|
Part One of the series: Won-Lost Records coming into the game
No one doubts the importance of a team's motivation coming into a game in so far as what constitutes a sound wager. Indeed we have in the past written articles on "schedule letdowns", the "horrible defeat syndrome", the "bitter heart" and other mental considerations of a matchup. These past efforts were all however rather unscientific, and so we have set out now to quantify motivation -- to provide fixed numeric data standards that can be applied to an NFL contest to produce a better understanding of the likely forces at work on a team's psyche...
We have elected to begin with a simple enough task: let's look at the comparison between the two teams' records on the season, coming into the game. To categorize teams we have set-up four classes:
- Poor -- teams with a winning percentage of under 40%
- Mediocre -- teams with a winning percentage of 40% to 49%
- Good -- teams with a winning percentage of 50% to 59%
- Excellent -- teams with a winning percentage of 60% or higher
Running these guidelines through all the match-ups from week 5 on from the period of 1991-2000 (eg a ten year span) produced the following results:
| Team |
Opponent |
Won (vs Spr) |
Lost (vs Spr) |
Win % |
| Poor |
Mediocre |
54 |
59 |
47.8 % |
| Poor |
Good |
103 |
79 |
56.6 % |
| Poor |
Excellent |
208 |
185 |
52.9 % |
| Mediocre |
Poor |
59 |
54 |
52.2 % |
| Mediocre |
Good |
45 |
32 |
58.4 % |
| Mediocre |
Excellent |
86 |
88 |
49.4 % |
| Good |
Poor |
79 |
103 |
43.4 % |
| Good |
Mediocre |
32 |
45 |
41.6 % |
| Good |
Excellent |
126 |
109 |
53.6 % |
| Excellent |
Poor |
185 |
208 |
47.1 % |
| Excellent |
Mediocre |
88 |
86 |
50.6 % |
| Excellent |
Good |
109 |
126 |
46.4 % |
ANALYSIS: Please note the same class matchups (eg Good vs Good) are excluded since there would always be one win and one loss. Of course a key dilemma with any kind of research you do for the NFL is that on the one hand you have a small sample size, on the other hand going back in time is dangerous since the NFL, and particularly NFL betting habits, do change from year to year. Still, the above "first pass look" is interesting in that the two strongest matchups for a gambler are Mediocre vs Good (58.4%) and Poor vs Good (56.6%). Now, one concern that instantly comes to mind is the point in the season these games occur -- in week five for instance it would be natural for some teams to be over-rated based on a couple of good efforts. Here then are the same numbers, only looking at week ten on:
| Team |
Opponent |
Won (vs Spr) |
Lost (vs Spr) |
Win % |
| Poor |
Good |
60 |
52 |
53.6 % |
| Mediocre |
Good |
38 |
27 |
58.5 % |
So the Mediocre vs Good is no different in the later weeks, however the Poor vs Good is much stronger in that "second quarter" of the season, 43-27 (61%) and then 53% thereafter. Now we could of course search for the optimum week cutoffs, but that would be a mistake since all we would do is backfit the data. Rather we can take away the notion that yes, Good teams are slightly over-rated when playing a lesser bunch.
The next worthwhile step would appear to be to separate the match-ups based on which team was at home and which was away. Going back then to our initial week five on setting (and we go with the week five since the early weeks are another matter altogether and we will visit them shortly), here's what we find:
| Home Team |
Away Team |
Won (vs Spr) |
Lost (vs Spr) |
Home W% |
| Poor |
Poor |
81 |
80 |
50.3 % |
| Poor |
Mediocre |
30 |
33 |
47.6 % |
| Poor |
Good |
52 |
45 |
53.6 % |
| Poor |
Excellent |
122 |
89 |
57.8 % |
| Mediocre |
Poor |
26 |
24 |
52.0 % |
| Mediocre |
Mediocre |
13 |
22 |
37.1 % |
| Mediocre |
Good |
22 |
17 |
56.4 % |
| Mediocre |
Excellent |
46 |
40 |
53.5 % |
| Good |
Poor |
34 |
51 |
40.0 % |
| Good |
Mediocre |
15 |
23 |
39.5 % |
| Good |
Good |
33 |
31 |
51.6 % |
| Good |
Excellent |
53 |
51 |
51.0 % |
| Excellent |
Poor |
96 |
86 |
52.7 % |
| Excellent |
Mediocre |
48 |
40 |
54.5 % |
| Excellent |
Good |
58 |
73 |
44.3 % |
| Excellent |
Excellent |
176 |
135 |
56.6 % |
To get the away record for a pairing, simply reverse the home results. Be careful not to look instead at the reversed matchup (eg thinking Poor vs Good is the away version of Good vs Poor). A few interesting areas here:
- Poor home team versus Excellent away team: the poor home squad, most likely an underdog, covers 57.8%
- the Battle of the Mediocres: a very small sample, but the away team gets the cover 63%
- Good Home team versus Poor/Mediocre away team: the lowly away sides get the job done 60% of the time.
- Primetime Match-Ups: in a game between two Excellent sides, the home team, suitably fired up one would presume for a big game, covers almost 57% of the time. The reason why there are so many teams classified as Excellent is because we use a fairly low standard and mid-season it's relatively easy to be 60%+
.
Reviewing the above using our week 10 on guideline shows the Poor vs Excellent is not much different (56.7%), the Mediocre-Mediocre is even more pronounced (67% away covers), the Good vs Poor/Mediocre is less strong (56% aways), and the Excellent-Excellent is much reduced to 53% for the home guys. Now the Excellent-Excellent data means in weeks 5 through 9 that the E-E pairings produces a home cover 62% of the time -- in other words an 'early big game' when both teams are faced with a true test, the home side should get the nod.
All right, well that's a good beginning with a few "early angles" to test out against the 2001 season. A couple of more things we want to check in this arena though, include the "extreme teams" and the pre-week 5 games.
For the extremes our first choice is those sad, frustrated winless franchises:
| Home Team |
Away Team |
Won (vs Spr) |
Lost (vs Spr) |
Home W% |
| Poor |
Winless |
3 |
8 |
27.3 % |
| Mediocre |
Winless |
2 |
4 |
33.3 % |
| Good |
Winless |
0 |
9 |
0.0 % |
| Excellent |
Winless |
15 |
10 |
60.0 % |
| *Any* |
Winless |
20 |
31 |
39.2 % |
| Winless |
Poor |
5 |
4 |
55.6 % |
| Winless |
Mediocre |
4 |
2 |
66.7 % |
| Winless |
Good |
6 |
7 |
46.2 % |
| Winless |
Excellent |
19 |
7 |
73.1 % |
| Winless |
*Any* |
34 |
20 |
63.0 % |
| Winless |
Winless |
3 |
0 |
100.0 % |
This is one category that's very impressive and should be taken seriously! A winless team from week five on is indeed a dangerous team against the spread. The record shows 65-40 (62%) overall and in those rare winless-winless games the home field advantage has been strong. This is one angle we can already attest to the 2001 results -- Detroit were very good bets despite continually falling just short of the outright win, going 8-2 against the line, culminating in the week 14 win over the Vikings.
Now let's look at the undefeateds:
| Home Team |
Away Team |
Won (vs Spr) |
Lost (vs Spr) |
Home W% |
| Poor |
Undefeated |
11 |
4 |
73.3 % |
| Mediocre |
Undefeated |
3 |
3 |
50.0 % |
| Good |
Undefeated |
2 |
5 |
28.6 % |
| Excellent |
Undefeated |
8 |
5 |
61.5 % |
| *Any* |
Undefeated |
24 |
17 |
58.5 % |
| Undefeated |
Poor |
7 |
4 |
63.6 % |
| Undefeated |
Mediocre |
1 |
2 |
33.3 % |
| Undefeated |
Good |
1 |
5 |
16.7 % |
| Undefeated |
Excellent |
8 |
2 |
80.0 % |
| Undefeated |
*Any* |
17 |
13 |
56.7 % |
| Undefeated |
Undefeated |
3 |
2 |
60.0 % |
This is different from the winless group -- overall you'd recommend playing against the Undefeated team on the road, but siding with them at home.
Lastly then, we will look at the early week 2 to week 4 pairings, and for this we will adjust our basic classing system to the following:
Poor -- less than 50%
Average -- 50%
Good -- 51%+
| Home Team |
Away Team |
Won (vs Spr) |
Lost (vs Spr) |
Home W% |
| Poor |
Poor |
31 |
48 |
39.2 % |
| Poor |
Average |
4 |
7 |
36.4 % |
| Poor |
Good |
32 |
42 |
43.2 % |
| Average |
Poor |
10 |
6 |
62.5 % |
| Average |
Average |
12 |
16 |
42.9 % |
| Average |
Good |
9 |
10 |
47.4 % |
| Good |
Poor |
34 |
37 |
47.9 % |
| Good |
Average |
12 |
8 |
60.0 % |
| Good |
Good |
36 |
31 |
53.7 % |
Some seeming wagering opportunities here: Poor teams have been terrible at home (67-97 for 40.8%), so go against them when possible!
All told then this first step in motivation research has been quite promising! There are a number of areas that suggest strong wagering situations, including:
- Back Mediocre teams against Good teams -- 58.4%
- Back Poor home teams against Excellent away teams -- 57.8%
- Back away Mediocre teams against home Mediocre teams -- 63%
- Back Poor/Mediocre away teams against Good home teams -- 60%
- Back Excellent home teams against Excellent away teams, at least for the second quarter of the season
- Back Winless teams from week 5 on -- 62%
- Play against Poor home teams in weeks 2 thru 4 -- 59%
As mentioned we'll recap these plays for the 2001 season (eg fresh data not included in our stats above)...for the results, check out Testing the Angles, part I
In the meantime, get ready for part II of the "Quantifying Motivation" series when we'll look at how a team's record compares to the expectations from the record of the previous season!
Copyright © 2002 by TwoMinuteWarning.com, All Right Reserved
|