≡ Menu

Stanford makes terrible quarterbacks.

Twice in four years, the Buffalo Bills have teased their fans. In 2008, the Bills started 5-1, and Peter King named Trent Edwards his MVP after the first quarter of the season. Edwards had led fourth-quarter comebacks and game-winning drives in the second and third weeks of the year, and he ranked in the top-five in AY/A after seven weeks.

The former Stanford Cardinal would rank 24th in AY/A the rest of the season while Buffalo finished the year with a 2-8 record.

Fitzpatrick was smart enough to shave before the game.

The Bills won just ten games the next two seasons, but made some noise at the start of the 2011 season. Going back to the nerd well at quarterback, Buffalo raced out to a 5-2 start. Fitzpatrick led two fourth-quarter comebacks/game-winning drives and ranked 11th in AY/A after eight weeks. Unfortunately, the former Harvard star ranked 32nd in that metric over the last nine weeks of the season, and the Bills finished the season 1-8.

How unusual is it for a team to have such a hot start and cold finish? It’s simple enough to look at first-half/second-half splits, but I prefer a more nuanced approach by weighing each team game based on when it occurred; e.g., game 1 counts 16 times as much as game 16, game 2 counts 15 times as much as game 16, game 3 counts 14 times as much as game 16, and so on. I looked at each team since 1990 and calculated their actual winning percentage and their “weighted” winning percentage.

The 2011 Bills had a 0.375 winning percentage last year, but by placing greater weight on games earlier in the season, Buffalo had a 0.507 weighted winning percentage. In 2008, the Bills had a 0.438 actual winning percentage and a 0.566 weighted winning percentage. As it turns out, those were two of the five “strongest-starting” teams of the last five years. The table below lists the “strongest-starting teams” since 1990, along with their actual and weighted winning percentages. The last column represents the difference between the two winning percentages.

TmYearAct. Win%Wt. Win%DIFF
DEN20090.5000.6760.176
SDG20010.3130.4850.173
SDG20020.5000.6690.169
OAK19950.5000.6690.169
PHI19940.4380.5960.158
MIA19930.5630.7130.151
STL19950.4380.5880.151
NOR20090.8130.9560.143
MIN20000.6880.8310.143
MIA19990.5630.7060.143
PIT19980.4380.5810.143
TAM20110.2500.3900.140
NOR20020.5630.6990.136
BUF20110.3750.5070.132
NOR19930.5000.6320.132
PHO19910.2500.3820.132
BUF20080.4380.5660.129
NYJ20000.5630.6910.129
NYG20090.5000.6250.125
DET20070.4380.5590.121
MIN20030.5630.6840.121
KAN19960.5630.6840.121
IND20060.7500.8680.118
NYJ20040.6250.7430.118
STL20000.6250.7430.118

We can also look at the “fastest finishers.” These teams all had significantly higher actual winning percentages than weighted winning percentages, because of how well they played at the end of the season:

TmYearAct. Win%Wt. Win%DIFF
CIN20080.2810.129-0.153
CLE20090.3130.162-0.151
TEN20060.5000.353-0.147
TEN20090.5000.353-0.147
DET20100.3750.235-0.140
BUF20040.5630.426-0.136
MIA20110.3750.243-0.132
TEN20020.6880.559-0.129
MIA20080.6880.559-0.129
IND20080.7500.625-0.125
SEA20020.4380.316-0.121
MIN20020.3750.257-0.118
HOU20080.5000.382-0.118
ARI20110.5000.382-0.118
MIA20050.5630.449-0.114
SDG20070.6880.574-0.114
GNB20060.5000.390-0.110
SDG20090.8130.706-0.107
CAR20090.5000.397-0.103
PHI20110.5000.397-0.103
GNB20030.6250.529-0.096
MIA20040.2500.154-0.096
DAL20100.3750.279-0.096
NYJ20020.5630.471-0.092
CAR20040.4380.346-0.092

In case you were wondering, there doesn’t appear to be much of a correlation between whether a team is a “strong starter” or a “fast finisher” and how well the team does the following season. I ran a regression using each team’s winning percentage and the differential column in the tables above. As it turns out, the differential variable was not statistical significant at the 10% level (p-value = 0.13), and even if it was, it was not practically significant (but with a very slight tendency towards the ‘strong starters’ playing better the next year). My guess is that the data gets clouded by teams who suffer injuries versus teams that simply see their play dropoff, and also by teams resting opponents late in the season and/or playing teams resting starers.

I looked at the top 24 “strongest starters” and top 24 “fastest finishers” and found similar results. On average, both groups won 53% of their games in Year N. The “strong starters” had a weighted winning percentage of 0.665 while the “fast finishers” had a weighted winning percentage of 0.383. In Year N+1, the strong starters won 54% of their games, while the fast finishers won 49% of their games.

Previous “Random Perspective On” Articles:
AFC East: Buffalo Bills, Miami Dolphins, New England Patriots, New York Jets
AFC North: Baltimore Ravens, Cincinnati Bengals, Cleveland Browns, Pittsburgh Steelers
AFC South: Houston Texans, Indianapolis Colts, Jacksonville Jaguars, Tennessee Titans
AFC West: Denver Broncos, Kansas City Chiefs, Oakland Raiders, San Diego Chargers
NFC East: Dallas Cowboys, New York Giants, Philadelphia Eagles, Washington Redskins
NFC North: Chicago Bears, Detroit Lions, Minnesota Vikings, Atlanta Falcons
NFC South: Atlanta Falcons, Carolina Panthers, New Orleans Saints, Tampa Bay Buccaneers
NFC West: Arizona Cardinals, San Francisco 49ers, Seattle Seahawks, St. Louis Rams

{ 5 comments }