So, we're literally favored to win in all of our remaining games. Wow!
So, we're literally favored to win in all of our remaining games. Wow!
@ UAB will be a toughie, as will @ Marshall. Homefield provides an 8-point advantage. That's a lot to overcome, even if you're the better team. I like our chances better at UAB than at Marshall, even if UAB is better than MU. They whipped us last year in Ruston, and Holtz seems able to turn the tables on teams that do that.
But, heck, just win 'em all!
HFA is less than 3 points.
Everything I have always read said HFA was worth an average of 3pts. I have read some articles lately since metrics is so intertwined in athletics, that several of the HFA formulas factor in stadium size, average home crowd etc. one mathematician even went as far as to break it down over a 10 year period for FBS pee team and for most teams it was between 1.8 to 3.2 pts.
The stats game has evolved so much in the last 10 or so years.
There is a website, I had the link...thought I posted it....that had EVERY college football game ever played in its database. Not a representative sample, but 100% of that population. And, at that time, home teams won by an average of 8.2 points. I would think that would still be the case, or maybe, since the site is active and dynamic, it might be 8.1 or maybe 8.3. I just use "8" since you can't score .2 points.
If that is "crap" to you, challenge that website's operatives. But I'm not sure what your counter-argument is gonna be. Maybe: facts be damned! Y'all must be like the libtards of the world. Facts mean nothing to you.
Found it! I had it bookmarked.
Latest from that site. They updated their process, limiting the data to "the last 10,000 CFB games played." And they now limited it to just FBS schools. There is no mention how or if there is any different treatment of FCS schools' history when they move up to FBS. Still, given all of that the new HFA is 7.897. Still about 8 points.
Another site uses "How does a team fair on the road vs. at home." Using that methodology, guess who has the highest HFA at 5.5 points. You got it! Hawaii does. But that methodology is flawed, IMO, since it skews toward the quality of the program rather than the unbiased factor of homefield.
Have to keep in mind that ALL such algorithms are flawed because, in CFB, there is unequal scheduling. The Bamas of the world get to play 75-100% of their non-conference games at home where they usually slaughter someone. But, this site uses "the last 10,000 FBS games played" in a floating average. Such biases, as the Bamas, should be smoothed out given the sheer volume of the database.
In the last 10,000 FBS games played the home team has won 61% of the time. When the road team wins (39%) the scores are considerably closer, on average. Home teams win close games too, but there are far more blowouts administered by the home side. The site shows that, in those last 10,000 games played, the home team outscored visitors by an average of 7.897 points per game. This is fact.
Like any method...computer-generated process...the designers can program in biases, intended or not. You can call this site "crap" if you want, but funny, I don't see you offering a method that is as all-encompassing as a 100% of the population. Other sites I have looked at admit they skew the data based on "quality of the program," "size of the stadium," "location of the stadium (geography)," and many other factors. You might say, yeah, such factors should be considered. But in each case some "human" bias pollutes the data. It becomes a matter of personal opinion.
This site simply lets the facts speak for themselves. No introduced biases.
8 points it is.
You are not applying statistics correctly regarding odds for betting purposes, which is what we are discussing. You are using a 10,000 sample size to evaluate one game.
Except that is not the HFA applied by bettors. It appears they are using results of games to get an avg score differential.
Wrong! HFA for Vegas lines is around 3 points in favor of the home team.
Historically, home teams win 61% of the time by an average margin of about 8 pts. But, how does that apply to a specific game...one you may be planning to bet on? Well, it provides an over-view of how HFA plays into that game. But, other factors can have a larger influence on the game. If a team loses its all-American QB the week before the game, that can affect the outcome to a greater extent than where the game is played. So too can motivation. A team eliminated from contention, from the league title, from a bowl, whatever, might "mail in" their effort and their opponent might be fighting for something. You bettors have to look at those other factors. Does Vegas? I don't know. Because, I don't bet on 18-22 year-olds.
Since this thread is about Sagarin ratings, the Sagarin Predictor homefield advantage hasn't been over 3 points since the ratings after games of Week 5 2015. And with that home advantage of under 3 points it has still correctly predicted game winners at over a 76% rate over that time.