clock menu more-arrow no yes mobile

Filed under:

Lost In the Numbers: How the Cowboys Fare in FPI

Everyone and their dog has a metric. ESPN's FPI is the new shiny toy of the group and for the most part, it is pretty accurate. There is still a lot of benefit of the doubt that is missing in all the measurements.

Mark J. Rebilas-USA TODAY Sports

From a local perspective, the expectations for the 2015-2016 Oklahoma State football season are relatively high. High, after a dramatic Bedlam win and strong bowl victory to cap off a season of frustration and at one point having readiness for major changes. Amazing what a couple of games can do.

Is it fair that the Cowboys have these high expectations after an overall, so-so year in 2014? Maybe not but the sample size of Mason Rudolph's career is so small that the optimism is going to be uncharacteristically high.

Nationally that optimism is not there. Where Cowboy fans believe they can compete for a conference title and maybe, just maybe squeak into the College Football Playoff, the national media says "not so fast my friend."

The Football Power Index is one of the new flavors of the month for the stat-driven folk. It uses 10,000 simulations of the season to measure the teams strength overall and how that equates into wins and losses along with being able to forecast things like conference championship chances with percentages.

Here's a look at how the Cowboys stack up in the FPI:

2015 Football Power Index

32 Oklahoma State, Big 12 0-0 7.5 - 4.5 0.1 3.0 43 10.8

There is a lot of benefit of the doubt missing in a metric like this. Just like how the Cowboy faithful are led to be more optimistic, this metric is made to not be realistic but slightly below that. Pessimistic.

That pessimism in numbers is shown in that first projection of wins and losses. The half game represented is that there are still a lot of questions to be answered about this football team. Will Mason Rudolph continue his success now that teams have had an entire offseason to prepare for him. A little different than what he did last year. He was a fresh face that was a incredibly pleasant surprise for the Cowboys and a nightmare for a team like OU. Also have to remember that with Tyreek Hill gone, that can also make up for that half game in the projection. He's on the team, win Bedlam, not on the team, likely don't win Bedlam. Pretty cut and dry stuff there.

Oklahoma State based on their schedule could do much better than 7.5 wins. By the TCU game, the team will have one loss, either to Texas or West Virginia on the road. If the Cowboys lose in Austin, that loss column could be at two by the time the Horned Frogs come to town.

Best case scenario, Cowboys finish 10-2, losing one of those aforementioned road games and Bedlam. Stringing consecutive Bedlam wins is not a strong suit for OSU. Let's not start acting like it is.

Worst case, 7-5. The Cowboys lose both Texas and WVU along with their three biggest home games, TCU, Baylor and OU.

Adjust the projected Win/Loss to 8.5-3.5 and that is a likely more accurate representation of the 2015 Cowboy season.

The other measurements are pretty self-explanatory. The Cowboys have a tenth of a percent to go undefeated. I don't think even the most optimistic Cowboy fans have qualms with that number.

Having such a low percentage to win the conference though goes back to the initial pessimism and general unknown about Oklahoma State. Will the offensive line pick up where it left off at the end of the season? Does the interior defensive line make an impact with such inexperience. Does play calling become a head scratcher again? I can see TCU before the odds on favorite, but the fact that Baylor is ahead of TCU in percentage is odd. Weak non-conference doesn't win you the conference. Baylor might have a better record than TCU, but would still finish second.

OSU should finish fourth in the conference but the conference race is a lot more wide open than the metric is allowing itself to be. It is not a two and a half team race. It is closer to a four team race.

Numbers and metrics are great, but they don't measure the benefit of the doubt.

Or punting twice.


Disagree with what you've read? Something missing from the analysis? Comment below.