Revised Similarity Scores: Pitt

The team similarity scores have been revised thanks to some input from Sam. As always if you have any suggestions feel free to let me know in the comment section. The new methodology is detailed at the end of this post. The team I'm going to examine today is a team that both I and Phil Steele believed to be a Big East sleeper in the preseason: the Pitt Panthers.

The most similar teams to Pittsburgh (2006) from last season--similarity score in parentheses and final record following

1. Michigan State (861.2) 5-6
2. Oregon (839.4) 10-2
3. Nebraska (802.2) 8-4
4. Kansas State (793.2) 5-6
5. Minnesota (684.4) 7-5
6. Colorado (681.6) 7-6

Michigan State may be the most similar team, but I wouldn't expect Pitt to end the season with a losing record. For all the absurd supernatural and endemic qualities sportswriters tend to assign to teams, I think they are on to something in the case of the Spartans. Michigan State seems to just not be able to rebound from tough defeats whether the coach is John Smith, Bobby Williams, or Nick Saban. I don't think they are bound for Oregon near-BCS territory (not with both Louisville and West Virginia left on the schedule), but I think Nebraska is a pretty good comp. A regular season finish of somewhere between 7-5 and 9-3 depending on how well they play on the road seems about right.

Here's the methodology.

1. Start with 1000 points

2. Through 'x' number of games take the difference in winning percentage multiply by 1000 and subtract from 1000example: Team A is 4-0 and Team B 3-1, then the difference in winning percentage would be 1-.75=.25, multiplying this by 1000=250, subtract this number from 1000

3. Multiply the difference in points scored per game by 3 and subtract this amount (similar teams should score similar amounts)

4. Multiply the difference in points allowed per game by 3 and subtract this amount (similar teams should allow a similar amount of points)

5. Multiply the difference in average opponents' Sagarin Rating (I think its a pretty good measure of schedule strength) by 2 and subtract this amount

6. Subtract the difference multiplied by 1000 in previous year's record (we need to know how good the team's were in the previous season)

7. Subtract the difference multiplied by 1000 in previous year's Pythagorean Winning Percentage (a better indicator of team strength than actual record)

8. The remaining points are the teams' similarity score (the higher the better)

The home/road inequality has been eliminated in order to increase the sample of possible similar teams. I think similar record is more important than similar schedule when prospecting forward.