Tuesday, September 29, 2009

Isaacson-Tarbell Postulate 2009 Predictions: Week 4

DET (1-2-0) @ CHI (2-1-0) --> CHI 
TB (0-3-0)  @ WAS (1-2-0) --> WAS 
OAK (1-2-0) @ HOU (1-2-0) --> HOU 
SEA (1-2-0) @ IND (3-0-0) --> IND 
TEN (0-3-0) @ JAC (1-2-0) --> JAC 
BAL (3-0-0) @ NE (2-1-0) --> BAL 
NYG (3-0-0) @ KC (0-3-0) --> NYG 
CIN (2-1-0) @ CLE (0-3-0) --> CIN 
BUF (1-2-0) @ MIA (0-3-0) --> BUF 
NYJ (3-0-0) @ NO (3-0-0) --> NO 
DAL (2-1-0) @ DEN (3-0-0) --> DEN 
STL (0-3-0) @ SF (2-1-0) --> SF 
SD (2-1-0)  @ PIT (1-2-0) --> SD 
GB (2-1-0)  @ MIN (3-0-0) --> MIN 

Isaacson-Tarbell Postulate 2009 Results: Week 3

A much more resectable 75% this week.


CLE (0-2-0) @ BAL (2-0-0)   Pred: BAL  Act: BAL +
TEN (0-2-0) @ NYJ (2-0-0)   Pred: NYJ  Act: NYJ +
NYG (2-0-0) @ TB  (0-2-0)   Pred: NYG  Act: NYG +
GB  (1-1-0) @ STL (0-2-0)   Pred: GB   Act: GB +
WAS (1-1-0) @ DET (0-2-0)   Pred: WAS  Act: DET -
JAC (0-2-0) @ HOU (1-1-0)   Pred: HOU  Act: JAC -
SF  (2-0-0) @ MIN (2-0-0)   Pred: MIN  Act: MIN +
KC  (0-2-0) @ PHI (1-1-0)   Pred: PHI  Act: PHI +
ATL (2-0-0) @ NE  (1-1-0)   Pred: ATL  Act: NE -
NO  (2-0-0) @ BUF (1-1-0)   Pred: NO   Act: NO +
CHI (1-1-0) @ SEA (1-1-0)   Pred: SEA  Act: CHI -
PIT (1-1-0) @ CIN (1-1-0)   Pred: CIN  Act: CIN +
MIA (0-2-0) @ SD  (1-1-0)   Pred: SD   Act: SD +
DEN (2-0-0) @ OAK (1-1-0)   Pred: DEN  Act: DEN +
IND (2-0-0) @ ARI (1-1-0)   Pred: IND  Act: IND +
CAR (0-2-0) @ DAL (1-1-0)   Pred: DAL  Act: DAL +





Wk Corr  Wk Inc  Wk Pct  Cum Corr  Cum Inc  Cum Pct
     12  4   75.0%        29       19   60.4%




Tuesday, September 22, 2009

Isaacson-Tarbell Postulate Predictions for 2009 Week 3

CLE (0-2-0) @ BAL (2-0-0)   Pred: BAL  
TEN (0-2-0) @ NYJ (2-0-0)   Pred: NYJ  
NYG (2-0-0) @ TB (0-2-0)    Pred: NYG  
GB (1-1-0)  @ STL (0-2-0)   Pred: GB  
WAS (1-1-0) @ DET (0-2-0)   Pred: WAS  
JAC (0-2-0) @ HOU (1-1-0)   Pred: HOU  
SF (2-0-0)  @ MIN (2-0-0)   Pred: MIN  
KC (0-2-0)  @ PHI (1-1-0)   Pred: PHI  
ATL (2-0-0) @ NE (1-1-0)    Pred: ATL  
NO (2-0-0)  @ BUF (1-1-0)   Pred: NO  
CHI (1-1-0) @ SEA (1-1-0)   Pred: SEA  
PIT (1-1-0) @ CIN (1-1-0)   Pred: CIN  
MIA (0-2-0) @ SD (1-1-0)    Pred: SD  
DEN (2-0-0) @ OAK (1-1-0)   Pred: DEN  
IND (2-0-0) @ ARI (1-1-0)   Pred: IND  
CAR (0-2-0) @ DAL (1-1-0)   Pred: DAL  

Isaacson-Tarbell Postulate 2009 Results: Week 2

Week 2 is a tough one..... we're just .500 this week.

CAR (0-1-0) @ ATL (1-0-0)   Pred: ATL  Act: ATL +
MIN (1-0-0) @ DET (0-1-0)   Pred: MIN  Act: MIN +
CIN (0-1-0) @ GB (1-0-0)    Pred: GB   Act: CIN -
ARI (0-1-0) @ JAC (0-1-0)   Pred: JAC  Act: ARI -
OAK (0-1-0) @ KC (0-1-0)    Pred: KC   Act: OAK -
NE (1-0-0)  @ NYJ (1-0-0)   Pred: NYJ  Act: NYJ +
NO (1-0-0)  @ PHI (1-0-0)   Pred: PHI  Act: NO -
HOU (0-1-0) @ TEN (0-1-0)   Pred: TEN  Act: HOU -
STL (0-1-0) @ WAS (0-1-0)   Pred: WAS  Act: WAS +
TB (0-1-0)  @ BUF (0-1-0)   Pred: BUF  Act: BUF +
SEA (1-0-0) @ SF (1-0-0)    Pred: SF   Act: SF +
PIT (1-0-0) @ CHI (0-1-0)   Pred: PIT  Act: CHI -
CLE (0-1-0) @ DEN (1-0-0)   Pred: DEN  Act: DEN +
BAL (1-0-0) @ SD (1-0-0)    Pred: SD   Act: BAL -
NYG (1-0-0) @ DAL (1-0-0)   Pred: DAL  Act: NYG -
IND (1-0-0) @ MIA (0-1-0)   Pred: IND  Act: IND +



Wk Corr Wk Inc  Wk Pct  Cum Corr  Cum Inc  Cum Pct
      8  8 50.0%        17  15  53.1%





Tuesday, September 15, 2009

Isaacson-Tarbell Postulate Predictions for 2009 Week 2

CAR (0-1-0) @ ATL (1-0-0)   Pred: ATL  
MIN (1-0-0) @ DET (0-1-0)   Pred: MIN  
CIN (0-1-0) @ GB  (1-0-0)   Pred: GB  
ARI (0-1-0) @ JAC (0-1-0)   Pred: JAC  
OAK (0-1-0) @ KC  (0-1-0)   Pred: KC  
NE  (1-0-0) @ NYJ (1-0-0)   Pred: NYJ  
NO  (1-0-0) @ PHI (1-0-0)   Pred: PHI  
HOU (0-1-0) @ TEN (0-1-0)   Pred: TEN  
STL (0-1-0) @ WAS (0-1-0)   Pred: WAS  
TB  (0-1-0) @ BUF (0-1-0)   Pred: BUF  
SEA (1-0-0) @ SF  (1-0-0)   Pred: SF  
PIT (1-0-0) @ CHI (0-1-0)   Pred: PIT  
CLE (0-1-0) @ DEN (1-0-0)   Pred: DEN  
BAL (1-0-0) @ SD  (1-0-0)   Pred: SD  
NYG (1-0-0) @ DAL (1-0-0)   Pred: DAL  
IND (1-0-0) @ MIA (0-1-0)   Pred: IND  

Isaacson-Tarbell Postulate 2009 Results: Week 1

Wk+ Wk- WkPct Cum+ Cum- CumPct
 9 7 56.3% 9 7 56.3%



TEN (0-0-0) @ PIT (0-0-0)   Pred: PIT  Act: PIT +
MIA (0-0-0) @ ATL (0-0-0)   Pred: ATL  Act: ATL +
KC  (0-0-0) @ BAL (0-0-0)   Pred: BAL  Act: BAL +
PHI (0-0-0) @ CAR (0-0-0)   Pred: CAR  Act: PHI -
DEN (0-0-0) @ CIN (0-0-0)   Pred: CIN  Act: DEN -
MIN (0-0-0) @ CLE (0-0-0)   Pred: CLE  Act: MIN -
NYJ (0-0-0) @ HOU (0-0-0)   Pred: HOU  Act: NYJ -
JAC (0-0-0) @ IND (0-0-0)   Pred: IND  Act: IND +
DET (0-0-0) @ NO  (0-0-0)   Pred: NO   Act: NO  +
DAL (0-0-0) @ TB  (0-0-0)   Pred: TB   Act: DAL -
SF  (0-0-0) @ ARI (0-0-0)   Pred: ARI  Act: SF  -
WAS (0-0-0) @ NYG (0-0-0)   Pred: NYG  Act: NYG +
STL (0-0-0) @ SEA (0-0-0)   Pred: SEA  Act: SEA +
CHI (0-0-0) @ GB  (0-0-0)   Pred: GB   Act: GB  +
BUF (0-0-0) @ NE  (0-0-0)   Pred: NE   Act: NE  +
SD  (0-0-0) @ OAK (0-0-0)   Pred: OAK  Act: SD  -

Saturday, September 5, 2009

Isaacson-Tarbell Postulate Predictions for 2009 Week 1

Week 1  -- Clean slates, home teams all to win!
Results posted Tuesday, September 15, 2009

TEN (0-0-0) @ PIT (0-0-0) --> PIT
MIA (0-0-0) @ ATL (0-0-0) --> ATL
KC (0-0-0) @ BAL (0-0-0) --> BAL
PHI (0-0-0) @ CAR (0-0-0) --> CAR
DEN (0-0-0) @ CIN (0-0-0) --> CIN
MIN (0-0-0) @ CLE (0-0-0) --> CLE
NYJ (0-0-0) @ HOU (0-0-0) --> HOU
JAC (0-0-0) @ IND (0-0-0) --> IND
DET (0-0-0) @ NO (0-0-0) --> NO
DAL (0-0-0) @ TB (0-0-0) --> TB
SF (0-0-0) @ ARI (0-0-0) --> ARI
WAS (0-0-0) @ NYG (0-0-0) --> NYG
STL (0-0-0) @ SEA (0-0-0) --> SEA
CHI (0-0-0) @ GB (0-0-0) --> GB
BUF (0-0-0) @ NE (0-0-0) --> NE
SD (0-0-0) @ OAK (0-0-0) --> OAK

The Isaacson-Tarbell Postulate Garners Attention

As of today a Google search for "isaacso-tarbell postulate" yields 376 hits, which I found both startling and delightful.  Here are a few of my favorites:
This one from 2008 shows that historically, the Isaacson-Tarbell postulate correctly predicts the winning team 62.5% of the time
This entry in a discussion forum at the website of Beatpaths ("automatic graphical power-ranking system based off of only wins, losses, and who beat who").  Here's another entry from the same site. 
This prediction of the 2009 season records claims to use the Isaacson-Tarbell postulate to predict the final records of all 32 teams.  I don't if it starts from scratch or uses last season's records as a starting point, but it's good news if you're a fan of the Ravens or Saints, but not so good for the Cardinals, Bengals (we could have guessed that anyway!), or Bucs.

About the Isaacson-Tarbell Postulate

I have been a longtime reader of Gregg Easterbook's weekly Tuesday Morning Quarterback column.  ONe of Easterbrook's recurring features is a summary of the often dubious success rate of the many people who publish predictions of each week's NFL games.  This led him to propose this brain-free formula: "Home team wins."  As he notes, this requires no knowledge of any kind except which teams are playing and where the game is being held.  Given a modest home-team-advantage edge (put at 3 points by Jeff Sagarin), this formula should be right a little better than half the time over the season, and in fact it was in 2006 (53%) and 2007 (57%).  During the 2006 season, it occurred to me that this formula ignored the fact that, despite the home field advantage, it was more likely that the better team would win.  How do we know the better team?  Well, on average the better team will win more games.  So I started tracking this simple two-part formula: The team with the better record wins; if the records are the same, home team wins.

Easterbrook first published this prediction in his Bad Predictions Roundup column in Feb. 2007, noting a success rate of 56%.

Then in a 2007 pre-season column, Eastbrook wrote: "This year, I will endorse the generic forecast advocated in the offseason by about 20 readers, pulling from a hat the name Catey Tarbell of Kirkland, Wash. Catey's Law: Best Record Wins -- Unless Records Equal, Then Home Team Wins. TMQ will track this off-price house-label generic forecast against the predictions of full-time football experts who possess incredible insider information."  As any good academic does, I recognized immediately the plagiarism of my own work, fired off a note to Easterbrook, who kindly corrected the record in the next day's reader-response column:

"Tuesday I wrote, 'This year I will endorse the generic forecast advocated in the offseason by about 20 readers, pulling from a hat the name Catey Tarbell of Kirkland, Wash. Catey's Law: Best Record Wins -- Unless Records Equal, Then Home Team Wins.' Professor Eric Isaacson of Indiana University notes that last year I attributed this idea to him. Hey Eric -- you don't think I can actually remember everything in Tuesday Morning Quarterback, do you?! At any rate, from now on I will call the generic prediction in question the Isaacson-Tarbell Postulate."

Easterbrook has reported interim and final results several times since then.  Before I link to these, I want to note that, unfortunately, Easterbook consistently over-reports the success of the formula.  I've not verified the cause but believe it is because he is looking at the teams' records after the game has been played, not before.  This has the effect of changing the formula's prediction in favor of the winning team after the game is over!  After each of the links below I'll provide the correct results.

October 10, 2007 (TMQ reports: 52-24, 68%.  Actual: 48-28, 63%)

November 7, 2007 (TMQ reports: 94-36, 72%.  Actual: 87-43, 67%)

February 8, 2008 -- Quoting from the season wrap-up:

"The Isaacson-Tarbell Postulate ruled the landscape of NFL predictions, finishing the season 183-84, or 69 percent correct. Of the dozens of NFL predictors TMQ tracked, John Czarnecki of Fox Sports had the best finish, at 183-84; ESPN's best predictor was Keyshawn Johnson at 176-91. Since its first mention in TMQ, the Isaacson-Tarbell Postuate has been referenced throughout the blogosphere."

The correct figure is: 175-92, 66%

November 18, 2008 (TMQ reports 107-51-1, 69%; Actual: 96-64, 60%)

February 10, 2009 -- Quoting from the season wrap-up:

"The Isaacson-Tarbell Postulate, proposed by TMQ readers Eric Isaacson of Indiana University and Catey Tarbell of Kirkland, Wash., holds: Best Record Wins Unless Records Equal, Then Home Team Wins. If you'd picked 2008 NFL games using this simple algorithm, you would have picked 175 of 267 contests correctly (.665), counting London as a "home" game for New Orleans, Toronto as a "home" game for Buffalo and the Super Bowl as a "home" game for Arizona, which was the home team of record. The Isaacson-Tarbell Postulate is a huge time-saver because you don't have to think about picks -- you don't even need to know who's playing!"

The correct results are 159-108 (60%).