Pittsburgh Steelers at Kansas City Chiefs - A Case of the Bye Week Blues?

Well, our beloved Steelers came perilously close to laying a stinker of an egg at Arrowhead Stadium Sunday night. It was almost exclusively because of the nearly-stellar play of the defense that the men in black and gold didn't leave Kansas City hanging their heads in shame. And, to be honest, they didn't seem to be bubbling over with pride after the game.

As usual, I'm looking for things that indicate that this ugly win isn't the start of something even uglier, a la 2009, except that we didn't give up an opening kickoff return for a TD. Perhaps we should be giving up some belated thanks for the new kickoff rules. Naturally I'm hoping that there is a way to put a more positive spin on the results than it being entirely due to moving up the kickoffs.

As I gave it some thought, I remembered that, at least earlier in the season, teams coming off their bye week had lost in practically historic numbers. So it seemed worth looking at the results this season.

All teams have now had their bye week. 16 teams won their game after the bye and 16 lost. The games after the first bye week (Week 5) certainly show why I had the impression that teams were losing after the bye—only one of the six teams with a bye in Week 5 won their game in Week 6. The following week continued the trend, with only two teams winning and four losing.

Week 8 looked much better for the bye teams—five out of the six teams won. Week 9 had four out of six winners, but Week 10 was back to only one team winning out of the four with bye weeks. Week 11 had no bye teams, and Week 12, the week after the final bye week, has two winners, including the Steelers, and two losers.

But it isn't really enough information to know who won or lost—one needs to know at least something about the relative records at that point in the season of the two teams playing any given game. I went through and compiled the information. The chart should be pretty obvious for the most part. The winning team is in bold. "Home" means that the bye team was playing at home—if there is no x in the box, they weren't.  All figures are for that point in the season. The order is by apparent strength of win or vice versa, with Buffalo apparently being the strongest opponent to lose a game and Carolina being the weakest opponent to win one.  

As one looks more closely, only five of the 16 teams that won their games after their bye week were playing a team with a winning record at that point in the season. Three of the winning teams played teams with an even record, and the remaining eight played teams with a losing record. Four of the eight bye teams played opponents that had not won a game at that point in the season.

Eight of the teams that lost after their bye week were playing teams with a winning record, five were playing teams with a losing record, and three were playing teams with an even record. This is almost suspiciously symmetrical, but that's how it happened.

Given the information one had at the time of each game, six teams that would have been expected to win lost. Two teams that would have been expected to lose won. Four of the games looked too close to call. Three of those were wins for the bye team. Two were at home, and the third was what might be an anomaly—Tim Tebow's first game vs. Miami, at Miami. The fourth was a loss for a home team, so that perhaps should go in our unexpected losses column.

So there were seven unexpected losses, over three times as many as the unexpected wins.  There are also wins by considerably less than expected. The Steelers' win on Sunday night was of course in that category, and I think you could reasonably put Green Bay, Houston, San Francisco, and the Giants there as well.

Two cases in which the final score was more lopsided than one would expect are the Kansas City and Baltimore games. In both cases the result can be at least partially laid at the feet of injuries. The Kansas City game was the first game Carson Palmer played with Oakland, and in the Baltimore/Houston game Matt Schaub sustained the injury that would put him on IR the following week.

So the question is, if there really is a "bye week" effect, is it different from previous years, and if so, what is it a function of? Well, the obvious factor is the new regulation in the CBA that players must be given four consecutive days off. As Abby Geisel argued four weeks ago in this article, 

Coaches have regarded a bye week as a time to correct persistent problems with extra drill time, but now, they actually lose that valuable time...

Additionally, teams rely on consistency to carry them through adversity, and they bye totally disrupts that flow. It becomes obvious with a team like the Patriots. They were 5-1 coming into the Steeler[s] game (their one loss being early in the season to Buffalo). Both the Steelers and Patriots were riding winning streaks into Sunday's game, but the major difference was that the Pats were coming off of a bye and the Steelers were not. The Steelers had a normal week of practice and preparations. The Pats had a choppy, interrupted week. Maybe the Steelers just played a great game and are truly hitting their stride. On the other hand, maybe the Pats bye week woes really changed the outcome of the game.

The always-entertaining Bill Simmons wrote a "Q & A" article last season that is worth checking out. He noted

Successful coaches tend to thrive after bye weeks. Andy Reid? 12-0. Bill Belichick? Eight straight wins by an average of 15 points. Jeff Fisher? Nine of his past 12. Mike Tomlin? Four of five. Peyton Manning? Eight for nine. (Wait, he's not Indy's coach?) Does it make sense that well-prepared teams are more dangerous with an extra week to prepare? Of course.

In an October 2009 article in the New York Times, George Bretherton noted that the overall win percentage since bye weeks were introduced in 1990 is 52.7%. At that point in 2009 teams were 4-0 after their bye. Out of curiosity I checked what happened the rest of that season, and the season ended up at a perfect 50%. Unfortunately, however, after the Week 7 bye four of the six teams with a bye played each other, which possibly skewed the results. I don't know if that is common, but certainly this year no bye teams played each other. 

Last season 20 teams coming off their bye won, and 12 lost. Once again four teams played each other, giving a 50% split for two games. The win percentage, averaged over 2009 and 2010, is 56.3%. This is not taking into consideration the record of the opponents or any of the other things I calculated for this season—it is strictly wins and losses.

I suppose it is too early to tell whether the modest advantage teams seemed to enjoy after their bye week in prior seasons is going to disappear. It also doesn't really tell us whether the Steelers' sloppy game last Sunday is attributable to the bye week, to the Steelers overlooking Kansas City on their way to the Bengals and 49ers, some unknown factor or factors, or all of the above.

I think it is interesting that nobody seems to be attributing it to our quarterback playing with a broken thumb. Everyone seems to accept as a given that Ben plays as well or better when he's hurt than when he's not. And frankly, several of his incompletions can be attributed to the receivers dropping perfectly good passes. The infamous timeout and the boneheaded pick are just the typical Roethlisberger brain farts we've all come to expect once or twice per game.

Hopefully the coaching staff will send the wide receivers out to catching drills along with the DBs this week. The DBs clearly put some time in last week. Maybe the coaches will do a few time management drills themselves, and Ben will spend some time reviewing the differences between the San Francisco and Pittsburgh jerseys. Hopefully this will lead to a glorious win in San Francisco.