ESPN College Football - Computer Ranking - Top 10

LOL  trash

Wow, what a stupid list. 

lol

GIGO

Nate Silver keeps falling further and further in my estimation /s

Okay, so the most mind boggling ranking is actually TCU at 10.

Let’s see if I can be open minded about this:

I have season tickets to the Longhorns and it appears they are ranked that high because they were close to being pretty good. They couldn’t get it done, but points for effort I guess? I’m guessing star power is also a factor here.

Same for Alabama. Star power and they were close to being undefeated. Let’s just ignore those close wins that were almost losses.

Then you look at TCU. No credit for going undefeated. No credit for getting to a championship game. No credit for beating #6 TX on the road. You can’t give exceptions to all the teams above TCU and then not do the same here. This makes no sense at all.

OSU over UM is laughable.

I agree with the poster who said they wanted to look at the code to find out what they rank as important or not.

I bet since ESPN has vested interest in SEC, a win over an SEC team counts as 1.5.

Edit: I completely missed USC not in the top 10, but Utah at 9 lol.

Algo:

SEC = B1G + x$(Saban slurp ratio)

OSU = UM * (10 year Points history OSU/UM)

FPI gets updated every week.  None of this is new.  It's a similar ranking to S&P+ and they do explain it in multiple places but here's the basic explanation:

"Each team’s FPI rating is composed of a predicted offensive, defensive and special teams component. These ratings represent the number of points each unit is expected to contribute to the team's net scoring margin on a neutral field against an average FBS opponent.

In the preseason, these components are made up entirely of data from previous seasons, such as returning starters, past performance, recruiting rankings and coaching tenure (more on the preseason component below). That information allows FPI to make predictions (and make determinations on the strength of a team’s opponents) beginning in Week 1, and then it declines in weight as the season progresses. It is important to note that prior seasons’ information never completely disappears, because it has been proved to help with prediction accuracy even at the end of a season. Vegas similarly includes priors when setting its lines."

As a statistical model it is by definition unbiased.  I'm not sure if it takes data in at the play level (like S&P+) or the drive level (like FEI) or more broadly on based on efficiencies (I suspect this is the case) but like any other statistical model it just says, what did you do against the teams on your schedule (ALL the teams) compared to what other teams did against those teams and how does that, along with some other meaningful data like recruiting rankings and coaching history, inform what you might do in the future.

What this is suggesting is that if Michigan played OSU again on a neutral field right now, OSU would still be favored by about a point (and that would be the case in Vegas which tends to make lines very close to these predictive metrics).  That's a few points less than the model said before The Game so that's a lot of movement.  But what the model is saying is that one game does not necessarily outweigh all the other data which has proven to be also predictive.

The one part of it that is human is the recruiting factor which plugs in ESPNs recruiting rankings but even those have been proven to be pretty accurate predictors of future performance.

Another important thing to keep in mind is that these are predictive rankings and NOT at all resume based.  ESPNs Strength of Record is their resume metric and the CFP tends to closely align with that metric. Like for example TCU is number 10 in FPI but number 1 in SoR and they will of course be ranked top 3 by the committee.

They obviously don't put much into this predictive FPI ranking.

And no USC at all.  This thing is a fraud.

I'm out.

Well this is what we get for ditching E-SEC-PN and heading to FOX/NBC/CBS. 

I won't be too surprised if, in the coming years, you can't even catch highlights of Big Ten games on Sports Center anymore, or if Gameday just stops showing up to any Big Ten location. Maybe scores are mentioned in passing on Sports Center, but if the Big Ten isn't going to give them any games to cover, I'm betting they in turn will do what they can to not even mention the Big Ten as an entity that even exists. 

Aside from putting Michigan below OSU, the fact that this has Texas and Clemson in the top 10 is laughable. 

Alabama is a 4 loss team. The refs bailed them out of 2 of them, but they are not a good football team this year. 

We're all focused on UM-OSU but Texas at 6 is some kind of crazy.

Texas as 6 should be enough to pull the plug on this computer.

Wow.  They need some new computers.  They using an Apple IIC there or what?

LMAO!!!!  ESPN couldn't be any more bold than this!

So Michigan, with wins over #3 and #7 (neither of which was close), is behind 3 teams with one win each among the 'top ten'.

Clearly they should stick to sportsball reporting because they're terrible at maths and logic.  They do not deserve clicks.

This is why we don't have the BCS anymore

I always knew I was smarter than computers.

Yeah. It would be interesting to see that algorithm. Not sure how that happens. 

Texas in front of Penn State. Yeah ok. Dont even need to be a fan of a team to see the crap that is going on here.

Bama/Tennessee/Texas (outside of each other ) losses

LSU, Texas tech, Oklahoma st, Tcu, Georgia and South Carolina by 25 points

UM/OSU/Penn state (outside of each other) losses

............

Think about that and the circle jerk those computers must be doing. 

The ESPN “algorithm” is just a transcript of Gary Danielson’s snore-lab from Saturday night. 

Wow, is that for real?? I'm feeling incredulous as my go to reaction as well.

This is why advanced metrics are dumb with small sample sizes. This isn’t baseball and the teams you play against have the largest influence on ratings/ rankings.