Drinen's Notebook: Wednesday, December 12, 2002

What kind of teams are successful in the playoffs?

Once again this week, I'm serving up a column that has little to no fantasy value. This is about real football, and the real playoffs that will be upon us very shortly.

The Old-School Book of Football Cliches has taught us that the formula for playoff success is to build a team with good defense and a good running game. It also helps to have lots of playoff-tested veterans. And for goodness sakes, make sure you secure home field advantage. This article is an attempt to put these cliches (and more) to the test. Is it really true that teams with good defenses do better in the playoffs than teams with good offenses? In playoff football, is a strong running game more important than a strong passing game? How important is home field advantage? What about opportunistic teams (i.e. those with a good turnover margin)? How do they tend to fare in the postseason? How about teams with lots of playoff experience? How about teams that have played well in close games? How about teams that are hot coming into the playoffs?

To try and get a handle on these questions, I looked at every playoff game from 1983 to 2001. Because Super Bowls really seem to me to be a different animal than the rest of the postseason (bye week for both teams most years, neutral site, unimaginable hype, odd start time, etc.), I left them out of the analysis. So this article should be viewed as examining what kind of teams are best suited to get to the Super Bowl, not necessarily win it.

How important is homefield advantage?

During the regular season, the home team wins about 56 percent of the time. During the playoffs, the home team wins about 76 percent of the time. So homefield is more important in the playoffs than during the regular year, right?

Not necessarily.

The problem is that, in the playoffs, the home team is always the higher seed, and the higher seed is almost always the better team. The question is: how much of that huge advantage is due to being the better team, and how much is due to the homefield? It's impossible to say, but consider this: from 1983-2001, when two teams with identical records have met in the postseason, the visiting team has won more often. OK, so it's a razor-thin margin, and a smallish sample (29 games, 14 wins for the home team, 15 wins for the visitors), but very interesting nonetheless.

I'm certainly not going to try to convince you there is actually a homefield disadvantage, but this result is enough to call into question the claim that there is some sort of extra home field advantage in the playoffs. It certainly is possible that home field advantage is more important for some teams than others, but in general, there is simply no reason to believe that it's a huge factor.

Now, if it weren't for the NFL's realignment this year, it wouldn't matter anyway, because we could just treat the homefield advantage as being built into the advantage of having a better record (see the next section). But realignment changes all that. With the division winners being seeded ahead of both wildcard teams, there is a very good chance that at least one road team will have a better record than its host. For example, if the playoffs started today, the 7-5-1 Steelers would be hosting an 8-5 wildcard team. We'll have to wait a few years to find out if the home field helps, and by how much.

Stating the obvious

There's an old bit by some standup comic (maybe it was Steve Martin) who claims to have a foolproof plan for retiring rich. That grabs the attention of the audience, so he starts to unveil the plan: "First," he begins, "you get a million dollars..."

It's not foolproof, but the plan for building a successful playoff team is much the same: first, you build a good football team. Not rocket science, I admit, but the following fact is overwhelmingly clear: by far the number one determining factor in who wins a playoff game is which team had the better regular season record. As we saw in the last section, part of that may be due to home field advantage the team with the better record has, but check out these numbers:

When the two teams' records differ by one game, the team with the better record wins 67.2 percent of the time.
When the two teams' records differ by two or more games, the team with the better record wins 83.8 percent of the time.

That's remarkable. Teams that are two games (or more) better than their opponent have a combined record of 62-12 since 1983. This ain't baseball, where anything can happen in a short series and you can throw the records out the window. Well, I suppose anything can happen, but what actually ends up happening the overwhelming majority of the time is that the better team wins.

Immovable object vs. irresistable force

OK, so let's say you're not the team with the better record. What kind of team is best positioned to pull an upset? Or, to look at the flip side, what kind of team is less vulnerable to the upset: an offensive-minded team, or a team whose strength is defense?

First, let's get some terms straight here. Just to save some typing and avoid writing unnecessarily complicated sentences, I'll use the word underdog to refer to the team with the worse record, and the term favorite to refer to the team with the better record. This may or may not correspond to what the Vegas folks were thinking at the time, but like I said, it's just to save some ink. Also, in everything that follows, I'm defining offense in terms of points scored and defense in terms of points allowed. (I won't go through the details, but I've checked it out, and you get similar results if you use yards instead of points.)

As you read through the following facts, the important numbers to keep in mind -- the baselines -- are these: overall, the favorite wins 76 percent of all playoff games, and the underdog wins 24 percent.

When a team is a favorite and it has a better offense than its opponent, it wins 80 percent of the time.
When a team is an underdog but it has a better offense than its opponent, it wins 35 percent of the time.

So, as you would expect, a good offense helps you. If you're a favorite, then having the better offense makes you slightly more of a favoite (80 percent vs. 76 percent). If you're an underdog, having the better offense doesn't make you a favorite, but it makes you somewhat less of an underdog (35 percent vs. 24 percent).

What about defense?

When a team is a favorite and it has a better defense than its opponent, it wins 75 percent of the time.
When a team is an underdog but it has a better defense than its opponent, it wins 23 percent of the time.

This is surprising. According to these numbers, having a better defense doesn't help underdogs pull the upset, and it doesn't help favorites avoid the upset. So based on what we've seen above, it appears that offense is more important than defense to determining playoff success. Still don't believe it? Check this out:

In playoff games where one team has the better offense and the other team has the better defense, the team with the better offense wins 59 percent of the time. In those games, the better offensive team was the favorite 41 times, the better defensive team was the favorite 39 times, and the teams had the same record 18 times. So if offense and defense were equally important, we would expect a record of about 50-48 in favor of the offensive teams. But we get a record of 58-40 in favor of the offensive teams.

Based on this, I have no choice but to conclude that teams whose strength is on offense are better suited to win playoff games than teams whose strength is defense.

Ground game vs. aerial attack

We continue as above, defining the team with the "better rushing game" to be the one that gained more rushing yards, and likewise for passing. (Again, I also ran similar tests using things like yards per rush and yards per pass instead, but the results were the same.) And again, keep the baselines in mind: favorites win 76 percent of the time, underdogs win 24 percent of the time.

When a team is a favorite and it has a better rushing game than its opponent, it wins 80 percent of the time.
When a team is an underdog but it has a better rushing game than its opponent, it wins 31 percent of the time.

When a team is a favorite and it has a better passing game than its opponent, it wins 83 percent of the time.
When a team is an underdog but it has a better passing game than its opponent, it wins 33 percent of the time.

The passing-oriented teams seemed to have fared very slightly better, but not really enough to conclude anything definitively. Let's see what happens when a passing team meets a running team.

In playoff games where one team has the better passing game and the other team has the better rushing game, the team with the better passing game wins 48.5 percent of the time. In those games, the better passing team was the favorite 37 times, the better running team was the favorite 46 times, and the teams had the same record 20 times. So if rushing and passing were equally important, we would expect a record of about 54-49 in favor of the rushing teams. And we get almost exactly that: a record of 53-50 in favor of the rushing teams.

Conclusion: I find no evidence that rushing-heavy teams play better than passing-heavy teams in the postseason, or vice versa.

Other factors

In addition to the things we've already talked about, I tested just about every stat I could get my hands on: turnover margin, record in close (decided by 7 points or fewer) games, playoff experience (as measured by number of playoff games that team played in the past two years, which I realize is not a perfect indicator of a team's playoff experience, but it's the best I could do with the data I had), and late-season (week 10+) record.

Of those, the only ones that look meaningful are record in close games and late-season record. Record in close games is just barely worth mentioning, although teams that have done well in close games have done just a tad better than expected in the playoffs. Late-season record, on the other hand, seems to be a major factor.

When a team is a favorite and it has a better late-season record than its opponent, it wins 81 percent of the time.
When a team is an underdog but it has a better late-season record than its opponent, it wins 54 percent of the time.

So "hot" teams have done extremely well in the playoffs, even as underdogs. I suppose this isn't too surprising.

Putting it all together

So here's the recipe for playoff success:

INGREDIENT 1: be a good football team;
INGREDIENT 2: be playing well late in the year;
INGREDIENT 3: have a good offense;
INGREDIENT 4: have a good record in close games.

As near as I can tell, nothing else matters. And when I say that, keep in mind that I am not saying defense doesn't matter. Defense does matter. It is, in fact, hugely important in achieving step 1 of the plan: being a good football team. What I'm saying is that a good defense doesn't seem to be of any extra value in the playoffs above and beyond the fact that it helps you to be a good football team (whereas, surprisingly, a good offense does).

If I had to guess, I'd say that the recipe calls for about 10 parts of ingredient 1, 5 parts ingredient 2, 2 parts ingredient 3, and 1 part ingredient 4. Maybe before the season ends, I'll tinker around and try to come up with a more "precise" formula, although we're dealing with such broad general trends here, that I doubt making the formula more precise would accomplish anything (other than keeping me off the streets for a few hours, that is).

The Raiders are quickly becoming the overwhelming favorite in the AFC, and this analysis implies that that's a good bandwagon to be on. The Raiders have the best record in the AFC, and the best offense. And they've won five straight. Their only blemish is a less-than-stellar record in close games, but that's a fairly minor issue. On the NFC side, the Eagles, Bucs, and Packers are tied for the best record in the league. The uncertainty at QB complicates things, but the formula for playoff success outlined above seems to favor the Eagles slightly, on the basis of their current hot streak and the fact that their offense is essentially equal to the Packers' and comfortably ahead of the Bucs'. All the NFC contenders are within a game and a half of each other, however, so the pecking order could change drastically in the season's final three weeks.

Unless otherwise noted, all stats come from football-reference.com and the disclaimer applies