View Full Version : The Payoff of Playoffs

01-25-2011, 11:55 PM
Joe Posnanski

We, as American sports fans, like endings. I think that speaks a little bit to who we are. We tend to think of September baseball games being more important than April games. We tend to think of heroics in the fourth quarter being more meaningful than heroics in the second. We tend to put more stock into great Sunday finishes in golf than great Thursday opening rounds. I think the vast majority of us believe in the fairness of playoffs over the fairness of extended excellence, the value of single-elimination games over the value of many weeks of consistent winning. Like I say: I think that speaks a little bit to who we are.

Letís start with a quick review of the NFL playoff system. This is the 21st year of the bye system as we know it in the NFL playoffs. Between 1978 and 1989 (not counting the 1982 strike season), there were only two games the first weekend ó the two games featured the four wild-card teams. There were only six divisions in the NFL then, so the six division winners would all get a first-week bye. The two wild-card weekend winners would match up with the six division winners in Week 2. In those years, wild-card í80 Oakland won the Super Bowl (winning at Cleveland in the Red Right 88 game) and the wild-card Patriots reached the Super Bowl before getting pulverized by the í85 Bears. Other than that, wild-card teams had fairly limited influence on the playoffs.

Starting in 1990, the NFL changed the system, adding one wild-card team to each conference. That meant that the division winner with the worst record in each conference stopped getting a bye and had to play a wild-card team that first weekend. Thatís when the system we know it began ó four teams got byes, the other eight (six which were wild-cards) did not.

From 1990 to 2001, teams that had byes the first weekend went 39-9 in their first playoff game. Thatís an 81% winning percentage. And that makes a lot of sense. Teams with byes SHOULD win a vast majority of the time, right? You have the best of the division winners, rested and playing at home, so they should win something like 80% of the time. It was set up so that the best teams during the season were given huge advantages. And those advantages paid off almost every time. Itís instructive to take a look at those nine games in which the bye team lost:

1992: Buffalo beat Pittsburgh 24-3.
The teams had the same record (11-5), but the Bills had to play the first weekend because of tiebreakers. They beat Houston that first weekend in the famous Frank Reich game, coming back from 35-3 in the second half. And they manhandled Pittsburgh; itís pretty clear they were the better team.

1993: Kansas City beat Houston 28-20
Again, the teams were pretty close during the season, Houston was 12-4, Kansas City 11-5. There always seemed something insubstantial about those run-and-shoot Oilers.

1995: Green Bay beat San Francisco 27-17
Two division winners again, both with the same 11-5 record. San Francisco got the bye because of tiebreakers. Green Bay had a young and ascending Brett Favre.

1995: Indianapolis beat Kansas City 10-7
Our first major upset, and people in Kansas City have never stopped thinking about it. Lin Elliott missed three field goals for the Chiefs.

1996: Jacksonville beat Denver 30-27
Our second major upset, and people in Denver probably have lived it down, since the Broncos won the next two Super Bowls.

1997: Denver beat Kansas City 14-10
The Chiefs had beaten the Broncos in the regular season on a last-second 54-yard field goal by Pete Stoyanovich to secure the division and the bye. And the game itself, like all close NFL games, has been dissected again and again in Kansas City (you can ask any obsessed Chiefs fan about phantom holding penalties and whether or not Tony Gonzalez was in bounds). The Broncos went on to win the Super Bowl.

1999: Tennessee beat Indianapolis 19-16
Both teams had gone 13-3 during the regular season, though the Titans had lost the division to the 14-2 Jaguars. They were very close in quality, I would say, and the game was very close. Peyton Manning, in only his second year, had a bad game and the rumblings about his ability to win playoff games would begin right around this time.

2000: Baltimore beat Tennessee 24-10
Like in 2000, the two best teams in the conference were probably in the same division. Tennessee had won the division with a 13-3 record. But you could argue convincingly that the 12-4 Ravens were better. In fact, the Ravens made a rather convincing argument on the field, and in the Super Bowl, too.

2001: Philadelphia beat Chicago 33-19
Two division winners, and though the Bears had the better record (13-3 to Philadelphiaís 11-5) and the teams had the same point differential, the Eagles really beat up the Bears in Chicago.

There was order in the NFL playoffs. Yes, there were upsets, but they clearly WERE upsets, things that did not happen often, things that usually happened for a reason.

In 2002, the system changed ó but it didnít seem a particularly big change. The league expanded to eight divisions. So that meant that there were now eight division champions instead of six. To compensate, the NFL wisely (methinks) eliminated two wild-card teams, going back to four. So that meant that there were still 12 teams getting into the playoffs. And four of those 12 ó the two division winners with the best records in each league ó got a first-round bye.

On the surface, it would not seem that this system should change much. The same number of teams were making the playoffs. Two of the wild-cards were replaced with division winners Ö but that just seems to be cosmetics. In 2002, everything looked about the same. The four bye teams all won their first playoff games, and by a total of 115-52. Only one of those games ó Tennesseeís 34-31 overtime win over Pittsburgh ó was even remotely close.

But something kind of bizarre has happened since 2003. That something might just be a fluke or a statistical anomaly, but that doesnít make it any less fascinating.

Since 2003, bye teams have gone just 18-14 in their first playoff games.

Since 2005, itís even more stark ó bye teams are just 12-12.

Think about that for a moment. Bye teams have:

(1) The best regular-season records.
(2) Home field advantage.
(3) An extra week to rest and prepare.

Thatís a pretty sizable advantage, isnít it? You take what looks like the superior team, you play the game at their stadium in front of their fans and you give them an extra weekís preparation. You would expect that team to win almost every time, wouldnít you? But over the past six years, the bye team has lost as many times as it has won.

Is this good for pro football? I would say largely that it is. I love the NFL playoffs. I love the randomness of it. The NFL is built around that Any Given Sunday credo, and the game thrives largely because of that. You really donít know whatís going to happen. But the question I think about, the question I want to ask here, is: WHY do we love that sort of randomness?

I bring up the BCS again. Lately, it feels like I have been arguing a lot in favor of the BCS, which is a weird thing because I donít like the BCS system, donít have any desire to argue for it, and I absolutely would prefer a well-designed college football playoff. My problem, I guess, is that I want to have a discussion, and it seems that almost nobody wants to talk about it. It seems like just about any time I bring up the question ó is a playoff really MORE FAIR ó I get yelled at, even by close friends. The BCS has been demonized past the point of absurdity, past the point where anyone even LISTENS when someone suggests that, hey, maybe itís not that bad.

Is a playoff really MORE FAIR? What does fair even mean? This year in college football, the BCS system had Oregon play Auburn for a trophy they called the national championship trophy. This left out other very good teams, particularly undefeated TCU. This wasnít fair. There was much griping about it, and rightfully so. It is absurd and somewhat arrogant to believe that we can use our eyes and our computer systems and our innate sense of the game to look at more than 100 Division I football teams playing somewhat self-determined schedules and simply pick the two best teams. The flaws in the system are obvious.

But arenít the playoff flaws obvious, too? This year in the NFL, the playoff system included a seven-win team and took one 10-6 wild-card team while leaving two other 10-6 teams at home. The system made a 12-win team and two 11-win teams go on the road for their first game, while three teams with 10 or fewer wins (including the NFLís first seven-win playoff team) played home games. This year, the NFL rewarded New England and Atlanta for their 14- and 13-win seasons by giving them an extra week to heal and home field advantage. This seems like a seismic advantage. But is it really? We cannot argue that they promptly lost convincingly ó making that one loss much more important than their stellar 16-game seasons. We cannot argue that 12 of the last 24 bye teams have lost their first playoff game.

There might not be any specific REASONS why bye teams have lost the last few years. It could just be one of those things. But I can think of a few reasons why it might be happening.

1. Thereís the NFL scheduling system. As you know, the scheduling system is intended to reward teams that had terrible years. In 2009, the Kansas City Chiefs went 4-12. As a result, their non-conference schedule featured these 10 teams (in parentheses Iíve included their 2009 records):

Cleveland Browns (5-11)
San Francisco 49ers (8-8)
Indianapolis Colts (14-2)
Houston Texans (9-7)
Jacksonville Jaguars (7-9)
Buffalo Bills (6-10)
Arizona Cardinals (10-6)
Seattle Seahawks (5-11)
St. Louis Rams (1-15)
Tennessee Titans (8-8)

The 10 teamsí combined record was 73-87. This was intended to be an easy schedule. It turned out to be even EASIER, because the five teams that were .500 or better on the list ó the 49ers, Colts, Texans. Cardinals and Titans ó ALL took huge steps backward in 2010. The Chiefs went 10-6, won their division, and beat one playoff team all year, that one playoff team being the 7-9 Seahawks. Were the Chiefs a lot better in 2010? Sure. How much better? Why donít we ask the question next year when the Chiefs play the four teams that remain in the playoffs (Chicago, Green Bay, Pittsburgh and the New York Jets) along with New England and Indianapolis.

This is how the system works. Lose and they try to ease your path. Win and they try to put boulders in your way. Scheduling is a big, big part of what the NFL calls parity. And so records can be illusions.*

*Since I wrote this several people have pointed out that since 2002, teams in the same division play 14 of the same games ó only two games are determined by how good or bad a team is supposed to be (the rest by divisions matching up with other divisions). So, for instance, the only schedule difference between the 2010 Chargers (who had won the division in 2009) and Chiefs (who had finished last) was that the Chargers played New England and Cincinnati while the Chiefs played Buffalo and Cleveland.

Itís a fair point ó and I missed it. Two games out of 10 non-division games is not insubstantial ó and if the Chiefs had played the Patriots instead of the Bills they probably would not have won the division. But the NFL schedule does not tilt as much as it once did, and so I would agree that this is not quite as big a factor as I had originally thought.

2. Homefield advantage seems to be losing some of its advantage in the NFL. For some of this, read Tobias J. Moskowitz and L. Jon Wertheimís fascinating piece on homefield advantage in the Jan. 17 issue of Sports Illustrated.* They point out that a big reason (the biggest reason?) for homefield advantage is unintended referee bias. Well, youíll have to read the piece.

*I just got their book, Scorecasting, and am ready to dive in. Iíll give you a full report and see if we can get Jon to come on here for a conversation.

But Ö with instant replay, the NFL might be taking a lot of unintended referee bias out of the game. Add in that officials surely bear down for playoff games (and they tend to be the best officials), and maybe homefield advantage isnít quite what is used to be. Maybe there arenít as many penalties called against the road team as there used to be. Maybe fumbles (or non-fumbles) that used to be called for the home team are overturned a little more. The speaker in the helmet thing seems to help, too ó it doesnít seem that teams are nearly as bothered by crowd noise as they used to be.

The numbers donít exactly bear this out ó teams ARE winning a little less often at home, but it doesnít seem earth shattering:

1970-79: 1087-813, 57.2%
1980-89: 1344-998, 57.4%
1990-98 (year before replay): 1239-839, 59.6%
1999-10: 1593-1205, 56.9%

Not a big deal. But the last five years, the numbers are down a bit more (56.1%). Anyway, with the playoffs you are dealing with small margins. In baseball, you hope and expect that over 600 plate appearances you will get something close to true value. But the NFL in many ways IS about small numbers. In the NFL, especially in the playoffs, one loss is devastating. Until 2005, road teams in the playoffs won 30% of the time. The last six years, they have won 45% of the time.

These are just thoughts, of course. I am not trying to suggest that they are right. Weíre just talking.

3. It seems like there is more REAL parity in the NFL than ever before ó not just the illusion of schedules, but a real tightening of talent. These would be the effects of the salary cap and the draft and various other things. It does seem true that there really isnít a lot separating the top two or three teams from the 10th or 11th teams.

The question, I think, is this: Whatís the competitive point of an NFL season? Is it to determine the BEST team in the NFL? Or is it to give us a fun and easy-to-follow trail on the way to our Super Bowl party? The New England Patriots won 14 of 16 games, including their last eight. They beat all four of the remaining playoff teams during the regular season (they also lost to the Jets in Week 2 during the regular season). In those four wins, only Green Bay even stayed close. They outscored their opponents by 205 points ó the best point differential in the NFL since New Englandís 16-0 season, and the second best in the NFL since 2001.

And on Sunday, after getting a weekís vacation and playing on their home field, they were obliterated by a New York Jets team that they had already played twice. The Jets had a great gameplan, and they played a sharp game, and Tom Brady looked confused, and the Patriots looked flat. And now their 2010 record is meaningless. Their season is mud. All the winning they did, well, nobody cares. That happened BEFORE the playoffs, before it really mattered. Is this fair? I think most of us would say that absolutely itís fair. We are a playoff nation. The Patriots lost on the field. Fair or unfair, either way, it made for good television.