Statistical Score Estimate vs. Marshall

[Ed. Note: Bumped from FanPosts for nerdiness.]

I've always been a total statistics nerd and I like translating it in sports, so I thought I'd try and give some the same sort of analysis for college football teams and Virginia Tech specifically.  I'm sure some other blog probably already has some analysis based on the same premise taken to a whole other level that blows mine away, but I'm a sucker for punishment so I'll share mine here.  I'll give it a shot if free time permits and if it doesn't totally suck for each game the rest of the year.

Details of the calculations after the jump.

Overview

I used the same premise that I gained from a baseball betting book which uses historical or forecasted stats to make power ratings and derive all of your adjusted calculations from there.  You take the league average as a power rating of 1 and adjust each team from there.  Example: if the league average was 200 yards per game and Virginia Tech gained 300 yards per game, then they would have a power rating of 1.5.  I took that basic premise and split it up into Rushing Offense, Passing Offense, Rushing Defense, and Passing Defense.

Once each team took a power rating in each category, I wanted to take a look at the quality of play of teams they've played so far.  If a team had a 1.25 in rushing offense, but only played 1.25 rushing defenses (defensive ratings are inverse as you obviously want under a 1 rating) all year then they really are just an average rushing team and wouldn't do as well against a good rushing defense.

Now, I wanted to take this one step further.  Say the team who had a 1.25 rushing offense faced a bad rushing defense, but that was actually because they faced a bunch of good running teams .  That actually helps the validity of the original team who had a 1.25 rushing offense.  So in the end, you're looking at the team, its opponents, and its opponents-opponents.

There are some things that I could have done to make this more practical and accurate, but I'm just too lazy.  Such as I'm too lazy to exclude FCS schools (they all earned a .8 rating in everything) and obviously if you're including an opponent's opponent, then one of the teams is the original team you're analyzing but I'm too lazy to exclude those too.

Game Calculation

Once you gain an adjusted power rating for each category, you can start to analyze how the teams would match up.  You multiply your offensive rushing power rating vs. the opponents defensive power rating and can determine the teams expected rushing yards.  Doing the same, you determine the expected yardage rushing and passing for both teams.  Then I determined each team's offensive and defensive scoring efficiency by simply dividing the number of points they have scored by their total yardage and then you can calculate the amount of points that each team will score.  Voila!

Hokie Offense

These are the Hokies offensive numbers versus an average college football defense.

Hokie Defense

These are the Hokies defensive numbers versus an average college football offense.

Marshall Game

Below are the yardage and point estimates based on all of those calculations.  These don't take into effect home/away advantage and obviously each field would be different from that.  Still all very immature, but I thought it'd be fun nonetheless.