Writers of Pro Football Prospectus 2008

Most Recent FO Features

LancasterJar15.jpg

» FEI Primer: Elite Characteristics

There will be four teams in the inaugural College Football Playoff at the end of the season. What common characteristics will distinguish these teams above all others?

31 Aug 2004

Guest Column: Strength of Schedule -- A Dynamic Analysis

Guest Column by Anthony Brancato

How does one go about determining how "tough" or "easy" an NFL team's schedule is? If you search the Internet to find out, chances are you'll run across a chart that looks like this:

2004 Strength of Schedule
Team Opp.
Win
Opp.
Loss
Overall
Season
Rank
Overall
Season
Win%
Team Opp.
Win
Opp.
Loss
Overall
Season
Rank
Overall
Season
Win%
Dallas 120 136 1 .469 Baltimore  130 126 17 .508
Philadelphia  121 135 2 .473 Green Bay  130 126 18 .508
Pittsburgh 122 134 3 .477 Indianapolis  130 126 19 .508
N.Y. Giants 123 133 4 .480 San Diego  130 126 20 .508
Cleveland 124 132 5 .484 Buffalo 131 125 21 .512
Tampa Bay 124 132 6 .484 Cincinnati 131 125 22 .512
Tennessee 124 132 7 .484 Houston 131 125 23 .512
Atlanta 125 131 8 .488 New England 131 125 24 .512
Denver 125 131 9 .488 N.Y. Jets 131 125 25 .512
Kansas City 125 131 10 .488 Oakland 131 125 26 .512
Carolina 126 130 11 .492 San Francisco 131 125 27 .512
Detroit 126 130 12 .492 St. Louis 131 125 28 .512
Washington 126 130 13 .492 Jacksonville 132 124 29 .516
Chicago 127 129 14 .496 Seattle 132 124 30 .516
Minnesota 127 129 15 .496 Arizona 134 122 31 .523
New Orleans 129 127 16 .504 Miami 136 120 32 .531
Rank is based on 2004 full season schedule strength.
Rank 1 is the easiest schedule, 32 is the hardest.

But what kind of useful information does this really provide? So what if a team is going to play weak opponents in the coming year; if they also played a cream-puff schedule last year, how will it help them improve? And the same basic point of course would apply, in reverse, to a team drawing a tough schedule.

Far more insight can be gained by comparing a team's strength of schedule for the upcoming season with the one they played the previous season. The following chart shows the 2003 winning percentage of each team's 2003 opponents, and that of the opponents each team is scheduled to play in 2004:



Team 2003
Win %
2004
Win %
Diff.
N.Y. Giants .555 .480 +75
Buffalo .570 .512 +58
Houston .570 .512 +58
Cleveland .539 .484 +55
Atlanta .539 .488 +51
Detroit .535 .492 +43
Washington .531 .492 +39
Jacksonville .543 .516 +27
Tampa Bay .508 .484 +24
Pittsburgh .500 .477 +23
Arizona .543 .523 +20
N.Y. Jets .527 .512 +15
Denver .500 .488 +12
Oakland .516 .512 +4
Philadelphia .477 .473 +4
San Francisco .512 .512 0


Team 2003
Win %
2004
Win %
Diff.
New Orleans .500 .504 -4
San Diego .504 .508 -4
Chicago .488 .496 -8
Dallas .461 .469 -8
Tennessee .473 .484 -11
Indianapolis .492 .508 -16
Miami .512 .531 -19
Green Bay .488 .508 -20
New England .484 .512 -28
Minnesota .457 .496 -39
Carolina .445 .492 -47
Baltimore .457 .508 -51
Seattle .465 .516 -51
Cincinnati .457 .512 -55
Kansas City .418 .488 -70
St. Louis .434 .512 -78

Note that the gap between the most and least favorable schedules widens dramatically when the previous year's figures are factored in.  In fact, it more than doubles, from 62 percentage points (or 16 games) to 153 percentage points (or 40 games).

But what about the history of all this? Well, the last four teams that took the sharpest drop in strength of schedule in the entire league for that year improved their records by a combined 17 games the following season; and even more significantly, three of them made the playoffs "the year after" when they had not done so "the year before":

Year(s) Team Previous
Year's
S.O.S.
Upcoming
Year's
S.O.S.
Previous
Year's
Record
Following
Year's
Record
2002-2003 Seahawks 129.5 113.5 7-9 10-6
2001-2002 Colts 144.0 112.0 6-10 10-6
2000-2001 Chargers 145.0 119.0 1-15 5-11
1999-2000 Eagles 141.0 109.0 5-11 11-5
S.O.S. = Strength of Schedule

This table uses number of opponents' victories instead of opponents' winning percentage because, thanks to Houston, the latter was impossible to quantify for 2002. Games against Houston count as zero opponent victories in the upcoming year's strength of schedule for the 2002 Colts.  Ties count as one-half victory (applicable for 2002-03 because Atlanta and Pittsburgh played a tie game in 2002).

The foregoing should be awesome news for the Giants, who take the biggest drop in schedule difficulty this year; but then again the Giants (and also the Raiders) will have to deal with a very ugly if unrelated trend: They won only four games in 2003 after having made the playoffs in 2002. Not counting the war-shortened seasons of 1943 through 1945 and the strike-shortened season of 1982, 312 NFL/AFL teams won fewer than five games in a season from 1933 through 2002 -- and of those 312 teams exactly one of them made the playoffs both the year before and the year after: the 1962 Chargers. So the strength-of-schedule factor may be meeting its match this season.

Also note how including the previous year's schedules in the calculations greatly alters the relationship between the schedules of the AFC West's top two teams, Kansas City and Denver. If one looks at only the 2004 strength-of-schedule chart, the two teams have identically-difficult schedules (both teams' 2004 opponents were a combined 125-131 in 2003). But when the schedules each team played last season are added in, it is revealed that the Broncos will be taking a marginal drop in strength of schedule -- from 128 opponents' victories to 125 -- while the Chiefs will be moving way up in company, as their 2003 foes won a mere 107 games. Which approach paints a more accurate picture?

Of course just because a particular team was good (or bad) one year doesn't mean they will be the same the next, and spectacular anomalies have occasionally arisen.  In 1992, for example, the Dallas Cowboys were supposed to play the toughest schedule in the NFL using the 1991 final records -- yet Dallas ended up playing the league's EASIEST schedule based on where everyone finished in 1992 itself. And, yes, the Cowboys went 13-3 that year and won the Super Bowl.

Tom Flores was definitely onto something when he observed that it's not so much who you play as when you play them. Still, it doesn't hurt to know the truth -- the whole truth -- about every team's schedule; knowing only half the story amounts to a half-truth, and as the old Yiddish proverb goes, a half-truth is a whole lie.

Anthony Brancato is a moderator of the NFL discussion boards at Sports-Central.org, where he hosts a weekly NFL pick 'em contest.  Comments? Make them in our discussion thread, or contact Anthony at ajbrancato @ yahoo.com. If you are interested in writing a guest column, something that takes a new angle on the NFL, please email us your idea at info @ footballoutsiders.com.

Posted by: Guest on 31 Aug 2004

comments