Boomer
Is this you? Claim Profile »
1 week ago
in Saturday Night Rantings on Fanblogs.com
I think Alabama would still make the BCS championship if they fall to Auburn and beat Florida in the SEC title game. Voters will think that losing the Iron Bowl is not as bad as losing to Oregon State and USC doesn't have much strength of schedule in their remaining games.
2 replies
Ben Prather
If Alabama loses to Auburn then beats Florida it would be a Big 12 vs Big 12 match up. Alabama would be below the Texas Tech vs Oklahoma winner and Texas.
Kevin @ Fanblogs
Late season losses are killers - just ask Penn State.
1 month ago
in BCS Standings - Oct. 19th 2008 on Fanblogs.com
Do you have sources for this? I don't not believe you, just interested to see how the computers have done predicting bowl game results compared to the pollsters.
1 reply
Ben Prather
I did a detailed analysis back just prior to 2004, using 2003 data.
I found that the computers were less reliable than the pollsters at ranking the winning teams higher before the game. For programs offereing MOV and W-L only options the MOV algorithms performed better. The only computer that out performed the human polls was Massey's MOV algorithm, not used by the BCS.
I found that the computers were less reliable than the pollsters at ranking the winning teams higher before the game. For programs offereing MOV and W-L only options the MOV algorithms performed better. The only computer that out performed the human polls was Massey's MOV algorithm, not used by the BCS.
1 month ago
in BCS Standings - Oct. 19th 2008 on Fanblogs.com
Yeah, I don't get it. The whole point of including computers, and part of the justification for the entire BCS system, is to take power out of the hands of pollsters. But then when the computer results didn't align with the polls, people screamed "WHAT'S GOING ON????" and demanded a fix to the craaaazy situation--even though it was not only expected, but preferable. The solution was to marginalize the unbiased computers and give all the power back to an even worse group of pollsters.
I am generally in favor of the BCS over a playoff (especially a large playoff), but the way they have implemented it makes the system very hard to defend.
I am generally in favor of the BCS over a playoff (especially a large playoff), but the way they have implemented it makes the system very hard to defend.
1 reply
Ben Prather
If the computers could out perform the voters in predicting the outcome of the bowls they could be trusted more to determine the outcome. They are not to that level yet.
1 month ago
in BCS Standings - Oct. 19th 2008 on Fanblogs.com
The only BCS dynamic that really matters is polling. The pollsters will decide who goes and who doesn't. They will override the computers by artificially inflating or lowering a team whose BCS standing is deemed "wrong" and will respond to sleazy lobbying by coaches and the media. Those numbers might be accurate, but in the end, people will make all the decisions.
1 reply
Tommy_Trojan
Skip:
Bravo! ding,ding,ding...You've nailed it! Actual play on the field doesn't have anything to do with it. Truthfully, I was not so upset back in 2003 (even though USC was ranked # 1 in both polls), when USC was passed over for an LSU/Oklahoma game. I am quite sure, that the BCS system as it was, was alot better than what they have now. At that time, the system, gave equal weight to "the voters" and "the computers". The computers overrode what the voters believed (probably based on SOS) and left USC out. It was probably correct.
Now, "the system", gives about 2/3 of the weight to "the voters". We now have the same system as we had prior to the invention of the BCS system. It is obviously very difficult for the computers to overtake the voters. Therefore, Auburn was burned in 2004, as the voters decided that USC should play Oklahoma. I don't know, if the actual computer data, would have gotten Auburn into one of the top two spots - but it would not surprise me at all.
So, I'm not gonna cry too hard about a system that was overhauled, because of what was deemed my teams slight. But, I do know that it did hurt the whole system. The idea was to get rid of "the voter bias", by using actual computer models to determine actual play on the field, but the whole thing has been discarded for what the media's popular opinion is.
Bravo! ding,ding,ding...You've nailed it! Actual play on the field doesn't have anything to do with it. Truthfully, I was not so upset back in 2003 (even though USC was ranked # 1 in both polls), when USC was passed over for an LSU/Oklahoma game. I am quite sure, that the BCS system as it was, was alot better than what they have now. At that time, the system, gave equal weight to "the voters" and "the computers". The computers overrode what the voters believed (probably based on SOS) and left USC out. It was probably correct.
Now, "the system", gives about 2/3 of the weight to "the voters". We now have the same system as we had prior to the invention of the BCS system. It is obviously very difficult for the computers to overtake the voters. Therefore, Auburn was burned in 2004, as the voters decided that USC should play Oklahoma. I don't know, if the actual computer data, would have gotten Auburn into one of the top two spots - but it would not surprise me at all.
So, I'm not gonna cry too hard about a system that was overhauled, because of what was deemed my teams slight. But, I do know that it did hurt the whole system. The idea was to get rid of "the voter bias", by using actual computer models to determine actual play on the field, but the whole thing has been discarded for what the media's popular opinion is.
4 months ago
in Conference honchos say no-way to playoffs on Fanblogs.com
Can we at least all agree that a 16-team playoff would be bad for college football?
1 reply
Football Picks
Compared to what? To the current system, no. Compared to an 8 team playoff, perhaps. But I really don't see that a 16-team playoff is bad for college football.