Part I --
Can preseason records help us in week one of the regular season?
So you have all the NFL teams slogging through an arduous preseason schedule of games, and while the results are typically discounted when it comes time to play games that mean something, the question we'll try to answer is can the won-lost preseason records be predictive for week one matchups?
Our data sample covers the seasons from 1997 to 2001 (a five-year span) and we will begin by simply tracking all teams based on how many preseason wins they accrue:
Preseason
Wins |
Spread Record
Home |
Home % |
Spread Record
Away |
Away % |
Total
Spread Record |
Total % |
| 0 |
4-4 |
50% |
2-0 |
100% |
6-4 |
60% |
| 1 |
7-8 |
47% |
10-13 |
43% |
17-21 |
45% |
| 2 |
10-11 |
48% |
10-10 |
50% |
20-21 |
49% |
| 3 |
9-10 |
47% |
11-7 |
61% |
20-17 |
54% |
| 4+ |
2-6 |
25% |
6-2 |
75% |
8-8 |
50% |
The first thing to mention is that there is a bias in the past five years data (that may nor may not continue), which is that in week one away teams have gone 39-32 (55%). Consequently the table above reflects this trend in that regardless of the number of preseason wins, no group has fared well at home. At the same time, there are a couple of positive "away" patterns, with teams that had great preseasons (3+ wins) mustering a nice 17-9 away mark, and the lowly winless preseason sides, who have a 2-0 history opening on the road.
These sample sizes are so tiny that we can have little confidence in the trends, but it's a beginning. The next step we will take is to break it out by favorite/underdog roles:
Note: with pick'em games the home team is treated as the underdog
Preseason
Wins |
Spread Record
as Favorite |
Fav. % |
Spread Record
as Underdog |
Dog. % |
| 0 |
1-3 |
25% |
5-1 |
83% |
| 1 |
4-10 |
29% |
13-11 |
54% |
| 2 |
9-12 |
43% |
11-9 |
55% |
| 3 |
11-11 |
50% |
9-6 |
60% |
| 4+ |
5-5 |
50% |
3-3 |
50% |
This is a perhaps a little more interesting, as the teams coming off poor preseasons (0 or 1 wins) are lousy bets as favorites at the start of the regular season -- posting a dismal 5-13 (28%) record. Again this is a very small number of games, but at least logically it makes sense that a bad preseason may leave a team underprepared for when the real action begins.
On the contrary though, a winless preseason has not been a bad thing for week one underdogs, who have covered five of six.
In conclusion then there is not huge predictive value for the preseason records, but the following points are worth considering in the big picture of a week one handicapping analysis:
- Teams with 3+ preseason wins have been solid on the road (17-11)
- Teams with 1 or fewer preseason wins have been bad bets as favorites (5-13)
- Teams with 0 preseason wins have been good as underdogs (5-1)
Copyright © 2002 by TwoMinuteWarning.com, All Right Reserved
| 
