atomicblue224
Well-Known Member
I wouldn't worry about them, their systems of linear equations all screwed up ! They got y axis and x axis all screwed up. They're failing big time!
ESPN College Football Power Index
- Thread starter dirtwrap
- Start date
The Index is bogus, but I think this is the answer to your question, at least to the extent ESPN is letting on:
http://espn.go.com/blog/statsinfo/post/_/id/109828/reintroducing-espns-college-football-power-index
PhoenixHawk
Well-Known Member
Sagarin doesn't fully explain his formula. He admits using the hugely flawed 'margin of victory' element, but how does that compare a team that beats a team by 20 pts, 60-40, vs. a team that beat someone by only 10 pts, 20-10? Notice that the team that won by 10 actually doubled the other teams score, while the team that won by 20 only outscored them by 50%? Which is then a better win? Outscoring by 100%, or by 20 pts? He also uses last year's performance at the start to put his finger on the scale to bias for his favorites. This biases his results in the same way traditional 'eyeball test' polling is biased.
Espn's FPI is pure voodoo garbage. Remember, Esecpn is hosting the playoffs! They want ratings, and believe a B12 high-octane O in the playoffs will be a huge ratings boost! They want those Texas TV sets tuned in! FPI is about what they want to get their ratings and profits up.
Espn's FPI is pure voodoo garbage. Remember, Esecpn is hosting the playoffs! They want ratings, and believe a B12 high-octane O in the playoffs will be a huge ratings boost! They want those Texas TV sets tuned in! FPI is about what they want to get their ratings and profits up.
HuckFinn
Well-Known Member
I am actually fascinated by the whole media thing for one simple reason: America LOVES an under dog. And, Iowa is the classic underdog. Why, then, are we the most hated team in the nation by the media? We should, given all past events, be the media darling, the under dog, the cinderella team, with glowing reviews about the most meaningless of our accomplishments. We so often see this in college basketball, and frankly, we used to see it in football as well. Oh, my, what is this world coming to when only the brand names are eligible for media love?
BlindDogScooter
Well-Known Member
I have absolutely no problem whatsoever ignoring it. The people who produce the only poll that matters (CFP) don't seem to be influenced by it at all.
Feel free to ignore it... really... you can do it!
Thank you! This is what I was looking for!
After reading those articles, I am aware now of what the FPI is and what it is not.
1. First of all, it is a predictive tool that seems to really only be useful before the season to predict all games. The inputs as I was stating were based off of past years success, coaching tenure, returning starters and recruiting rankings. We know recruiting rankings are very subjective, as well as the assumption that longer tenured coaches are more successful and that more returning starters is always better. Most of the time those are the case, but many exceptions happen to that rule. So for that reason it is a good tool for preseason magazines only, because...
2. The backbone of the rating is the expected points added (EPA). Reading between the lines it is attempt to evaluate each of the team's units on how well they impose their will or do what they set out to do. There seems to be a lot to it and that is fine, however, the main flaw is that it was developed for the NFL. The problem with using an NFL metric like this for college is that it does not account for tempo, it is a total of the points accumulated by all three units. So basically, if two teams are identically , but one is higher tempo than the other and runs twice as many plays, their FPI will be much higher due solely to this.
This explains why the Big 12 teams are so high in the FPI. And why we are getting torched in that metric because everything goes against us. Alabama would be much the same way, however, they get weighted for recruiting and past success.
Thats all I needed to know. Thanks again.
1. First of all, it is a predictive tool that seems to really only be useful before the season to predict all games. The inputs as I was stating were based off of past years success, coaching tenure, returning starters and recruiting rankings. We know recruiting rankings are very subjective, as well as the assumption that longer tenured coaches are more successful and that more returning starters is always better. Most of the time those are the case, but many exceptions happen to that rule. So for that reason it is a good tool for preseason magazines only, because...
2. The backbone of the rating is the expected points added (EPA). Reading between the lines it is attempt to evaluate each of the team's units on how well they impose their will or do what they set out to do. There seems to be a lot to it and that is fine, however, the main flaw is that it was developed for the NFL. The problem with using an NFL metric like this for college is that it does not account for tempo, it is a total of the points accumulated by all three units. So basically, if two teams are identically , but one is higher tempo than the other and runs twice as many plays, their FPI will be much higher due solely to this.
This explains why the Big 12 teams are so high in the FPI. And why we are getting torched in that metric because everything goes against us. Alabama would be much the same way, however, they get weighted for recruiting and past success.
Thats all I needed to know. Thanks again.
I really don't want to ignore it, I would rather understand it. I enjoy watching ESPN and their football coverage because they do a very good job of adding hype and drama to the sport. Thats what I love about sports.
