In the third and final article of our series examining touchdown rate regression, we focus our attention on passing.
Heading in, we knew our sample size would be smaller than the ones we enjoyed in the rushing and receiving departments, but we still wrangled up enough three-year data to draw some conclusions. The results, however, were not as telling as the 90% success rate we saw in the other two pieces.
In creating our sample size, I pulled all passing data from the NFL over the last three seasons (2008-2010). I cut the sample down to quarterbacks with a minimum of 100 pass completions in a single season. 37 quarterbacks fit the bill in 2008, followed by 34 each in both 2009 and 2010. Note: in order to increase our sample size as much as possible, this study will use data from playoff games.
Going Down
To open, we will focus on only those players with a Touchdown/Completion rate at or above 8.0%. The league average in this department is 6.8%, so 8% is considered to be relatively high. Since 2008, 29 players fit the bill, but 12 pulled it off in 2010. That cuts our sample to 17. Additionally, two players didn’t accrue enough pass completions in their follow-up season to warrant examination, which leaves us at 15.
Year | Player | Com | TD% | Next Com | Next TD% | Change |
2008 | Philip Rivers | 353 | 10.5% | 344 | 8.4% | -2.1% |
2009 | Drew Brees | 435 | 9.7% | 487 | 7.2% | -2.5% |
2008 | Tony Romo | 276 | 9.4% | 392 | 7.1% | -2.3% |
2009 | Brett Favre | 406 | 9.4% | 217 | 5.1% | -4.3% |
2009 | Aaron Rodgers | 378 | 9.0% | 402 | 9.2% | 0.2% |
2009 | Eli Manning | 317 | 8.5% | 339 | 9.1% | 0.6% |
2008 | Aaron Rodgers | 341 | 8.5% | 378 | 9.0% | 0.5% |
2009 | Philip Rivers | 344 | 8.4% | 357 | 8.4% | 0.0% |
2009 | Matt Ryan | 262 | 8.4% | 377 | 7.7% | -0.7% |
2008 | Kurt Warner | 493 | 8.3% | 386 | 8.3% | 0.0% |
2008 | Drew Brees | 413 | 8.2% | 435 | 9.7% | 1.4% |
2009 | Peyton Manning | 480 | 8.1% | 468 | 7.3% | -0.9% |
2009 | Donovan McNabb | 285 | 8.1% | 275 | 5.1% | -3.0% |
2009 | Jay Cutler | 335 | 8.1% | 282 | 8.9% | 0.8% |
2009 | Alex D. Smith | 225 | 8.0% | 204 | 6.9% | -1.1% |
2008 | Tarvaris Jackson | 103 | 8.7% | 14 | 7.1% | -1.6% |
2009 | Kurt Warner | 386 | 8.3% | 0 | 0.0% | 0.0% |
The chart above shows each of our 17 (including the two exceptions) players completion total and TD rate during the season noted in Column 1. ‘Next Com’ and ‘Next TD%’ refer to the completion total and TD rate the player put up the next season. Finally, we have the ‘Change’ column, which shows us the difference in the player’s TD rate from Year 1 to Year 2.
The results are the least impressive we’ve seen so far, but we still get a regression rate of 67% (10-of-15). This means that, of the 15 players in our sample, 10 saw a drop in TD% the next year. Additionally, note that the four passers with a TD rate above 9.0% all regressed in year two.
Of the five misses, two are Aaron Rodgers, who saw his TD rate progress from 8.5% in 2008 to 9.0% in 2009 and 9.2% in 2010. Talk about an exception to the rule, although we should definitely expect a drop in 2011. The other misses were Eli Manning (09-10), Jay Cutler (09-10), and Drew Brees (08-09). Note that Brees did finally regress in 2010, but the jury is still out on Cutler and Manning.
Going Up
Next, we’ll focus on the players who struggled to score touchdowns, racking up a TD rate below 5.1%. Again, the league average is 6.8%, so anything under 5.1% should be considered low/unlucky. 21 quarterbacks fit the bill, seven of which came in 2010. Our sample drops even lower once we remove the four passers on our list who failed to throw a pass the following season. This leaves is with a low, but usable sample of 10 quarterbacks.
Year | Player | Com | TD% | Next Com | Next TD% | Change |
2008 | Ryan Fitzpatrick | 221 | 3.6% | 127 | 7.1% | 3.5% |
2008 | Jason Campbell | 315 | 4.1% | 327 | 6.4% | 2.3% |
2009 | Chad Henne | 274 | 4.4% | 302 | 5.0% | 0.6% |
2008 | Marc Bulger | 251 | 4.4% | 140 | 4.3% | -0.1% |
2008 | David Garrard | 335 | 4.5% | 314 | 4.8% | 0.3% |
2008 | Kerry Collins | 268 | 4.5% | 119 | 5.0% | 0.6% |
2009 | Jake Delhomme | 178 | 4.5% | 93 | 2.2% | -2.3% |
2009 | David Garrard | 314 | 4.8% | 236 | 9.7% | 5.0% |
2008 | Trent Edwards | 245 | 4.9% | 110 | 6.4% | 1.5% |
2009 | Kerry Collins | 119 | 5.0% | 160 | 8.8% | 3.7% |
2008 | Jeff Garcia | 244 | 4.9% | 0 | 0.0% | 0.0% |
2008 | Brian Griese | 110 | 4.5% | 0 | 0.0% | 0.0% |
2009 | Marc Bulger | 140 | 4.3% | 0 | 0.0% | 0.0% |
2009 | JaMarcus Russell | 120 | 2.5% | 0 | 0.0% | 0.0% |
The sample size here is awfully small, but we still see a regression rate of 80% (8 of 10). Half (5 of 10) of our sample ended the following season above our 5.1% threshold.
Our exceptions were Marc Bulger (08-09) and Jake Delhomme (09-10). Bulger didn’t regress in 2010, but he also didn’t play. Delhomme went from a struggling Panthers offense to a struggling Browns offense and actually saw a pretty steep drop.
2011 Losers
Year | Player | Com | TD% | Proj TD% |
2010 | Tom Brady | 353 | 10.8% | 9.0% |
2010 | Matt Cassel | 271 | 10.0% | 7.5% |
2010 | David Garrard | 236 | 9.7% | 7.2% |
2010 | Aaron Rodgers | 402 | 9.2% | 8.8% |
2010 | Eli Manning | 339 | 9.1% | 8.7% |
2010 | Ryan Fitzpatrick | 255 | 9.0% | 7.8% |
2010 | Jay Cutler | 282 | 8.9% | 7.8% |
2010 | Kerry Collins | 160 | 8.8% | UFA |
2010 | Michael Vick | 253 | 8.7% | 8.5% |
2010 | Josh Freeman | 291 | 8.6% | 7.5% |
2010 | Philip Rivers | 357 | 8.4% | 8.5% |
2010 | Joe Flacco | 347 | 8.1% | 7.5% |
Twelve quarterbacks make the cut in 2011. Each had a TD rate above 8.0% in 2010 and a whopping six were above 9.0%. Passing is on the upswing in the NFL, so we’re likely to see rising TD rates, but we did learn earlier that, over the 2008 and 2009 seasons, all four players who put up a 9.0%+ TD% saw a drop in Year 2.
My projected TD rates for 2011 show a drop for all but one (Philip Rivers) of the quarterbacks listed. We learned earlier that the regression rate will be closer to 67%, so players who I’m projecting close to, but lower than their 2010 rates could find themselves with an even higher rate. Some names to watch include Michael Vick, Joe Flacco, and of course, Aaron Rodgers.
2011 Winners
Year | Player | Com | TD% | Proj TD% |
2010 | Donovan McNabb | 275 | 5.1% | TBA |
2010 | Sam Bradford | 354 | 5.1% | 7.0% |
2010 | Brett Favre | 217 | 5.1% | N/A |
2010 | Chad Henne | 302 | 5.0% | 5.2% |
2010 | Colt McCoy | 135 | 4.4% | 5.0% |
2010 | Derek Anderson | 169 | 4.1% | UFA |
2010 | Jimmy Clausen | 157 | 1.9% | 3.5% |
Seven quarterbacks make the list for 2011. Each had a TD rate below 5.1% in 2010. Our 10-man-deep test from earlier told us to expect an 80% regression rate, which means that only one of the listed players should be expected to see a decline in 2011. Projections aren’t shown for retired Brett Favre, free agent Derek Anderson, and soon-to-be free agent/traded Donovan McNabb, leaving us with four projected 2011 TD rates. All show, at least, a small increase.
Final Thoughts
Combining the two tests we did today, we find a 72% regression rate. Our combined rate in the pass/rush experiment was 90%, so it’s clear that quarterbacks are better able to maintain extremely high/low TD rates. As mentioned earlier, I’m not in love with the sample size here, but I suspect that the coming years will show a similar trend. Nonetheless, when evaluating extremely high and extremely low offensive touchdown rates , it’s much safer to project regression for rushers and receivers than it is passers.
Follow Mike on Twitter: @PFF_MikeClay and be sure to follow our main Twitter feed as well: @ProFootbalFocus
[Editor's note: this article was previously published in our fantasy section where you can find much more outstanding work from Mike and the PFF Fantasy staff.]