How much did the defense contribute to the 8-2 finish?

Thursday saw the release of Football Outsiders Almanac 2010, the latest edition of the annual put out by Football Outsiders.  As I’ve mentioned on here before, I now write for FO, which means I wrote the Titans chapter in this year’s edition.  Over the next couple weeks or so, I’ll be writing a few posts on some of the things in that chapter, and some of the things that didn’t make into the chapter.  Just as a disclaimer, most of these posts will probably be heavily influenced by FO’s numbers.

When the Titans made the playoffs in 2007 and 2008, it was largely on the back of one of the league’s best defenses, and with every starter but Haynesworth back for 2009, it seemed reasonable to expect the same in 2009.  That clearly didn’t happen, as the Titans gave up 300 yards passing  and 20+ points in in 5 of their first 6 games in falling to 0-6.  The rest of the year, the Titans only gave up 300+ yards passing once and held 6 of their 10 opponents under 20 points.  Obviously, that improved performance was the result of the defense playing better, right?

Well, not really.

The Titans defense had 5 games in 2009 where it was either well above average or well below average, which I semi-arbitrarily define as 25% away from average (DVOA of +/- 25.0%).  By point of comparison, the top defense in the league last year was the Jets, who finished 23.4% better than average, and the ’08 Titans had a defensive DVOA of -16.6% (for defense, negative good, positive bad).  Three of those games, it was particularly bad.  As you might guess, two of those games came as part of the 0-6 start: Week 4 at Jacksonville and Week 6 at New England.  In the other four pre-bye games, the Titans had essentially a league-average defense.

The third game it was particularly bad was after the bye, and unsurprisingly was the Week 16 loss against the Chargers.  The two games where it was particularly good also came after the bye: the home wins against the Jaguars and the Rams.  In six of the other seven post-bye games, though, the Titans defense scored as below-average according to DVOA.  (Oddly, Miami was the only above-average game.)  Overall, for the 8-2 stretch, the Titans had a slightly below-average defense, roughly equivalent to the Texans’ at 19th, and in the seven non-extreme games had a below-average defense, roughly equivalent to Oakland’s 24th-ranked defense.

So, does this mean the Titans’ defense played as poorly those last 10 games as it had the first 6?  Again, not really.  There are two main extenuating factors as to why there was an apparent defensive improvement.  First, as I’ll write about in a future post, the offense was more productive, so the defense tended to start off in better situations.  Second, the “D” in DVOA stands for “Defense-adjusted”; that’s a slight misnomer in this case, since we’re talking about defense, but the key part is “adjusted.”  The Titans defense looked better because it did better.  The reason it did better, though, was they played worse teams.  4 of the first 6 games came against teams in the top 11 in offensive DVOA.  Only 3 of the last 10 did, plus the 4th best offense the Titans faced was Arizona, who played without Kurt Warner.

The bad news, though, is that, while less constant than offensive DVOA, defensive DVOA tends to be a better predictor of the subsequent year’s performance than how the defense looked.  If the defense plays the same in 2010 as it did in 2009, it’ll be below-average again, not average or a little above like it looked the second half of 2009.  With the turnover in defensive personnel, likely including at least 5 new defensive starters from Week 1 against Pittsburgh last year, we shouldn’t expect the defense to be at exactly the same level as it was last year, that the Titans were below average on defense for all of 2009 does suggest that if they want to return to the playoffs, it’ll have to be the offense that does it.

Quantcast