point is not the same as beating Michigan State by 14 points. You would have thrown both of those games out. I say if you can only beat the last place team by 1 point, then you deserve to be harmed in the ratings. If you choose to play easy teams, then you must blow them away or you aren't a good team.

I'd love a playoff system. USC vs LSU would be a great matchup. If you ask me, LSU would win by a field goal.

Nah LSU would lose because they'd commit too many stupid penalties. I also think their mediocre QB would throw 2 or 3 picks!

 

[dullard]

Originally posted by: royaldank
I think you did answer my next question, though. Do you think your rankings are accurate? Just wondering if you are defending them from a "this is true and how it is" stance or "this is what my system has and here's why." I figured it was the second one since the first is no better than the current crappy human polls. Props to sticking to your formula and not tweaking it to your liking (ie. getting LSU to finish first).

So, you started thinking about any adjustments you're going to make? Are you happy with USC finishing first?

I'm a pessimist and a realist. No ranking system is perfect. Nor can any system be perfect (when you have team X beat Y, Y beat Z, Z beat X then it is impossible to rank them all ahead of each other). My program is meant to be a predictor of future games. It predicts USC will beat LSU by 0.9 points. But as a realist, I know my error margin is about 11-12 points. So realistically it predicts USC to win by 13 all the way down to LSU winning by 11 - anything in that range would be acceptable to me and the program. Thus I look at the ratings as saying there is a tie between USC and LSU.

They are the best predictor that I have come up with using the data I feed it. My next step is to make things non-linear. At the moment you get X points for each win and lose Y points for each loss. I want to see what happens if I make those X and Y functions instead of constants. My previous (meager) attempts at doing this provided no better score predictions. But that doesn't mean that a really well thought out function won't provide a better prediction. I want to narrow the average point error from 11-12 points down to 9-10 points. Maybe I'll do that, maybe it cannot be done.

One adjustment I'd love to consider is using more data. What about using turnover margin or yards per game, etc? The biggest drawback so far is how to get that data for free each week.

 

[Nitemare]

Originally posted by: dullard

Originally posted by: Nitemare
Still think LSU should get the nod since their conference is 10 times better than the Pac-10

That may be true in the past, but this year I think the Pac-10 did quite well. If I counted right they were 1 point away from being 5/1 in the bowl games (and a 6 win 7 loss UCLA would never have been allowed to play a bowl in previous years due to the more wins than losses rule).

I never took the time to program conference tracking into my program but Sagarin did. Here is his average conference rating:
ACC: 79.11
SEC: 78.67
Pac-10: 76.79
Big-10: 76.44
Big-12: 75.61
Big East: 73.86
Mountain west: 72.14
Mid-American: 64.68
CUSA: 64.16

The Pac-10 is right up there this year.

Sagarin is saying that the ACC had the hardest conference this year???

 

[erikistired]

Originally posted by: Red Dawn

Originally posted by: dullard

Originally posted by: royaldank
I guess we agree to disagree. I don't see anything outside of the top 30 or so teams being very important, especially if you've beaten 5 top 30 teams. They were supposed to win those games, and both teams won those games. Point margin isn't all it's cracked up to be and one of the reasons why the human polls are so slanted. I still say that viewing the entire season, LSU had a better season. They beat GA twice whose #4. They beat #3. Granted, USC didn't get to play them, but I don't think that matters. USC can say they tried to schedule good teams, but in the end, they didn't play as good of a schedule and came up second. Most people just can't see how bad the human polls are when ranking teams or understand how slanted and biased they are towards public opinion. I'd have to say the BCS got the right teams and we had a heck of a game. The BCS was put in place as a tie-breaker between 3 tied teams. It was decided that rather than take a popularity contest, we'd run a bunch of mathematical equations and take that instead of whomever ESPN likes that week. Oh well. That is the wonder of college football.

Thanks for sharing this season Dullard. Hope you do so again next year. It's been fun.

A lot of people agree with you - that only your games against the top X teams should count. That is a valid argument and I cannot rebutt it with a good argument. My only argument is a philosophical questipn: why bother playing those games if they don't count? If you do throw away all the games against easier teams, then I too agree that point margin isn't too important. But if you do consider the whole season, all ~14 games, then point margin MUST be used. Beating Army by 1 point is not the same as beating Michigan State by 14 points. You would have thrown both of those games out. I say if you can only beat the last place team by 1 point, then you deserve to be harmed in the ratings. If you choose to play easy teams, then you must blow them away or you aren't a good team.

I'd love a playoff system. USC vs LSU would be a great matchup. If you ask me, LSU would win by a field goal.

Nah LSU would lose because they'd commit too many stupid penalties. I also think their mediocre QB would throw 2 or 3 picks!

people who know absolutely nothing about college football shouldn't join college football discussions. of course that hasn't stopped you from running off at the mouth concerning about a hundred other topics, so i guess it's to be expected.

 

[Tom]

"people who know absolutely nothing about college football shouldn't join college football discussions. "

there would be nobody to talk to.

 

[Red Dawn]

Originally posted by: fisher

Originally posted by: Red Dawn

Originally posted by: dullard

Originally posted by: royaldank
I guess we agree to disagree. I don't see anything outside of the top 30 or so teams being very important, especially if you've beaten 5 top 30 teams. They were supposed to win those games, and both teams won those games. Point margin isn't all it's cracked up to be and one of the reasons why the human polls are so slanted. I still say that viewing the entire season, LSU had a better season. They beat GA twice whose #4. They beat #3. Granted, USC didn't get to play them, but I don't think that matters. USC can say they tried to schedule good teams, but in the end, they didn't play as good of a schedule and came up second. Most people just can't see how bad the human polls are when ranking teams or understand how slanted and biased they are towards public opinion. I'd have to say the BCS got the right teams and we had a heck of a game. The BCS was put in place as a tie-breaker between 3 tied teams. It was decided that rather than take a popularity contest, we'd run a bunch of mathematical equations and take that instead of whomever ESPN likes that week. Oh well. That is the wonder of college football.

Thanks for sharing this season Dullard. Hope you do so again next year. It's been fun.

A lot of people agree with you - that only your games against the top X teams should count. That is a valid argument and I cannot rebutt it with a good argument. My only argument is a philosophical questipn: why bother playing those games if they don't count? If you do throw away all the games against easier teams, then I too agree that point margin isn't too important. But if you do consider the whole season, all ~14 games, then point margin MUST be used. Beating Army by 1 point is not the same as beating Michigan State by 14 points. You would have thrown both of those games out. I say if you can only beat the last place team by 1 point, then you deserve to be harmed in the ratings. If you choose to play easy teams, then you must blow them away or you aren't a good team.

I'd love a playoff system. USC vs LSU would be a great matchup. If you ask me, LSU would win by a field goal.

Nah LSU would lose because they'd commit too many stupid penalties. I also think their mediocre QB would throw 2 or 3 picks!

people who know absolutely nothing about college football shouldn't join college football discussions. of course that hasn't stopped you from running off at the mouth concerning about a hundred other topics, so i guess it's to be expected.

Just because you suck dick doesn't make you an expert at relationships. I guess your fanboism blinded you last night to the crappy play by LSU's Offense.

 

[dullard]

Originally posted by: Nitemare
Sagarin is saying that the ACC had the hardest conference this year???

While it isn't always the best measure, the ACC had the most bowl wins (of any conference) and the fewest bowl losses (of conferences with 3 or more bowl games). That is if I counted correctly. Is there any place online that kept track?

 

[gistech1978]

Originally posted by: Red Dawn

Originally posted by: fisher

Originally posted by: Red Dawn

Originally posted by: dullard

Originally posted by: royaldank
I guess we agree to disagree. I don't see anything outside of the top 30 or so teams being very important, especially if you've beaten 5 top 30 teams. They were supposed to win those games, and both teams won those games. Point margin isn't all it's cracked up to be and one of the reasons why the human polls are so slanted. I still say that viewing the entire season, LSU had a better season. They beat GA twice whose #4. They beat #3. Granted, USC didn't get to play them, but I don't think that matters. USC can say they tried to schedule good teams, but in the end, they didn't play as good of a schedule and came up second. Most people just can't see how bad the human polls are when ranking teams or understand how slanted and biased they are towards public opinion. I'd have to say the BCS got the right teams and we had a heck of a game. The BCS was put in place as a tie-breaker between 3 tied teams. It was decided that rather than take a popularity contest, we'd run a bunch of mathematical equations and take that instead of whomever ESPN likes that week. Oh well. That is the wonder of college football.

Thanks for sharing this season Dullard. Hope you do so again next year. It's been fun.

A lot of people agree with you - that only your games against the top X teams should count. That is a valid argument and I cannot rebutt it with a good argument. My only argument is a philosophical questipn: why bother playing those games if they don't count? If you do throw away all the games against easier teams, then I too agree that point margin isn't too important. But if you do consider the whole season, all ~14 games, then point margin MUST be used. Beating Army by 1 point is not the same as beating Michigan State by 14 points. You would have thrown both of those games out. I say if you can only beat the last place team by 1 point, then you deserve to be harmed in the ratings. If you choose to play easy teams, then you must blow them away or you aren't a good team.

I'd love a playoff system. USC vs LSU would be a great matchup. If you ask me, LSU would win by a field goal.

Nah LSU would lose because they'd commit too many stupid penalties. I also think their mediocre QB would throw 2 or 3 picks!

people who know absolutely nothing about college football shouldn't join college football discussions. of course that hasn't stopped you from running off at the mouth concerning about a hundred other topics, so i guess it's to be expected.

Just because you suck dick doesn't make you an expert at relationships. I guess your fanboism blinded you last night to the crappy play by LSU's Offense.

so OU's defense had nothing to do with it? right...
who is blinded or biased now?

 

[erikistired]

Originally posted by: Red Dawn

Originally posted by: fisher

Originally posted by: Red Dawn

Originally posted by: dullard

Originally posted by: royaldank
I guess we agree to disagree. I don't see anything outside of the top 30 or so teams being very important, especially if you've beaten 5 top 30 teams. They were supposed to win those games, and both teams won those games. Point margin isn't all it's cracked up to be and one of the reasons why the human polls are so slanted. I still say that viewing the entire season, LSU had a better season. They beat GA twice whose #4. They beat #3. Granted, USC didn't get to play them, but I don't think that matters. USC can say they tried to schedule good teams, but in the end, they didn't play as good of a schedule and came up second. Most people just can't see how bad the human polls are when ranking teams or understand how slanted and biased they are towards public opinion. I'd have to say the BCS got the right teams and we had a heck of a game. The BCS was put in place as a tie-breaker between 3 tied teams. It was decided that rather than take a popularity contest, we'd run a bunch of mathematical equations and take that instead of whomever ESPN likes that week. Oh well. That is the wonder of college football.

Thanks for sharing this season Dullard. Hope you do so again next year. It's been fun.

A lot of people agree with you - that only your games against the top X teams should count. That is a valid argument and I cannot rebutt it with a good argument. My only argument is a philosophical questipn: why bother playing those games if they don't count? If you do throw away all the games against easier teams, then I too agree that point margin isn't too important. But if you do consider the whole season, all ~14 games, then point margin MUST be used. Beating Army by 1 point is not the same as beating Michigan State by 14 points. You would have thrown both of those games out. I say if you can only beat the last place team by 1 point, then you deserve to be harmed in the ratings. If you choose to play easy teams, then you must blow them away or you aren't a good team.

I'd love a playoff system. USC vs LSU would be a great matchup. If you ask me, LSU would win by a field goal.

Nah LSU would lose because they'd commit too many stupid penalties. I also think their mediocre QB would throw 2 or 3 picks!

people who know absolutely nothing about college football shouldn't join college football discussions. of course that hasn't stopped you from running off at the mouth concerning about a hundred other topics, so i guess it's to be expected.

Just because you suck dick doesn't make you an expert at relationships. I guess your fanboism blinded you last night to the crappy play by LSU's Offense.

now that's a topic i'm sure you're an expert on. 😀

however, my fanboism would have blinded me if i was an lsu fan, which i'm not. thanks for playing tho.

lsu played a GOOD defense last night, something that usc saw maybe once all season. but then, having your head up your ass probably blinded you to that.

 

[erikistired]

Originally posted by: dullard

Originally posted by: Nitemare
Sagarin is saying that the ACC had the hardest conference this year???

While it isn't always the best measure, the ACC had the most bowl wins (of any conference) and the fewest bowl losses (of conferences with 3 or more bowl games). That is if I counted correctly. Is there any place online that kept track?

check out yahoo sports, they list bowl games by conference if you look in the right place (unfortunately you have to dig for that info sometimes).

SEC bowl games:

wins:
auburn over wisconsin
arkansas over missouri
georgia over perdue
mississippi over oklahoma st (even tho yahoo sports still has the scores backwards)
lsu over oklahoma

losses:
iowa over florida
clemson over tennessee

ACC bowl games:

wins:
nc state over kansas
virginia over pittsburgh
maryland over west va
clemson over tennessee
georgia tech over tulsa

losses:
miami over florida state

same amount wins, one more loss for the sec. the only head to head was an acc win, not sure if that factors into it. does who is playing in the games make a difference?

 

[Mill]

Originally posted by: Red Dawn

Originally posted by: fisher

Originally posted by: Red Dawn

Originally posted by: dullard

Originally posted by: royaldank
I guess we agree to disagree. I don't see anything outside of the top 30 or so teams being very important, especially if you've beaten 5 top 30 teams. They were supposed to win those games, and both teams won those games. Point margin isn't all it's cracked up to be and one of the reasons why the human polls are so slanted. I still say that viewing the entire season, LSU had a better season. They beat GA twice whose #4. They beat #3. Granted, USC didn't get to play them, but I don't think that matters. USC can say they tried to schedule good teams, but in the end, they didn't play as good of a schedule and came up second. Most people just can't see how bad the human polls are when ranking teams or understand how slanted and biased they are towards public opinion. I'd have to say the BCS got the right teams and we had a heck of a game. The BCS was put in place as a tie-breaker between 3 tied teams. It was decided that rather than take a popularity contest, we'd run a bunch of mathematical equations and take that instead of whomever ESPN likes that week. Oh well. That is the wonder of college football.

Thanks for sharing this season Dullard. Hope you do so again next year. It's been fun.

A lot of people agree with you - that only your games against the top X teams should count. That is a valid argument and I cannot rebutt it with a good argument. My only argument is a philosophical questipn: why bother playing those games if they don't count? If you do throw away all the games against easier teams, then I too agree that point margin isn't too important. But if you do consider the whole season, all ~14 games, then point margin MUST be used. Beating Army by 1 point is not the same as beating Michigan State by 14 points. You would have thrown both of those games out. I say if you can only beat the last place team by 1 point, then you deserve to be harmed in the ratings. If you choose to play easy teams, then you must blow them away or you aren't a good team.

I'd love a playoff system. USC vs LSU would be a great matchup. If you ask me, LSU would win by a field goal.

Nah LSU would lose because they'd commit too many stupid penalties. I also think their mediocre QB would throw 2 or 3 picks!

people who know absolutely nothing about college football shouldn't join college football discussions. of course that hasn't stopped you from running off at the mouth concerning about a hundred other topics, so i guess it's to be expected.

Just because you suck dick doesn't make you an expert at relationships. I guess your fanboism blinded you last night to the crappy play by LSU's Offense.

When in doubt call the poster a homo!

Seriously, I've yet to see any facts or stats from you RD. Just rhetoric.

 

[Red Dawn]

Originally posted by: Mill



When in doubt call the poster a homo!

You mean she is a he?

Seriously, I've yet to see any facts or stats from you RD. Just rhetoric.

I'm basing my opinion on what I witnesseed in the two bowl games. Fsck stats, if they were so perfect then OU wouldn't have looked like ass last night! Dropped passes, wrong routes run, missed blocking assignment galore from both teams.

 

[Mill]

Originally posted by: Red Dawn

Originally posted by: Mill



When in doubt call the poster a homo!

You mean she is a he?

Seriously, I've yet to see any facts or stats from you RD. Just rhetoric.

I'm basing my opinion on what I witnesseed in the two bowl games. Fsck stats, if they were so perfect then OU wouldn't have looked like ass last night! Dropped passes, wrong routes run, missed blocking assignment galore from both teams.

I think Erik Bush is a fairly masculine name. Anyway, dropped passes, wrong routes run, missed blocking from both teams is the results of heavy pressure and great D on both sides. The offense fizzles under a good D. Didn't you see all the mistakes Michigan made vs USC? It was because USC suffocated Navarre and made their offense make numerous crappy mistakes.

 

[Red Dawn]

Originally posted by: Mill

Originally posted by: Red Dawn

Originally posted by: Mill



When in doubt call the poster a homo!

You mean she is a he?

Seriously, I've yet to see any facts or stats from you RD. Just rhetoric.

I'm basing my opinion on what I witnesseed in the two bowl games. Fsck stats, if they were so perfect then OU wouldn't have looked like ass last night! Dropped passes, wrong routes run, missed blocking assignment galore from both teams.

I think Erik Bush is a fairly masculine name. Anyway, dropped passes, wrong routes run, missed blocking from both teams is the results of heavy pressure and great D on both sides. The offense fizzles under a good D. Didn't you see all the mistakes Michigan made vs USC? It was because USC suffocated Navarre and made their offense make numerous crappy mistakes.

You know maybe if it was a regular season game both teams might have performed better but to me it seemed that neither was adequately prepared for last nights game. That's just my opinion from observing the game. In the past Championship games at least one team fired on all cylinders, last night neither team did.

 

[Mill]

Originally posted by: Red Dawn

Originally posted by: Mill

Originally posted by: Red Dawn

Originally posted by: Mill



When in doubt call the poster a homo!

You mean she is a he?

Seriously, I've yet to see any facts or stats from you RD. Just rhetoric.

I'm basing my opinion on what I witnesseed in the two bowl games. Fsck stats, if they were so perfect then OU wouldn't have looked like ass last night! Dropped passes, wrong routes run, missed blocking assignment galore from both teams.

I think Erik Bush is a fairly masculine name. Anyway, dropped passes, wrong routes run, missed blocking from both teams is the results of heavy pressure and great D on both sides. The offense fizzles under a good D. Didn't you see all the mistakes Michigan made vs USC? It was because USC suffocated Navarre and made their offense make numerous crappy mistakes.

You know maybe if it was a regular season game both teams might have performed better but to me it seemed that neither was adequately prepared for last nights game. That's just my opinion from observing the game. In the past Championship games at least one team fired on all cylinders, last night neither team did.

Honestly, I think LSU and OU were scared of each other and USC wasn't scared of Michigan at all. That might explain some of it.