College Football season is about to kick off. Here are our record projections for every team

[Rolltydr]

What? I could throw darts and come up with better rankings.

If anyone's interested in seeing a ranking system based on performance, not speculation, they should check out the Colley Matrix rankings -- an unbiased computer based ranking system that only takes results into account. Strength of schedule is inherently "baked in" to the rankings, as a team gets more ranking points for beating other highly ranked teams and fewer points for beating lower ranked teams. Similarly, a loss to a "good" team isn't as penal to ranking as a loss to a "bad" team. Kind of how anyone with any common sense and/or analytical ability would do it. Algorithm was developed by a PhD in astrophysics who also happens to be a huge college football buff.

Imo, there is no such thing as an “unbiased computer based” ranking system. And, yes, I know about AI and all the things computers can do. I spent 35 years in IT. However, it still boils down to what information a human feeds the computer. Unless the hunan is totally unbiased, extremely rare, the results still are skewed by human bias. The closest way to getting an unbiased result, is from viewing a number of different polls and average them out. That way, the inherent biases will somewhat even out. Also, I’ve never met an unbiased college football buff.

 

[Elan]

Imo, there is no such thing as an “unbiased computer based” ranking system. And, yes, I know about AI and all the things computers can do. I spent 35 years in IT. However, it still boils down to what information a human feeds the computer. Unless the hunan is totally unbiased, extremely rare, the results still are skewed by human bias. The closest way to getting an unbiased result, is from viewing a number of different polls and average them out. That way, the inherent biases will somewhat even out. Also, I’ve never met an unbiased college football buff.

Of course the algorithm can be biased. But, Colley makes no attempt to predict or speculate on outcomes, which is somewhat unique. All teams are rated identically at the beginning of the season and only the game results dictate their movement up or down. The matrix of results is recalculated every week. One can actually go through the season, week by week, and see how a given team moves up and down the rankings. There's no input for margin of victory, or "game control" or other subjective BS. Colley Matrix was one of the computer algorithms used in the BCS era.

There's no perfect "answer", but I find Colley to be the best way to check my biases (and those of others ).

Edited to add: I'm not sure how anyone can be fearful of a computer algorithm being biased and simultaneously advocate for a "consensus of opinions", i.e polls, approach. Talk about biased! Now, if one wants to opt for a consensus of multiple computer rankings, I'd be much more likely to get on board.

 

[Rolltydr]

Of course the algorithm can be biased. But, Colley makes no attempt to predict or speculate on outcomes, which is somewhat unique. All teams are rated identically at the beginning of the season and only the game results dictate their movement up or down. The matrix of results is recalculated every week. One can actually go through the season, week by week, and see how a given team moves up and down the rankings. There's no input for margin of victory, or "game control" or other subjective BS. Colley Matrix was one of the computer algorithms used in the BCS era.

There's no perfect "answer", but I find Colley to be the best way to check my biases (and those of others ).

Edited to add: I'm not sure how anyone can be fearful of a computer algorithm being biased and simultaneously advocate for a "consensus of opinions", i.e polls, approach. Talk about biased! Now, if one wants to opt for a consensus of multiple computer rankings, I'd be much more likely to get on board.

Fearful? It’s football polls, nothing to fear. As I stated, tongue in cheek, in post #174, I’m glad to use whatever polls support my biases. It just always irks me a little when someone suggests computers aren’t or can’t be biased. I believe they are inherently biased because they can only do what humans tell them to do. I’m not bashing them. I love them! I’m enjoying a nice retirement largely due to my former profession.

 

[Elan]

Fearful? It’s football polls, nothing to fear. As I stated, tongue in cheek, in post #174, I’m glad to use whatever polls support my biases. It just always irks me a little when someone suggests computers aren’t or can’t be biased. I believe they are inherently biased because they can only do what humans tell them to do. I’m not bashing them. I love them! I’m enjoying a nice retirement largely due to my former profession.

I wasn't necessarily advocating for "computers" in general. I was advocating more specifically for Colley's algorithm, because it's only input is the game results, W or L. I am not a fan of some of the other computer algorithms, such as Sagarin, because they attempt to incorporate a predictor of future games, presumably for sports betting purposes. FWIW, Alabama would make the field using the Colley rankings. He has them at #11, the highest ranked 3 loss team. I'm fine with that as it also seems reasonable.

BTW, I also "retired off computers". 35+ years in semiconductors as a logic design engineer, the last 20+ concurrently writing logic verification software to find and fix errors made by other logic experts.

 

[Rolltydr]

I wasn't necessarily advocating for "computers" in general. I was advocating more specifically for Colley's algorithm, because it's only input is the game results, W or L. I am not a fan of some of the other computer algorithms, such as Sagarin, because they attempt to incorporate a predictor of future games, presumably for sports betting purposes. FWIW, Alabama would make the field using the Colley rankings. He has them at #11, the highest ranked 3 loss team. I'm fine with that as it also seems reasonable.

BTW, I also "retired off computers". 35+ years in semiconductors as a logic design engineer, the last 20+ concurrently writing logic verification software to find and fix errors made by other logic experts.

You were much deeper in the weeds than I was. I was more in the hardware and networking installation and support area most of my career. I worked in the IT department for one of the largest insurance companies in the U.S. I finished the last 10 years or so selecting equipment and vendors we could work with to support our catastrophe response teams and then writing and communicating the process and procedures for implementation. Early in my career 70’s and early 80’s, I dabbled in a little programming, but as you know, that’s a totally different world today.

 

[Brett]

Jimmy Madison wins a bowl !
https://www.msn.com/en-us/sports/ot...-first-bowl-win-in-school-history/ar-AA1w8t4x

 

[pedro47]

Jimmy Madison wins a bowl !
https://www.msn.com/en-us/sports/ot...-first-bowl-win-in-school-history/ar-AA1w8t4x

That was a great win for an up and coming football team from the Commonwealth of Virginia. IMHO.

Look out University of Virginia.

 

[TheHolleys87]

Alabama vs. Notre Dame would have been a much better game.

 

[TUGBrian]

after that game most would agree that anyone vs ND would have been a better game than the one tonight. (though id also bet most people would have rather watched indiana play oklahoma instead of alabama a few weeks back) however I still believe Indiana deserved a spot in the top 12 based on their season.

I have a bigger issue with some of the byes given out than I do the teams that made it.

 

[pedro47]

IU scheduled was very weak compared to Alabama.
IU overall schedule and players were not a playoff team and especially playing ND at ND….LOL

 

[Elan]

after that game most would agree that anyone vs ND would have been a better game than the one tonight. (though id also bet most people would have rather watched indiana play oklahoma instead of alabama a few weeks back) however I still believe Indiana deserved a spot in the top 12 based on their season.

I have a bigger issue with some of the byes given out than I do the teams that made it.

I agree, but the committee played themselves with the bye protocol. They were so sure the B10/SEC/ACC/B12 would be the 4 highest ranked conference champions. I'd be shocked if it doesn't change going forward.
The other aspect is will it be all that advantageous to have a bye? 3 weeks without a game playing a team that just played. We'll see...

 

[Rolltydr]

I agree, but the committee played themselves with the bye protocol.

I hope so. Boise State getting a 3 seed is laughable. The #1 seed, Oregon, actually has the hardest path to the championship. It doesn’t make any sense. The basketball committee does a great job seeding 68 teams. The CFP committee needs to sit with them in March and learn how to do it!

I know I’ve already lost the argument on the size of the playoff, but… Indiana was untested due to their schedule, losing badly to the only ranked team they played. SMU, Boise St. and Arizona St. also had very weak schedules and I expect will meet the same fate as IU. Yes, Alabama beat Georgia and two other ranked teams. They also lost to Vandy and were completely embarrassed by a bad Oklahoma team. Ole Miss lost 3 games, including one to a bad Kentucky team. I don’t think any of them should be in a championship playoff and the fact they are just shows 12 is at least four teams too many. It is going to give us more games, but it isn’t going to give us better games.