If a team is playing badly, then a bye week means an extra week to reevaluate and turn things around. If a team is playing well, then the bye is a time to rest and keep players fresh for the postseason.
The interesting thing about bye weeks is that they occur at very different times in the season. Green Bay and Philadelphia had the earliest bye weeks this year, during Week 4, while Cleveland and Tennessee have the latest, during Week 13.
When a bye week occurs has interesting implications for fans and bettors alike. Is it better to have a week off later in the season, when you may need it to recover from injuries? Or is it better to have it earlier in the season, to prevent injuries and fatigue from happening in the first place? Or perhaps right in the middle of the season, for a combination of those two effects?
The first thing we can immediately look at is how teams do after their bye week. Below we have plotted the win rate of teams straight up after their bye week.
We can see there exists a slight downward trend, especially during the last three weeks when teams really start to lose. Despite the fact that the two teams with the latest bye week this year are Cleveland and Tennessee, we don’t think it is usually the case that worse teams get later bye weeks, or that there is any relationship between bye week and team strength. Still though, we can look how teams do against the spread to see if there is any real effect or whether it has just been the case that more bad teams than good teams have had later bye weeks.
We can still observe the same trend – a slightly negative relationship between bye week number and win percentage ATS for most of the season, followed by a massive drop in win percentage ATS for teams that have Bye weeks in Weeks 11,12 and 13.
But is that drop significant? We can run a t-test on games played directly after a bye week in Week 11, 12 or 13 to check this. Doing so, we find that teams have won 34.1 percent of those games – good for a p-value of 0.06 (due to a small sample size). Although this isn’t statistically significant, it is definitely practically significant as if you had bet against them you would have won much more than 52.4 percent of your bets, which is the number required to break even when factoring in vig.
Why might this edge exist? We can think of two possible explanations: Having a later bye week actually affects the players, as perhaps they are more likely to get injured or significantly fatigued due to the long period of consecutive games at the beginning of the season.
Alternatively, this could be due to bettors mentality. They may overestimate the effect of a bye week later in the season, and therefore the lines may be too high for teams with bye weeks later in the season, leading to those teams not covering them as often.
Given the fact that teams actually lose these games, not just lose ATS, we would personally lean towards the first explanation, but think valid arguments could be made either way.
Finally, we thought it would also be interesting to see how bye week scheduling affects how a team does later in the season. Specifically, looking at the last week of the season, do teams that had a late bye perform better in those games than teams with an earlier bye week, since they have more time to rest?
Once again, we can look at this from a straight up and ATS perspective. This time, we have plotted them on the same graph, with ATS results in red, and straight up results in blue.
We can see that there’s almost an opposite effect as before. When looking at win percentage in Week 17, teams that have a later bye week are at an advantage. It is important to note that this is not a statistically significant advantage, as teams with bye weeks in Weeks 11,12 and 13 perform better ATS at a 58.9 percent rate, which only produces a p-value of .3367. Still, it is interesting to see the trend.
In conclusion, bye weeks that occur later on in the season are actually detrimental in the short term, as those teams perform worse ATS in the game immediately after their bye week, but beneficial in the long run, with those same teams outperforming others in Week 17.
Harrison Chase is the Co-President of the Harvard Sports Analysis Collective, a student-run organization at Harvard College dedicated to the quantitative analysis of sports strategy and management.
Tagged with: feature