Greetings fellow BTSCers. With the announcement of the schedule a week ago I decided to take a look at one element of it that is often neglected: The bye week.
Of course there's often a few comments about it, though it receives much less attention than the actual games. It led me to ask myself, does it really matter when you have your bye? Does it make any difference at all?
I've seen the arguments made for both early and late bye weeks. I've always found myself in the late camp. I tend to believe that it makes more sense to rest physically after several weeks of wear and tear. Considering that players weren't under such stress for the last 6-8 months, you would think they could handle the first few weeks pretty well. The late bye gives us a chance to recharge the batteries closer to the end of the season, closer to the playoffs.
In addition I've always felt that the Bye week is a good time to make changes. And it makes sense to make those changes after a few more weeks of evaluation to avoid knee-jerk reactions early in the year.
But I got curious as to how correct my hypothesis is. And so I've decided to test it by pulling up some data.
Dive on in and see what you think.
I want to measure how the timing of the bye week relates to success. And the most successful teams are Super Bowl teams. I decided to start there. I took the Super Bowl winner and loser from the 2000 to 2011 seasons, found their bye week, and averaged the byes of the super bowl winners and losers. I emphasize the word seasons here, please don't get confused with the calendar year as we're considering both regular season and super bowl teams.
it's good to look at the bottom too. I decided to take all the teams that finished in last place also to see if the bye week can also have a negative effect based on its location.
I calculated a few numbers, the average as I mentioned, and also the # of byes below 7, which is what I consider an 'early bye'. Number of byes between 7-9 (middle byes), and number of byes above week 9 (high byes). This would correspond to the number of games in a season and where you take your bye.
So here are some of the numbers, first the average bye week of the teams selected:
from 2000-2011 (12 teams):
Super Bowl Winner average bye: 8.33
# of Low Byes: 5
# of middle byes: 2
# of High byes: 5
We can see high and low are equal, but some of these byes were very very high, the Patriots and Ravens in 2001 and 2000 had weeks 16 and 14 respectively, there were 3 week 10 byes also. So our high byes were very high, and there were very few middle byes.
Super Bowl Loser:
Super Bowl Loser average bye: 6.56
# of Low Byes:5
# of middle byes: 6
# of High byes: 1
Interestingly, we see the same number of low byes. But we find a LOT more middle byes and only one high bye. That high bye was a week 10 Patriots time that was undefeated going into the 2007 super bowl. So not too shabby. I feel like this may be a slight confirmation of the hypothesis. A team with a late bye would be more likely to endure that last game.
Last place team average bye: 6.75
# of Low Byes: 6
# of middle byes: 5
# of High byes: 1
Last place teams had a higher number of low byes, and the numbers were quite similar to the Super Bowl Loser's numbers.
So what can we gather from that information?
Well the first thing I would say is that the bye week seems to have a minimal impact on the season's outcome. Obviously looking at the numbers between the 2nd place team, and the last place team, we see its virtually identical. Clearly the bye week isn't the culprit for their success or failure. And obviously the Super Bowl losers and winners are very close competitors and are not separated by a whole lot.
So I wouldn't say that having a higher bye guarantees you a better record, nor does having a low bye, as the numbers are nearly the same across the board.
But there is something interesting, something which may be of importance to winning it all. Considering that the first two numbers were teams that played against eachother, we can see that perhaps the extra rest DOES matter in winning the Super Bowl. After 3-4 more weeks of playing games, that late rest could prove to be a helpful factor. We can see that 5/6 high bye teams won the super bowl from 2000-2011.
To further look at this theory, I decided to compare the head to head matchups themselves. Did the team with the higher bye win more often?
The result was a 5-7 record for the team with the higher bye. So this would seem to disprove the theory. However I thought it would be interesting to look at the average margin between byes for the victor/loser. Certainly one week between the two teams wouldn't make as much of a difference as a several week gap.
My results confirmed this idea:
The 7 times the team with the higher bye lost, their bye was only 1.7 weeks more than the winning team's bye. In 6 cases, it was either 1 or 2 weeks more.
In the 5 times the team with the higher bye won, their bye was on average 6.6 weeks more than the losing team.
This would seem to suggest that having a much higher bye than your super bowl oppenent lends a heavy advantage. Whereas having a small bye advantage translates to a more equal game, with little to no competitive advantage.
However we all know that not all super bowl teams are created equal. So I decided to find Super Bowl matchups where the teams had close records. I took teams that were within 2 wins of eachother. So for example, our 2010 super bowl is counted, but the 2008 one is not.
The idea here is that if two teams are more equal, then a bye advantage could be more accurately measured by limiting the variable of team quality.
For 2006, the Colts (12-4) beat the Bears (13-3) with a -1 advantage
For 2005, the Steelers (11-5) beat the Seahawks (13-3) with a -4 advantage (note: the Seahawks really suck, this was also the largest bye advantage for any losing team).
For 2004, the Patriots (14-2) beat the Eagles (13-3) with a -2 advantage.
For 2002, the Bucs (12-4) beat the Raiders (11-5) with a +7 advantage
For 2000, The Ravens (12-4) beat the Giants (12-4) with a +6 advantage
What we can see here is that when there is a large positive difference, that team tends to win. The Seahawks lost with a +4 advantage, that being the worst bye advantage loss. On the other hand the teams with a very large bye advantage won in all cases.
For the record, the Pats in 2001 were 3 games worse than the Rams, but won with a whopping +8 advantage. Perhaps that explains how they were healthy and fresh for those last games, whereas the greatest show on turf couldn't keep it up. They also had a +7 advantage when they beat the panthers in 2003, but were also 3 games better than they were.
My conclusion here, by the numbers, is that the bye week won't have much of an effect on your season.
We could probably analyze this infinitely, but I've already spent 2 hours on this. I feel that the data is pretty scattered when assessing season results.
However, when examining the final game of the season, it seems to me that the team with the latest bye week is more likely to win, especially when that bye week is considerably later than their opponents. We can even see a few cases where equal or lesser ranked teams pulled out a win with a very large bye differential. We see 2 blowouts in 2002 and 2000 with a high bye differential, a solid win in 2010 with a high bye differential, and no blowouts with low differentials. Low differential games tended to be more even, with several going to the last drive (Steelers against the Cardinals had a -1 differential, same with the Pats in 2007).
To me this indicates that the healthiest team in the Super Bowl has the best chance of winning. And the closer your rest was to the playoffs, the more likely you are to secure that win.
This year our bye is set at week 4. My results would say we'll still probably have a standard awesome Steelers season. Should we make it to the big game, don't fear nation: The Steelers have gone to their last 3 super bowls with byes before week 7. And in 2005 we had a week 4 bye as well... Just sayin :)
So nation, what do you glean from this analysis? How do you feel about the timing of the bye?