NFL Playoffs: What Seeding History Really Says About Your Team’s Chances

By Scott Kacsmar

Cold Hard Football Facts’ Comeback King (@CaptainComeback)

While watching Notre Dame get crucified 42-14 in the BCS National Championship against Alabama, someone on Twitter amusingly pointed out the little “1” written next to the Fighting Irish, and the “2” next to the Crimson Tide in ESPN’s static scoreboard image. 

Fortunately, rankings mean a bit more in the NFL playoffs. They are playoff seeds in a real playoff system, and they determine home-field advantage, which is obviously huge this time of year.

Teams fight for first-round byes every year, though we are seeing an increase in lower seeds winning championships, as the No. 1 seeds continue to choke in January. There is a reason for that, which we will look at along with all the results for the various seeding matchups.

Last week’s Wild Card round saw all four favorites advance, but do not expect a repeat this week. The great Divisional round is one that almost always produces at least one shocking upset. The 2004 season was the last time all four home teams advanced, and that was one of the most top-heavy seasons in history.

This week features two classic matchups of a No. 2 seed (San Francisco and New England) hosting the No. 3 seed (Green Bay and Houston), while the No. 1 seeds (Denver and Atlanta) will host a No. 4 (Baltimore) and No. 5 (Seattle).

Think the top seeds have it easier in that case? Not so fast. Since 1990, a No. 2 seed is 21-7 (.750) against the No. 3 seed, while the No. 1 seed is a combined 22-9 (.710) against No. 4 and No. 5 seeds.

That winning percentage for the No. 1 seed would actually decrease if adding games against the No. 6 seed. Intrigued yet? This is exactly the kind of seedy data we are going to break down today.

The expectations for each seed (1990-2011)

Expanding to a 12-team playoff format in 1990, that is the starting point for our data. Looking at the last 22 seasons, which includes 264 playoff teams and 242 playoff games, here is the breakdown of success (44 teams for each seed).

The No. 1 seeds have gone one-and-done two more times than the No. 2 seeds, but still have reached more Super Bowl (21 to 12). At least one No. 1 seed has appeared in 18 of the last 22 Super Bowls. However, only two No. 1 seeds have won a Super Bowl since 2000, and one of those came in a matchup between the top seeds in 2009. Someone had to win.

Here is a table showing the percentages of three milestones, such as going one-and-done in the first game you play, reaching at least the Conference Championship, and reaching the Super Bowl.

It is not talked about much, but it really does not pay to be the No. 3 seed. Only the 2006 Colts and 1987 Redskins have ever won the Super Bowl from that seed. It is difficult as you would expect to have to beat the top two seeds on the road (Colts only beat No. 2 Baltimore), which is why it was such a disaster for Houston to slip to that spot this year.

Here are the average regular season stats for record and scoring, along with the playoff record for each seed.

No surprise teams who earn byes win a higher rate of games, score the most points and allow the least. They also do the best in the postseason. What is interesting is that the No. 4 seed has had more playoff success than the No. 3 seed. However, there is a good reason for this.

The NFL used to only have three divisions in each conference (East, Central and West). This went on through 2001, meaning a little more than half the data includes seasons where the No. 4 seed could have had the second best record in the conference.

For instance, the 1999 Tennessee Titans were 13-3, but had to settle for the No. 4 seed because Jacksonville (14-2) won the division. But since the league went to 32 teams split up in eight divisions, a team like Tennessee actually would be the No. 5 seed today. Due to this split, let’s take a look at the data broken up into two parts: 1990-2001 with the three divisions, and 2002-2011 with four divisions.

This table is a gold mine. While the top seeds are even greater in the regular season since 2002, their playoff success is much worse. The No. 1 seed won 69.1 percent of its playoff games in 1990-2011, and that is down to 56.1 percent since 2002.

We also see the No. 4 seeds are weaker now that they are the worst division winner – think 2010 Seahawks and their 7-9 record or last year’s Broncos – instead of possibly the second best team in the conference. They win 60 percent of their games instead of 68 percent, and their playoff success has gone down.

Finally, we see the improvement in the Wild Card teams. They win more games by bigger margins in the regular season since 2002, and the playoff success is much better. After going just 15-48 (.238) from 1990-2001 in the playoffs, the last two seeds are 31-37 (.456) since 2002. Amazingly the No. 6 seed is 18-18 thanks to two Super Bowl wins by the 2005 Steelers and 2010 Packers.

This really sums up the increasing competitive nature of the league:

• From 1990-2001, the playoff success for the six seeds ranged from 22.6 percent to 69.1 percent (46.5 percent gap).
• From 2002-2011, the playoff success for the six seeds ranged from 40.6 percent to 56.1 percent (15.5 percent gap).

Head-to-head results for different playoff seeds

Setting aside the Super Bowls for now, here are the results of the 220 playoff games since 1990 where there was a home-field advantage. The cells in red are the road records, which are the inverse of the records for the higher seed (home team).  

When the No. 1 seed hosts the Conference Championship against the No. 2 seed, the record is 14-8 (.636), which is just under the overall home-field advantage in these games, which was 149-71 (.677).

No. 4 Baltimore at No. 1 Denver: Before we worry about a Patriots/Broncos rematch, these teams have business to take care of this week. In the No. 4 at No. 1 games, the top seed is a solid 15-6 (.714). But it was just a year ago the No. 4 Giants (9-7) went into Green Bay (15-1) and shocked the Packers 37-20.

No. 5 Seattle at No. 1 Atlanta: Not a frequent occurrence, the top seed is 7-3 in these matchups. Seattle is better than your average No. 5 seed, though they will have to deal with the early start time on Sunday against an Atlanta team that was the last unbeaten (8-0) this year.

What is really bizarre is the 11-5 record the No. 1 seed holds over the No. 6 seed. Even more bizarre is the fact the No. 6 seed is 5-1 in the last six meetings ever since the 2005 Steelers broke through and took down the Colts. Look at this breakdown:

• From 1990-2004, the No. 1 seed was 10-0 against the No. 6 seed, with an average margin of victory of 19.1 points.
• Since 2005, the No. 1 seed is 1-5 against the No. 6 seed, with an average margin of defeat of 10.4 points. Only the 2009 Colts held up with a 20-3 win over Baltimore.
• These include devastating losses like Green Bay’s 48-21 dismantling of Atlanta (2010), and the Jets stunning New England a day later (28-21) just a month after losing 45-3 in Foxboro.

No more No. 6 seeds in this year’s playoffs, but remember this in the future. Houston and Green Bay took care of the No. 6 seeds this season, and now they must go on the road in premiere matchups against the No. 2 seeded Patriots and 49ers.

No. 3 at No. 2: You would expect these matchups to be common, which they are as they make up a leading 31.8 percent of all Divisional games since 1990. But it is the home-field advantage coming through 75 percent of the time (21-7). That is good news for San Francisco and New England.

In terms of the Wild Card games, the No. 6 seed has actually fared a little better (36.4 percent) over the No. 5 seed (31.8 percent) in pulling off the road upset. However, this is aided by the shift from three divisions to four and the weakening of the No. 4 seed.

• From 1990-2001, the No. 4 seed was 18-6 (.750) against the No. 5 seed, while the No. 3 seed was 17-7 (.708) against the No. 6 seed.
• From 2002-2012, the No. 4 seed was 13-9 (.591) against the No. 5 seed, while the No. 3 seed was 13-9 (.591) against the No. 6 seed.

There has yet to be a Conference Championship game between the two Wild Card teams, but maybe some day. The epic championship game between Tom Brady and Peyton Manning in 2006 was the only time a No. 3 hosted the No. 4 seed. Two years later the No. 4 Cardinals hosted the No. 6 Eagles in the NFC Championship, which is the only game of that seeding battle.

For your interest in distilling the idea of parity or just as a great reference point, here are tables showing the playoff seeds for each season with a summary of how many new teams from one year to the next.

The eight-division playoff system:

The four new playoff teams in 2012 ties 1995 for the fewest in this era, and the AFC became the first conference to repeat all four division winners and the same final four in the playoffs.

Super Bowl teams based on seeds

We leave you with a table of the 22 Super Bowl results since 1990. The winner had an average seed of 2.4, while the loser was the 1.8 seed. Conference strength may be a better predictor than seed in the Super Bowl.

Like two of the home teams this week to go the distance? Consider that only three of these Super Bowls featured both No. 1 seeds, while only one of the last seven featured two teams who had a bye week.

The importance of a playoff seed is just not what it used to be in the NFL. Expect another shining example of that this postseason.

Scott Kacsmar is a football writer/researcher who has contributed large quantities of data to Pro-Football-Reference.com, including the only standardized database of fourth quarter comebacks and game-winning drives. You can visit his blog for a complete writing archive. Please send any questions or comments to Scott at smk_42@yahoo.com, or you can follow him on Twitter at @CaptainComeback.