Wednesday, January 2, 2008

2007 Playoff Predictions (Full Bracket)

As my prediction system is going to be picking all the home teams from this point out, I thought I'd go through each round and pick games using the system's predictions, reasoning based on past research, and maybe a smidgen of personal bias. (No, the tattoo is German for "The Brady, The.")


WILD CARD ROUND








Predictions based on unadjusted stats
GamePredicted Final Score MarginP(Home Team Wins)
JAX @ PIT4.552865.8528
TEN @ SD5.198966.2033
NYG @ TB4.850767.0733
WAS @ SEA1.859757.9650


This year, my prediction system using unadjusted stats was wrong more often than not when the projected probability of the home team winning was between 50 and 60%. Therefore, I'm picking Seattle to lose to Washington.
When the project probability of the home team winning was between 60 and 70%, the system was correct 2 out of 3 times. There are 3 such games in the wild card round. Since Pittsburgh's starting running back is out and they have the lowest probability of winning of the three, I'm picking Jacksonville to beat Pittsburgh.
San Diego and Tampa Bay should win their games as predicted.

DIVISIONAL ROUND







Predictions based on opponent-adjusted stats
GamePredicted Final Score MarginP(Home Team Wins)
JAX @ NE5.210565.7171
SD @ IND7.574774.7456
WAS @ DAL4.612966.0576
TB @ GB2.250857.1654


Surprisingly, Green Bay is not seen as a very heavy favorite against Tampa Bay. The weather might tilt things more heavily in Green Bay's favor. But Tampa Bay is seen as the better team overall, despite their 9-7 record (a GB@TB game is given a 61% probability of a Bucs win). So that's the one upset I'd pick.


CONFERENCE CHAMPIONSHIP






Predictions based on opponent-adjusted stats
GamePredicted Final Score MarginP(Home Team Wins)
IND @ NE4.210063.4600
TB @ DAL4.536064.1551


In this article, I looked into the relative importance of defense in the postseason. In short, teams with good offenses tend to make the playoffs, so having a good defense in addition will get you deeper into the playoffs. The Patriots' pass offense has looked ridiculously good, but they are ranked 14th in rush offense VOLA and 26th in rush defense VOLA. Indy is ranked 21st and 7th, in the same categories respectively. Rush defense has shown to be of more importance in winning the Super Bowl than rush offense (though Baltimore and Tampa Bay might be skewing the sample there). Pass defense is essential, and they are both about equally good. Indy is 4th and New England is 5th in unadjusted pass defense VOLA. Indy's pass offense has also been hampered by injuries, weighing down their numbers. New England has the weather advantage over the dome team, but I'll go with the slightly better balanced Colts upsetting the Pats, leaving them at 17-1.

Dallas is simply a much better team than the Bucs.

SUPER BOWL





Predictions based on opponent-adjusted stats
GamePredicted Final Score MarginP(Home Team Wins)
DAL @ IND4.945865.7789
IND @ DAL3.729863.1363


Indy is seen as a stronger team. Given the neutral site, I'm picking the Colts to repeat as champions.

3 comments:

Doug Walters said...

Amen, Derek

JP said...

You have JAX and SD winning the wildcard round. In that case, JAX goes on to play NE, and SD plays IND. In your table this is backward.

Derek said...

Fixed. Thanks, JP.