Tuesday, February 19, 2008

Big East -- Pythagoras at the 70.3% Mark

208 of the 288 games are, as of Sunday, in the books. The current standings (though the teams are in various stages of completing their Big East regular season schedules) find Louisville and Georgetown on top, with a little separation between them and Connecticut & Notre Dame for the other bye seeds. If the season were to end today, Providence, Rutgers, St. John's and South Florida would stay home. But with nearly 3 weeks left and having either 2 or 4 wins each, none have lost contact with the three schools immediately above them (DePaul, Syracuse and Villanova) in the standings. The standings as of Sunday:

Notre Dame930.750
West Virginia750.583
De Paul570.417
Seton Hall580.385
St. John's490.308
South Florida2110.154

I applied the Pythagorean formula to each team's "points for" and "points against" (raised to the power of 9.6, per Pomeroy) in Big East games only. The results are certainly not authoritative, but interesting when compared, each team to every other Big East team, and when looking at the remaining part each team's schedule. I ran the numbers about a month ago and the top 4 projected finishers were Louisville, West Virginia, Georgetown and Pittsburgh. They were separated by at least a game from the next two teams (Connecticut and Providence). The bottom 4 teams were in order Cincinnati, St. John's, South Florida and Rutgers. The separation between teams #14-#16 teams and #13 (Cincinnati) was about 2 games (wins). Cincinnati was really clustered with 2 other teams, Notre Dame and Seton Hall, projected to win between 7 & 8 games. The projection today:

West Virginia12.45.60.691
Notre Dame11.07.00.609
De Paul6.511.50.363
Seton Hall6.111.90.339
South Florida4.713.30.260
St. John's4.113.90.226

Pomeroy (log5) vs. Pythagoras
This second look spotted a few trends that appear to have held up...
1. Connecticut started out at 3-3, the projection had the Huskies winning about 11 games this season. Looks like they are pretty much on track to do that.
2. The initial projection predicted that Rutgers, St. John's and South Florida would struggle to the point that there would be some separation between them and the rest of the conference. Despite several unexpected wins for Rutgers (and a slump for Providence...) those numbers have largely held up. The current projection suggests the trend will play out to the end of the season, and there will be a "gap" between those three and the rest of the conference.
3. Though Louisville stumbled out of the gate again this season, and played the first 5 games at a 0.600 pace, the projection suggested they would finish with a record closer to 0.750. The Cardinals are currently playing at that 0.750 level. The projection now is for them to improve yet again and finish the season with a 15-3 0.833 pace.

A few trends/projections to track as the season winds up...
Louisville -- for the Cardinals to finish at 15-3 they will have to run the table. That means sweeping Syracuse (H), Pitt (A), Notre Dame (H), Villanova (H) and Georgetown (A). In the case of Georgetown by the way, that would mean the Cards took both games this season. Current projections from Pomeroy's Schedule Page suggest Louisville will lose one of those games (the GTown away game) and finish 14-4.
Georgetown -- according to the projection the Hoyas will go 3-2 over their last 5 games. They have Providence (A), Cincinnati (H), St. John's (H), Marquette (A) and Louisville (H) remaining. Pomeroy's Schedule Page predicts a 4-1 record over the home stretch, the loss coming to Marquette. The Hoyas have gone 1-2 over the last 3 games, dropping decisions to Syracuse and Louisville (both road games), and beating Villanova on the last possession of the game. While I am inclined to favor Pomeroy's project over Pythagoras, a 3-2 finish would not surprise me.
Pittsburgh -- will have to go 5-1 to meet Pythagoras' expectations. Mindful that their schedule includes road games at Notre Dame, Syracuse and West Virginia, and home games with Louisville, Cincinnati and DePaul, 5-1 might be a reach. In a too hot/too cold contrast Pomeroy's projection has the Panthers stumbling 2-4 to close out the regular season. I suspect their finish will be somewhere in between, with the next 2 games for Pitt giving us a clue on which projection is closer. Go 2-0 through this stretch and the Panthers have to feel good about finishing with a 3-3 at worst, and more likely 4-2 (or even 5-1).
Connecticut -- according to this projection will finish the home stretch with a 2-4 record, slumping to an 11-7 finish. With a schedule that includes a home game (DePaul) followed by a 2 game road trip (Villanova and Rutgers), and then finish out with a home-away-home sequence (West Virginia, Providence and Cincinnati respectively), I see maybe 2 losses (maybe...). Pomeroy's Schedule Page predicts a six game winning streak to finish out the season. I think that is especially optimistic, given that the Huskies' current run is up to 9 games. That would mean the Huskies would take a 15 game run to finish the regular season. Do that and the Huskies are definitely back.
Notre Dame -- would, according to Pythagoras, take (like UConn...) a 2-4 header to finish the season. The Irish love home cooking; they are 6-0 so far, with Pittsburgh and Syracuse the next 2 coming into South Bend. They then take to the road (a far less hospitable place where they are currently 3-3 in Big East play) for a games at Louisville and DePaul. They then close out with a home-away sequence with the Johnnies and the Bulls. Pomeroy's Schedule Page, in sharp contrast to Pythagoras, predicts a 5-1 finish. The markers for who is right may come in those next two games, both at home. If the Irish can go 2-0 against Pitt and Syracuse I cannot see them dropping three of their last four.
West Virginia -- has perhaps been the most unpredictable team going into the season. And they have maintained that reputation going into the last 2 weeks. Pythagoras suggests the 'Eers will go 5-1 to close out the season, a run also predicted by Pomeroy's Schedule Page. Four of those games are, however, road games, and WVU is 2-3 on the road so far. The Villanova game, up next for both teams, will be a good test for these projections.
Villanova -- is currently 5-7. Pythagoras' projection is for the 'Cats to have a 3-3 finish, and to head into the BET with an 18-12 record. Pomeroy's Schedule Page by contrast predicts a 1-5 finish (2-4 if you check the predicted record) with the only win against South Florida. With a 3 game homestand coming up, West Virginia, Connecticut and Marquette, anything north of 0-3 would make Pythagoras look better than Pomeroy.

No comments: