The Importance of Week One Results

Week One Results can be weighted too heavily. Using football science, statistics, the spread and NFL history can help make sense of gambling and trends.


The first week of the pro football season is often chaotic. While the NFL is a league of parity, the opening games bring plenty of uncertainty. Players and coaches have changed teams, and rookies are playing in their first NFL contests. Some players have improved through off-season work, while others have begun to feel the effects of age. The first week of the season is just another week in the standings, but potentially carries with it significant informational value about what has transpired during the offseason. Put another way: How panicked should fans of the Eagles (10.5-point favorites as of this writing) be if the team loses to the Jaguars?

One way to examine the significance of a potential “Week One” effect is to review how teams fared in their openers and compare those results to expectations, as captured in the Team Totals. A Team Total is a bet offered where you can wager on a team to win over a given number of games, or under that number. For instance, the Patriots have a Team Total of 10.5, meaning you can bet on them to win more than 10.5 games, or less than 10.5 (Note: This doesn’t account for vig, a sportsbook’s advantage on each wager, but we’ll ignore that for now.) Drawing on a dataset of NFL closing lines spanning 1999-2013, which are derived from the Team Totals as of the first week of the season, we can take a examine how much Week 1 results can tell us about a team.

First, some background on the data we’re working with. We’re tracking 469 out of a possible 474 teams over those 15 seasons. Five teams were omitted from the dataset because there was no closing Team Total on them, such as the Colts’ 2011 season. This helps bump the average Team Total to 8.1 (from 8), so let’s recalibrate our results to adjust for the fact that all the Totals are shaded a bit high. Team Totals also tend to be tightly bunched, with a standard deviation of 1.7 wins versus an observed wins standard deviation of 3.1 wins.

So, how much did the opening week influence the Team Total prediction by the end of the season? We would expect our Week 1 winners to be a bit better than average, so their expectations before the game should have been to win a bit more than half the time. Getting a full win should bump their expectation by a bit less than half a win. With a reminder that results do not add up evenly because of excluded teams with no Team Total, what we see largely lines up with this:

 

Team Total Actual Wins Difference
Winners 8.275 8.862 0.587
Losers 7.728 7.149 -0.579

 

Week 1 winners were predicted to be slightly better than average (although only slightly). After getting an opening win, they went on to beat expectations by 0.587 wins. That’s a bit better result than what we’d anticipate – the Week 1 winners were expected to win 8.275 games before the season, which works out to 0.517 wins/week. They actually won one game that week, so they beat expectations in Week 1 by 0.483 wins. The difference between the expected 0.483 wins and the observed 0.587 wins could be the size of the “informational” value contained in a Week 1 win. This effect is modest, however, as beating your expectation by 0.1 wins just isn’t much to get excited over.

What about the big favorites who were upset victims, or the big underdogs who pulled off upsets themselves? Surely there is information to be gleaned from the big “surprise” results. Using point spread data from the excellent resource Pro-Football-Reference, we can try and tease out a bigger effect.

Historically, teams favored by 5.5 points or more win approximately two thirds of the time or more (this will be explored further in a future piece). We have 88 such favorites and 87 underdogs during the 15-year period. Overall the data looks like this:

 

Team Total Actual Wins Difference
5.5+ Pt Faves 9.398 9.445 0.046
5.5+ Pt Dogs 6.843 6.403 -0.440

 

Our 5-point favorites are by and large pretty good teams, and most of our 5-point dogs are not very good (I suspect the underperformance of the 5-point dogs relative to their Team Total is noise, but maybe not). So what did it mean when these pretty good teams lost in Week 1?

 

5+ Pt Spread Average Spread Team Total Actual Wins Difference Info
Fave & Win -7.908 9.463 9.833 0.370 0.100
Fave & Loss -7.696 9.260 8.613 -0.647 0.083
Dog & Win 7.648 6.943 7.603 0.661 0.069
Dog & Loss 7.908 6.798 5.863 -0.935 -0.205

 

I’ve added two extra columns here. The first is for the average spread we’re looking at, because 5.5+ point favorites come in many categories. As you can see, the 5.5-point favorites who won tended to be slightly bigger favorites (by about 0.2 to 0.25 points on average) than the 5-point favorites who lost, and vice versa.

The second added column, titled “Info”, attempts to measure the informational value contained in a win or loss. For instance, our bucket of 5.5+ point favorites who won were expected to get approximately 0.73 wins heading into the game (based on the historical winning percentage of 5.5+ favorites since 1978). They actually won 1 game, so they beat expectations by 0.248 wins. With no informational value, we’d expect their season win total to increase by 0.248 wins. Yet they went on to beat their predicted Team Total by 0.370 wins – about 0.1 win better, and a similar size effect to what we saw before.

There are a few takeaways here. First, we get some pretty decent stratification. Winning versus losing your first game as a favorite or underdog was tied to a 1.2-win difference in Team Totals for your actual season win total as a favorite, and a 1.8-win difference as an underdog (although this is largely the result of superior/inferior team quality in the first place). Second, favorites who lost should expect to lose 0.73 wins off their Team Total, but actually only lost .647. In other words, while they didn’t get a win in Week 1, they went on to perform just fine – and actually a bit better than we’d expect with no informational value. Finally, the underdogs that lost underperformed by nearly a full win, and the informational value contained in that loss was double the size of any other effect we’ve seen.

What if we increase our spreads?

 

7.5+ Pt Spread Average Spread Team Total Actual Wins Difference Info
Fave & Win -10.200 9.860 9.926 0.066 -0.184
Fave & Loss -9.955 9.572 8.335 -1.237 0.487
Dog & Win 10.050 6.002 6.976 0.973 -0.223
Dog & Loss 10.200 6.239 5.461 -0.777 -0.027

 

First, a major caveat: This only represents 71 teams and just 21 upsets. There are simply not that many situations where favorites by more than a touchdown have lost in Week 1 over the last 15 years, so sample size is a factor.

However, what we see here is that big favorites who win go on to only modestly outperform their preseason expectation. Meanwhile big favorites who get upset may be in for some real worries, -dropping off from 9.6 projected wins to 8.3 – a steep decline that’s almost a half-win larger than simply that one game in the standings. A similar but smaller effect is seen for big underdog winners. Some real information may be conveyed there.

One final pivot point – what if we ignore wins and losses and just look at margin of victory? In some ways, a 5-point favorite winning by 30 is more impressive than a 5-point underdog winning by 1 (although it counts for the same amount in the standings):

To start, here’s a summary of all teams that didn’t push in Week 1, divided into groups that covered the spread and groups that did not:

 

N = 224 W% Average Spread Cover By Team Total Actual Wins Difference
No Cover 0.174 -0.413 -10.618 8.010 7.474 -0.536
Cover 0.823 0.361 -10.618 7.950 8.502 0.551

 

Not too much here. Both buckets of teams looked largely the same before the season (a predicted win total of about 8), and the average spread they were up against was fairly neutral. The teams that beat the spread at all, regardless of margin, won a full game more on the season than those that didn’t. However, almost all of that effect was simply the extra Week 1 win. There’s no real informational value here yet.

How about teams that covered by at least a touchdown? Let’s see:

 

N = 140 W% Average Spread Cover By Team Total Actual Wins Difference
No Cover 0.057 -0.639 -15.175 8.100 7.075 -1.025
Cover 0.943 0.535 -15.175 7.903 8.743 0.839

 

Teams that cover by at least a touchdown almost always win. Such teams are winning .839 games more than expected preseason, and only about half (.943 – .500) are simply the result of getting a win booked.

We start to see diminishing marginal returns for the teams that cover here around this point. Teams that cover by 10.5 or more look pretty similar.

 

N = 92 W% Average Spread Cover By Team Total Actual Wins Difference
No Cover 0.011 -0.060 -18.582 7.931 6.792 -1.139
Cover 0.989 -0.060 -18.582 8.022 8.882 0.861

 

However, the effect continues to increase for the teams that fail to cover. Teams that cover by 15:

 

N = 56 W% Average Spread Cover By Team Total Actual Wins Difference
No Cover 0.000 -0.009 -22.295 8.021 6.495 -1.526
Cover 1.000 -0.188 -22.295 8.074 8.898 0.824

 

And by 20:

 

N = 30 W% Average Spread Cover By Team Total Actual Wins Difference
No Cover 0.000 0.633 -26.300 8.315 6.544 -1.771
Cover 1.000 -0.633 -26.300 8.216 8.936 0.719

 

At this point, the teams that have covered by a lot have taken a small step back. But the teams that wildly underperformed gambling expectations in game 1, despite being predicted as better teams than those that blew them out, totally collapsed the rest of the season. They lost 1.77 wins from their season long expectation, only about a half-win of which can be explained by the loss itself. The other ~1.2 wins are additional informational value, a reduction in aforementioned uncertainty that led to this inquiry in the first place.

This result continues to increase, to the point where teams that failed to cover by 30 or more went on to win fewer than 4.5 games the rest of the season, despite being rated average teams before the season.

Margin of victory can be tricky in football, where running up the score doesn’t seem to tell you much. However, from this, it looks like you can’t tell too much about which teams are going to wildly exceed expectations by their Week 1 results, but you can get a pretty good idea of which teams are going to struggle the rest of the way.

Follow Konstantin on Twitter @kmedved.

Konstantin Medvedovsky analyzes data, looking at home field advantage and weather, or ifearly-season results are predictive.

12 thoughts on “The Importance of Week One Results

  1. With all the salacious happenings in the NFL today, I don’t want this to get lost. 
     
    This is one of the finest things I’ve ever read on SoSH, and that includes some of the best baseball content we’ve featured over the years. Everyone, please take a few minutes and read – and comment – on this article written by a fellow SoSHer. 

  2. Great article.

    I was eager to exploit this knowledge for gambling purposes, but it appears the market may be efficient here. The four teams to bet against based on bowiac’s insights seem to be the Rams, Bucs, Chiefs and Patriots. The Rams and Bucs play each other. The other two lines have already moved 2-3 points compared to early lines that were published before yesterday’s games. In the case of Vikings-Pats, that movement could be attributed to the Vikings’ unexpectedly strong performance yesterday, but in the case of Broncos-Chiefs the movement has to be ascribed to KC’s struggles, as Denver’s performance almost perfectly matched pre-game expectations.

  3. Fantastic read. I’ll be interested to see if I can apply some your logic in week 2 if those inefficiencies still exist.

    I don’t know if I’d call them inefficiencies – this is based on information not yet available (Week 1 results), so it’s not really exploitable in the way we’d like an inefficiency to be. As maufman points out, Vegas is moving the line on the teams that this analysis flagged. If I have time before this weekend, I’ll try and look at how these teams did in Week 2 though.

  4. Interesting topic. I have a lot of this stuff easily accessible, so thought I’d run some numbers.

    Wins are the total for the season, including playoffs. So I don’t use winning percentages. To keep the numbers consistent, I used all the seasons from 1978-2013, eliminating the short seasons of 1982 and 1987. There were two fewer playoff games per season before 1990. That change consistently affects all weeks and is tiny. The differences between the league size of 28 (through 1994) and larger afterward, related to the number of playoff teams is similarly tiny. I didn’t separate week 18 because there was only one year that had a week 18.

    The average number of wins per team is 8.51 per season (this is far from the center of the week averages shown below because playoff weeks only contain teams with a lot of wins).

    Week / Average total wins for team that won / Average total wins for team that lost
    1, 9.64 – 7.05
    2, 9.56 – 7.07
    3, 9.74 – 6.89
    4, 9.57 – 7.00
    5, 9.65 – 7.10
    6, 9.65 – 7.09
    7, 9.58 – 6.94
    8, 9.51 – 7.12
    9, 9.68 – 7.08
    10, 9.68 – 7.00
    11, 9.67 – 7.01
    12, 9.72 – 6.97
    13, 9.58 – 7.08
    14, 9.65 – 7.06
    15, 9.50 – 7.19
    16, 9.56 – 7.09
    17, 9.61 – 7.13

    Total, weeks 1-18, 9.62 – 7.05

    So, week three stands out a tiny bit, maybe. Seems well within random fluctuation. Week one is pretty much exactly the average. And maybe a bit of a slope in weeks 15-17 as the top teams can start to prepare for the playoffs.

    I don’t see anything to indicate that a week one win or loss is any more important than any other week. Week one performance seems very representative of overall team quality.

  5. I don’t see anything to indicate that a week one win or loss is any more important than any other week. Week one performance seems very representative of overall team quality.

    Based on some stuff I’ve looked at before, and based on what you show here, I agree (although it’s interesting even Week 17 looks the same as ever).
     
    I wasn’t trying to suggest there’s anything more important about Week 1 than any other week – I was mostly trying to see how much we should adjust season expectations based on Week 1 results. 

  6. Yes, I was just extending the thought a little. It’s hard not to get into tiny sample size issues when trying to factor in expectations. Maybe the best way to do that is simply use last season’s win totals. So, looking at that piece, I have around 900 data points for week one. Of those, 58 had 13 wins (like New England, including playoffs), in the preceding season.

    The average win total for the following season was only 9.64. That’s quite a drop. I think we can agree that what New England has accomplished in the Brady/Belichick era is remarkable.

    Those 58 teams went 29-29 in week 1 the following year. The 29 that won averaged 10.79 wins, the 29 that lost averaged 8.48 wins.

    That’s really interesting. Maybe there’s something there. Not in the final results, wins versus losses. That’s consistent with the above chart. But the 29-29 record itself is quite interesting, as these teams wind up being a couple of wins better than average during the season.

    So, a deeper look shows that only 23 of the 58 were at home in week one. That really is the problem when you start isolating things – you get unexpected biases here and there.

    Going further, let’s see what happens with 12 wins the previous season…

    Record: 29-14 (25 of 43 at home). 10.97 wins for winning teams in week one, 7.79 for losing teams.

    And 14 wins…

    Record: 15-11 (11 of 26 at home). 13.50 wins for winning teams, 8.45 for losing teams.

    And 11 wins…

    Record: 47-27 (38 of 74 at home). 10.51 wins for winning teams, 8.74 for losing teams.

    I don’t have any really solid conclusions. The average team that won 13 games, including playoffs, wins close to 10 games the following season. So, at this point, with very small sample sizes to base this on, you could set an expected win total of 9 on the Patriots right now based on their loss. But anyone with even a tiny bit of understanding of sample size and how statistics work wouldn’t call that a confident prediction. After all, Miami is probably a decent team this year, and this was a road game in 90-degree heat and a heat index of 99.

  7. I don’t have any really solid conclusions. The average team that won 13 games, including playoffs, wins close to 10 games the following season. So, at this point, with very small sample sizes to base this on, you could set an expected win total of 9 on the Patriots right now based on their loss. But anyone with even a tiny bit of understanding of sample size and how statistics work wouldn’t call that a confident prediction. After all, Miami is probably a decent team this year, and this was a road game in 90-degree heat and a heat index of 99.

    I took a quick look at your heat idea, and there does seem to be something to it. Favorites from 1978-2012 (haven’t added 2013 yet) win at around a 63-64% clip overall. For games payed at least 85 degrees (I don’t have humidity info yet for heat index stuff, but am building out that database as well), that number drops to around 58%. This is quick and dirty – I’m not doing a comparison of spreads here, and I’m looking at all games, to avoid cutting my sample size too much (games that hot are rare regardless). 
     
    Push comes to shove, I do think this loss should lower expectations a bit, although as you note, the Belichick-Patriots have defied every other statistical rule, including regression to the mean (as shown in your analysis of 13 win teams above), so it’s hard to go to far with this. I’m a bit worried however. 

  8. I took a quick look at your heat idea, and there does seem to be something to it. Favorites from 1978-2012 (haven’t added 2013 yet) win at around a 63-64% clip overall. For games payed at least 85 degrees (I don’t have humidity info yet for heat index stuff, but am building out that database as well), that number drops to around 58%. This is quick and dirty – I’m not doing a comparison of spreads here, and I’m looking at all games, to avoid cutting my sample size too much (games that hot are rare regardless). 

    I’d love to see whether games with temperature extremes show more of a homefield advantage since the NFL went to the more limited practice schedule under the new CBA. Belichick had the team practicing in sweats last week to simulate the hot temperatures, but there’s a limit to how much of that sort of preparation they can do now. There’s probably not enough of a sample of such games to draw a conclusion, but it bears watching.

  9. The struggles of the Vikings and the Titans so far today support bowiac’s hypothesis that unexpectedly bad Week 1 performances tell us more than unexpectedly good ones.

Leave a Reply

Your email address will not be published.