Monday, January 16, 2012

Home Field Stuff

Over the course of many years, I keep a data-base for various trends in the NFL game. This morning, let's update two of the most interesting, Home Field Advantage, Seeding, and the Turnover Battles:

Home Field Advantage and Seeding:

The table you will see below is a simple tracking of seeding and home field advantage in the NFL playoffs since the field was expanded to 6 teams per conference in the 1990 season.

As you might notice, the 2011 column only represents 8 of the 10 games that are played before the Super Bowl (since the final 2 games have yet to be played). In that 22 seasons, the best year for home teams was 8-2. This season, if New England and San Francisco both win on Sunday, that would put the home teams at an all-time high 9-1, which I suppose would suggest that home field means something again. But, given that in 2010, the home teams were 4-6 in the playoffs, that would likely be a rush to judgement and a knee jerk reaction we should resist.

What makes that 9-1 home record all the more remarkable would be that the lone loss would be 15-1 Green Bay losing at home as the #1 seed in the NFC. A quick scan of the NFC side of the table below will reveal that the #1 seed of the NFC Playoffs played in the NFC Championship Game every year from 1990-2006. But, in 2007, when the Cowboys lost in their 1st playoff game, the #1 seed started trending in the wrong direction. Since then, the 2008 Giants, 2010 Falcons, and 2011 Packers have all gone one-and-done.

YearHome RecordAFC SeedsNFC Seeds
19908-21* vs 21 vs 2*
19918-21* vs 21* vs 2
19925-52 vs 4*1 vs 2*
19938-21* vs 31* vs 2
19948-21 vs 2*1* vs 2
19957-32* vs 51* vs 3
19968-22* vs 51* vs 2
19976-42 vs 4*1 vs 2*
19988-21* vs 21 vs 2*
19997-31 vs 4*1* vs 2
20008-22 vs 4*1* vs 2
20017-31 vs 2*1* vs 3
20028-21* vs 21 vs 2*
20036-41* vs 31 vs 3*
20046-41 vs 2*1* vs 2
20054-62 vs 6*1* vs 5
20068-23* vs 41* vs 2
20075-51* vs 32 vs 5*
20085-52* vs 64* vs 6
20097-31* vs 51* vs 2
20104-62* vs 62 vs 6*
20117-11 vs 22 vs 4

Table Tutorial


(* = Super Bowl Team)

The other thing that is extra fascinating about tracking these numbers since the format change is that the Super Bowl representatives have also opened up quite a bit in the last handful of years. Between 1990 and 2004, there were no #5 or #6 seeds to play in a Super Bowl - let alone to win one.

Since then, it seems like anything is possible if you simply "make the tournament". Get in and if you can play the hot hand, you can go all the way. It started with the 2005 Pittsburgh Steelers, who were the #6 seed in a crazy year where 11-5 was the last spot to make the playoffs. They won 3 straight road games at Cincinnati, Indianapolis, and Denver and won the Super Bowl over Seattle in Detroit at Ford Field.

In 2007, we all remember the #5 seed New York Giants, who finished at 10-6, caught fire with wins against Tampa, Dallas, and Green Bay and go on a run all the way to Glendale, Arizona to defeat the 18-0 Patriots in Super Bowl 42.

And in 2010, the #6 seed Green Bay Packers (10-6) barely scraped into the playoffs before getting on a huge roll and winning games in Philadelphia, Atlanta, and Chicago, before edging Pittsburgh in Super Bowl 45 in Arlington.

So, it only makes sense that the same Packers squad would get the #1 seed and home field throughout and then see their title defense end in less than 3 hours to a team that finished the season 9-7. Of course it does.

One other item that is interesting to track is the cumulative record for the home team during this 22-year study. 148-70 (67.8%) is the overall home record during the stretch, but you see the decline over 5 year increments that would seem to coincide with free agency and further NFL parity. By 2005, home field had almost no bearing at all:

1990-1994: 37-13 - 74%
1995-1999: 36-14 - 72%
2000-2004: 35-15 - 70%
2005-2009: 29-21 - 58%
2010-2011: 11-7 - 61% (only 1 complete year)
Totals: 148-70 68%

You would still always prefer to have a home game when everything is on the line, but the league is so close that the opportunity for the visitor to win a big game is as good as it has ever been.

Turnover Battle:

Despite the innovations of the NFL game over the year, there is nothing more constant than the impact of the turnover. It is staggering and reliable that the turnover - even one - impacts the game more than any statistic on the face of the earth.

Every year, the rate of winners of the turnover battle also winning the game hit at 78.4%. Now remember, this number has no relation to home field, point spreads, health, caliber of QBs, or anything. Put Team A against Team B - makes no difference which is which, and see that the winner of the turnover battle, even if it is only by a +1 margin, will win around 4 of every 5 games.

So how does this affect the playoffs? Since 1990, the winner of the turnover battle has a record of 171-33 in the playoffs. This winning percentage is even higher, hitting at 83.8! The results are far more convincing than home field advantage or favorite or anything.

Here is that 22-year trend by 5 year blocks:

1990-1994: 43-4 - 85%
1995-1999: 38-8 - 83%
2000-2004: 41-7 - 85%
2005-2009: 39-9 - 81%
2010-2011: 10-5 - 67% (only 1 complete year)
Totals 171-33 - 84%

You can certainly see that last year in particular was a crazy playoff, where 4 teams won games where they lost the turnover battle, but that is an extreme rarity.

This past weekend, New York was a +3, Baltimore a +4, and San Francisco a +4. Denver actually won the turnover battle with New England, but that cannot erase 6 Tom Brady touchdowns. In the wildcard round, there was only one game with a turnover differential, and that was Houston at a +3 over the Bengals.

So, next time you hear about key stats, just recall that sometimes we over-complicate matters. The NFL has changed, but it still comes back to giveaways and takeaways. 84% of the time, that shows you the winner. A trend that is almost certain to be unmatched.

No comments: