You Are Here > Pro-Football-Reference.com > PFR Blog > NFL and College Football Analysis

«
»

More on interception rates

Posted by Chase Stuart on March 1, 2010

Jason and I have written about interception rates before. Over at Footballguys.com, I concluded that interception rates were almost entirely random from season to season. Jason concluded that among the most popular stats, interception rate was by far the least consistent when a quarterback changed teams, indicating that either interceptions are largely random and/or aren't mostly the fault of the quarterback. He reached essentially the same conclusions when he analyzed what happened when teams changed quarterbacks.

Brian Burke essentially agrees with us, although in his win probability prediction model he includes interception rates. He also believes that interception rates are less predictable than other passing stats, but he does note some correlation between past and future interception rates when looking from half-season to half-season. All of the work Jason and I have done with respect to interception rates was across seasons; perhaps we'd see different results if we look at splits within the same season? That's what this post is about.

I looked at every 16-game season in NFL history and noted every QB who threw at least 150 passes in his team's first 8 games and at least 150 passes in his team's last 8 games. There were 533 such quarterbacks. I then calculated each QB's interception rate, relative to league average, in the first and second halves of the season. So how closely correlated were INT rates in the first and second halves of the season? Not very. The correlation coefficient was just 0.12, which shows a very small relationship between the two numbers. The R^2 was just 0.01, which means that (in loose mathematical terms) only 1% of a QB's interception rate in the second half of the season can be predicted by his interception rate in the first half of the season. Or in other words, they're almost entirely random.

When I spoke with Burke about this, he mentioned that he had looked at just the 2006 and 2007 seasons because that was all the data he had on hand at the time, and it's possible that his numbers were inaccurate due to a small sample size. That appears to have been the real culprit -- using the same data set but only looking at QBs during those two seasons brings the correlation coefficient up to 0.40, indicating a much stronger relationship. But in 2008 the correlation between first half INT rates and second half INT rates was again negligible, leading me to believe that 2006 and 2007 were just random occurrences and not the result of a fundamental shift in football.

So does this mean predicting who will throw an INT next week is impossible? The short answer is, probably. I looked at the correlation coefficients with some other metrics and only completion percentage appears to be correlated with future INT rate. However, the relationship is small -- a QB who completes 70% of his passes instead of 60% of his passes will probably only throw one fewer pick every 200 attempts.

I also looked at something I'm calling "interceptions per incomplete pass" which is just what it sounds like -- how many interceptions a player has divided by the difference between his attempts and his completions. Because completion rates have risen steadily while interception rates have dropped, the league average INT/IP rate hasn't changed as dramatically as the league average interception rate. From 1978 to 1992, the rate steadily dropped, but since then the league average INT/IP rate has hovered between 7.5 and 8.0. I decided to then look at QB numbers, without adjusting for era, from 1993 to 2008.

From 1993 to 2008, the correlation coefficient between raw INT rates in the first and second halves of the season (ignoring era adjustments, which slightly inflates the correlation) was 0.18. But the correlation coefficient between INT/IP in the first and second halves of the season (ignoring era adjustments) was 0.12. What's all that mean? There is only a small relationship between a quarterback's interception rate in the first and second halves of the season, and part of the explanation for the correlation has to do with how many incomplete passes a player throws. Once you look at just his INT rate on incomplete passes, it's even harder to predict which QBs will throw a lot (or few) interceptions in the second half of the season.

The conclusion, as it's always been when I've looked at the issue, is that interception rates are almost entirely random. And at least a sizable portion of the explanation when interceptions are not random has to do with two things: a quarterbacks' general inaccuracy and the sort of system in which he plays. Quarterbacks who are inaccurate will probably throw more interceptions than quarterbacks who are accurate. Similarly, quarterbacks who play on high risk/high reward offenses or on bad teams forced to throw in obvious passing situations will probably throw slightly more interceptions than the average quarterback will have on his stat sheet.

It's also important to remember that interceptions are incredibly rare events. Baseball fans like to say that the difference between a .250 batter and a .300 batter is just one hit every 20 at-bats, which would be very difficult to perceive if we didn't track stats. Well the difference between a high interception guy and a low interception guy is usually about one interception every 35 pass attempts, which is practically impossible to judge just by watching. We'll remember the momentum changing interceptions and the ones at the ends of games, but for the most part, most quarterbacks throw interceptions at pretty similar rates.

I'm reprinting the full dataset below of the 533 quarterbacks reference above. As an example, here's what the first line says. Jason Campbell, in 2008, threw 230 passes in the early half of the season (the first 8 games) and 0 interceptions, and an early-INT rate of 0.0%. That rate was 2.7% better than league average for the first half of the season. In the second eight games, Campbell threw 276 passes and was intercepted six times, for a late-INT rate of 2.2%. That was 0.8% better than the league average for the second half of the season. Campbell went from being 2.7% better than average to just 0.8% better than average, so he declined by 1.9%.

QB year att_E int_E rt_E avg_E att_L int_L rt_L avg_L diff
Jason Campbell 2008 230 0 0.0 2.7 276 6 2.2 0.8 -1.9
Jeff Garcia 2008 157 2 1.3 1.4 219 4 1.8 1.1 -0.3
Kerry Collins 2008 207 3 1.4 1.2 208 4 1.9 1.0 -0.2
Donovan McNabb 2008 288 4 1.4 1.3 283 7 2.5 0.5 -0.8
Chad Pennington 2008 242 4 1.7 1.0 234 3 1.3 1.7 0.6
Kyle Orton 2008 244 4 1.6 1.0 221 8 3.6 -0.7 -1.7
Tyler Thigpen 2008 151 4 2.6 0.0 269 8 3.0 0.0 -0.1
JaMarcus Russell 2008 216 4 1.9 0.8 152 4 2.6 0.3 -0.5
Aaron Rodgers 2008 262 5 1.9 0.8 274 8 2.9 0.0 -0.7
David Garrard 2008 263 5 1.9 0.8 272 8 2.9 0.0 -0.8
Eli Manning 2008 250 5 2.0 0.7 229 5 2.2 0.8 0.1
Tony Romo 2008 200 5 2.5 0.2 250 9 3.6 -0.7 -0.8
Marc Bulger 2008 200 5 2.5 0.2 240 8 3.3 -0.4 -0.6
Matt Ryan 2008 227 5 2.2 0.5 207 6 2.9 0.0 -0.4
Jake Delhomme 2008 231 5 2.2 0.5 183 7 3.8 -0.9 -1.4
Trent Edwards 2008 222 5 2.3 0.4 152 5 3.3 -0.4 -0.8
Kurt Warner 2008 295 6 2.0 0.6 303 8 2.6 0.3 -0.3
Philip Rivers 2008 234 6 2.6 0.1 244 5 2.0 0.9 0.8
Drew Brees 2008 304 7 2.3 0.4 331 10 3.0 -0.1 -0.4
Matt Cassel 2008 233 7 3.0 -0.3 283 4 1.4 1.5 1.9
Joe Flacco 2008 220 7 3.2 -0.5 208 5 2.4 0.5 1.1
Ben Roethlisberger 2008 198 8 4.0 -1.4 271 7 2.6 0.4 1.7
Matt Schaub 2008 227 8 3.5 -0.9 153 2 1.3 1.6 2.5
Peyton Manning 2008 294 9 3.1 -0.4 261 3 1.1 1.8 2.2
Jay Cutler 2008 300 10 3.3 -0.7 316 8 2.5 0.4 1.1
Brett Favre 2008 263 12 4.6 -1.9 259 10 3.9 -0.9 1.0
Tom Brady 2007 267 2 0.7 2.5 311 6 1.9 1.1 -1.3
Peyton Manning 2007 259 4 1.5 1.7 256 10 3.9 -0.8 -2.5
Donovan McNabb 2007 287 4 1.4 1.8 186 3 1.6 1.5 -0.3
Jon Kitna 2007 253 6 2.4 0.8 308 14 4.5 -1.5 -2.3
Ben Roethlisberger 2007 208 6 2.9 0.3 196 5 2.6 0.5 0.2
Matt Hasselbeck 2007 280 7 2.5 0.7 282 5 1.8 1.3 0.6
Jason Campbell 2007 227 7 3.1 0.1 190 4 2.1 1.0 0.9
Brett Favre 2007 308 8 2.6 0.6 227 7 3.1 0.0 -0.6
Jay Cutler 2007 219 8 3.7 -0.5 248 6 2.4 0.7 1.1
Philip Rivers 2007 231 8 3.5 -0.3 229 7 3.1 0.0 0.3
Vince Young 2007 151 8 5.3 -2.1 231 9 3.9 -0.8 1.3
Marc Bulger 2007 209 8 3.8 -0.6 169 7 4.1 -1.1 -0.4
Eli Manning 2007 249 9 3.6 -0.4 280 11 3.9 -0.9 -0.4
Derek Anderson 2007 257 9 3.5 -0.3 270 10 3.7 -0.6 -0.3
Drew Brees 2007 335 10 3.0 0.2 317 8 2.5 0.6 0.3
Carson Palmer 2007 294 10 3.4 -0.2 281 10 3.6 -0.5 -0.3
Tony Romo 2007 264 10 3.8 -0.6 256 9 3.5 -0.4 0.1
Marc Bulger 2006 290 1 0.3 2.7 298 7 2.3 1.0 -1.7
Peyton Manning 2006 281 3 1.1 1.9 276 6 2.2 1.1 -0.8
Philip Rivers 2006 238 3 1.3 1.7 222 6 2.7 0.6 -1.1
David Carr 2006 223 5 2.2 0.8 219 7 3.2 0.1 -0.6
Jake Delhomme 2006 270 5 1.9 1.1 161 6 3.7 -0.4 -1.6
Matt Leinart 2006 153 5 3.3 -0.3 224 7 3.1 0.2 0.5
Carson Palmer 2006 258 6 2.3 0.7 262 7 2.7 0.6 0.0
J.P. Losman 2006 212 6 2.8 0.2 217 8 3.7 -0.4 -0.5
Brett Favre 2006 310 7 2.3 0.7 303 11 3.6 -0.3 -1.1
Drew Brees 2006 284 7 2.5 0.5 270 4 1.5 1.8 1.3
Steve McNair 2006 223 7 3.1 -0.1 245 5 2.0 1.3 1.4
Alex Smith 2006 229 7 3.1 -0.1 213 9 4.2 -0.9 -0.8
Michael Vick 2006 206 7 3.4 -0.4 182 6 3.3 0.0 0.4
Matt Hasselbeck 2006 176 7 4.0 -1.0 195 8 4.1 -0.8 0.2
Tom Brady 2006 267 8 3.0 0.0 249 4 1.6 1.7 1.7
Chad Pennington 2006 218 8 3.7 -0.7 267 8 3.0 0.3 1.0
Brad Johnson 2006 258 8 3.1 -0.1 181 7 3.9 -0.6 -0.4
Eli Manning 2006 261 9 3.4 -0.5 261 9 3.4 -0.1 0.3
Joey Harrington 2006 178 9 5.1 -2.1 210 6 2.9 0.5 2.5
Jon Kitna 2006 296 10 3.4 -0.4 300 12 4.0 -0.7 -0.3
Rex Grossman 2006 260 10 3.8 -0.9 220 10 4.5 -1.2 -0.4
Ben Roethlisberger 2006 234 14 6.0 -3.0 235 9 3.8 -0.5 2.5
Kerry Collins 2005 283 3 1.1 1.9 282 9 3.2 0.0 -2.0
Trent Green 2005 265 3 1.1 1.9 242 7 2.9 0.3 -1.6
Jake Plummer 2005 242 3 1.2 1.8 214 4 1.9 1.3 -0.5
Mark Brunell 2005 255 3 1.2 1.8 199 7 3.5 -0.4 -2.2
Tom Brady 2005 282 4 1.4 1.6 248 10 4.0 -0.9 -2.4
Eli Manning 2005 258 5 1.9 1.1 299 12 4.0 -0.8 -1.9
Carson Palmer 2005 267 5 1.9 1.1 242 7 2.9 0.3 -0.9
Matt Hasselbeck 2005 256 5 2.0 1.0 193 4 2.1 1.1 0.1
Drew Bledsoe 2005 244 6 2.5 0.5 255 11 4.3 -1.1 -1.7
Steve McNair 2005 250 6 2.4 0.6 226 5 2.2 1.0 0.4
Peyton Manning 2005 241 6 2.5 0.5 212 4 1.9 1.3 0.8
David Carr 2005 192 6 3.1 -0.1 231 5 2.2 1.0 1.1
Michael Vick 2005 157 6 3.8 -0.8 230 7 3.0 0.1 0.9
Drew Brees 2005 242 7 2.9 0.1 258 8 3.1 0.1 0.0
Jake Delhomme 2005 218 8 3.7 -0.7 217 8 3.7 -0.5 0.2
Gus Frerotte 2005 267 9 3.4 -0.4 227 4 1.8 1.4 1.8
Aaron Brooks 2005 252 10 4.0 -1.0 179 7 3.9 -0.7 0.2
Brett Favre 2005 296 14 4.7 -1.7 311 15 4.8 -1.7 0.1
Drew Brees 2004 206 3 1.5 1.3 194 4 2.1 1.5 0.2
Chad Pennington 2004 216 3 1.4 1.4 154 6 3.9 -0.3 -1.7
Aaron Brooks 2004 281 4 1.4 1.3 261 12 4.6 -1.0 -2.3
Peyton Manning 2004 269 4 1.5 1.3 228 6 2.6 1.0 -0.3
Donovan McNabb 2004 270 4 1.5 1.3 199 4 2.0 1.6 0.3
Josh McCown 2004 227 4 1.8 1.0 181 6 3.3 0.3 -0.7
Daunte Culpepper 2004 277 5 1.8 1.0 271 6 2.2 1.4 0.4
Joey Harrington 2004 242 5 2.1 0.7 247 7 2.8 0.8 0.1
David Carr 2004 250 5 2.0 0.8 216 9 4.2 -0.6 -1.3
Trent Green 2004 263 6 2.3 0.5 293 11 3.8 -0.2 -0.6
Kyle Boller 2004 196 6 3.1 -0.3 268 5 1.9 1.7 2.0
Tom Brady 2004 246 7 2.8 -0.1 228 7 3.1 0.5 0.6
Drew Bledsoe 2004 221 7 3.2 -0.4 229 9 3.9 -0.3 0.1
Byron Leftwich 2004 273 7 2.6 0.2 168 3 1.8 1.8 1.6
Marc Bulger 2004 283 8 2.8 -0.1 202 6 3.0 0.6 0.7
Brett Favre 2004 269 9 3.3 -0.6 271 8 3.0 0.6 1.2
Matt Hasselbeck 2004 269 9 3.3 -0.6 205 6 2.9 0.7 1.2
Jake Plummer 2004 278 10 3.6 -0.8 243 10 4.1 -0.5 0.3
Vinny Testaverde 2004 263 10 3.8 -1.0 232 10 4.3 -0.7 0.3
Carson Palmer 2004 279 10 3.6 -0.8 153 8 5.2 -1.6 -0.8
Jake Delhomme 2004 285 11 3.9 -1.1 248 4 1.6 2.0 3.1
Kerry Collins 2004 224 11 4.9 -2.1 289 9 3.1 0.5 2.6
Daunte Culpepper 2003 160 2 1.3 2.0 294 9 3.1 0.1 -1.9
Steve McNair 2003 232 3 1.3 2.0 168 4 2.4 0.8 -1.2
Aaron Brooks 2003 248 5 2.0 1.3 270 3 1.1 2.1 0.8
Peyton Manning 2003 270 6 2.2 1.1 296 4 1.4 1.9 0.8
Jake Delhomme 2003 212 6 2.8 0.4 237 10 4.2 -1.0 -1.5
Tom Brady 2003 254 7 2.8 0.5 273 5 1.8 1.4 0.9
Trent Green 2003 239 7 2.9 0.3 284 5 1.8 1.4 1.1
Matt Hasselbeck 2003 248 7 2.8 0.5 265 8 3.0 0.2 -0.3
Donovan McNabb 2003 246 7 2.8 0.4 232 4 1.7 1.5 1.1
Brad Johnson 2003 296 8 2.7 0.6 274 13 4.7 -1.5 -2.1
Jon Kitna 2003 272 8 2.9 0.3 248 7 2.8 0.4 0.1
Drew Bledsoe 2003 256 8 3.1 0.2 215 4 1.9 1.3 1.2
Jeff Garcia 2003 235 8 3.4 -0.1 157 5 3.2 0.0 0.2
Marc Bulger 2003 259 9 3.5 -0.2 273 13 4.8 -1.6 -1.4
Quincy Carter 2003 234 9 3.8 -0.6 271 12 4.4 -1.2 -0.6
Byron Leftwich 2003 180 9 5.0 -1.7 238 7 2.9 0.3 2.0
Kerry Collins 2003 333 10 3.0 0.3 167 6 3.6 -0.4 -0.7
Brett Favre 2003 252 10 4.0 -0.7 219 11 5.0 -1.8 -1.1
Tommy Maddox 2003 274 11 4.0 -0.7 245 6 2.4 0.8 1.5
Joey Harrington 2003 269 13 4.8 -1.6 285 9 3.2 0.1 1.6
Michael Vick 2002 159 1 0.6 2.5 262 7 2.7 0.3 -2.2
Brett Favre 2002 258 4 1.6 1.6 293 12 4.1 -1.1 -2.7
Jeff Garcia 2002 239 4 1.7 1.5 289 6 2.1 0.9 -0.6
Mark Brunell 2002 234 4 1.7 1.4 182 3 1.6 1.3 -0.1
Chad Pennington 2002 168 4 2.4 0.8 231 2 0.9 2.1 1.3
Rodney Peete 2002 154 4 2.6 0.5 227 10 4.4 -1.5 -2.0
Drew Bledsoe 2002 318 5 1.6 1.6 292 10 3.4 -0.5 -2.0
Brad Johnson 2002 239 5 2.1 1.0 212 1 0.5 2.5 1.4
Tim Couch 2002 224 6 2.7 0.5 219 12 5.5 -2.5 -3.0
Rich Gannon 2002 342 7 2.0 1.1 276 3 1.1 1.9 0.8
Jon Kitna 2002 163 7 4.3 -1.2 310 9 2.9 0.0 1.2
Tommy Maddox 2002 165 7 4.2 -1.1 212 9 4.2 -1.3 -0.2
Kerry Collins 2002 265 8 3.0 0.1 280 6 2.1 0.8 0.7
Drew Brees 2002 239 8 3.3 -0.2 287 8 2.8 0.2 0.4
David Carr 2002 207 8 3.9 -0.7 237 7 3.0 0.0 0.7
Tom Brady 2002 315 9 2.9 0.3 286 5 1.7 1.2 0.9
Jake Plummer 2002 278 9 3.2 -0.1 252 11 4.4 -1.4 -1.3
Aaron Brooks 2002 268 9 3.4 -0.2 260 6 2.3 0.6 0.9
Steve McNair 2002 262 9 3.4 -0.3 230 6 2.6 0.3 0.6
Brian Griese 2002 284 9 3.2 0.0 152 6 3.9 -1.0 -1.0
Joey Harrington 2002 233 9 3.9 -0.7 196 7 3.6 -0.6 0.1
Trent Green 2002 244 10 4.1 -1.0 226 3 1.3 1.6 2.6
Peyton Manning 2002 308 11 3.6 -0.4 283 8 2.8 0.1 0.6
Daunte Culpepper 2002 289 14 4.8 -1.7 260 9 3.5 -0.5 1.2
Rich Gannon 2001 257 1 0.4 2.9 292 8 2.7 0.7 -2.2
Brad Johnson 2001 278 4 1.4 1.8 281 7 2.5 0.9 -0.9
Doug Flutie 2001 225 4 1.8 1.5 296 14 4.7 -1.3 -2.8
Donovan McNabb 2001 257 4 1.6 1.7 236 8 3.4 0.0 -1.7
Mark Brunell 2001 226 4 1.8 1.5 247 9 3.6 -0.2 -1.7
Kordell Stewart 2001 215 4 1.9 1.4 227 7 3.1 0.3 -1.1
Tom Brady 2001 200 4 2.0 1.3 213 8 3.8 -0.3 -1.6
Tim Couch 2001 216 5 2.3 0.9 238 16 6.7 -3.3 -4.3
Vinny Testaverde 2001 194 5 2.6 0.7 247 9 3.6 -0.2 -0.9
Tony Banks 2001 162 5 3.1 0.2 208 5 2.4 1.0 0.8
Aaron Brooks 2001 288 6 2.1 1.2 270 16 5.9 -2.5 -3.7
Jim Miller 2001 169 6 3.6 -0.3 226 4 1.8 1.6 1.9
Jeff Garcia 2001 264 7 2.7 0.6 240 5 2.1 1.3 0.7
Steve McNair 2001 214 7 3.3 0.0 217 5 2.3 1.1 1.1
Jon Kitna 2001 282 8 2.8 0.4 299 14 4.7 -1.3 -1.7
Elvis Grbac 2001 195 8 4.1 -0.8 272 10 3.7 -0.3 0.6
Jake Plummer 2001 265 9 3.4 -0.1 260 5 1.9 1.5 1.6
Kerry Collins 2001 266 10 3.8 -0.5 302 6 2.0 1.4 1.9
Brett Favre 2001 261 10 3.8 -0.6 249 5 2.0 1.4 2.0
Peyton Manning 2001 260 11 4.2 -1.0 287 12 4.2 -0.8 0.2
Kurt Warner 2001 285 11 3.9 -0.6 261 11 4.2 -0.8 -0.2
Chris Weinke 2001 269 12 4.5 -1.2 271 7 2.6 0.8 2.0
Brian Griese 2001 252 12 4.8 -1.5 199 7 3.5 -0.1 1.4
Jay Fiedler 2001 219 12 5.5 -2.2 231 7 3.0 0.4 2.6
Trent Green 2001 282 13 4.6 -1.3 241 11 4.6 -1.1 0.2
Jeff Garcia 2000 271 5 1.8 1.4 290 5 1.7 1.5 0.1
Rich Gannon 2000 234 5 2.1 1.1 239 6 2.5 0.7 -0.4
Steve McNair 2000 178 5 2.8 0.4 218 8 3.7 -0.4 -0.8
Drew Bledsoe 2000 269 6 2.2 1.0 262 7 2.7 0.6 -0.4
Shaun King 2000 229 6 2.6 0.6 199 7 3.5 -0.3 -0.9
Donovan McNabb 2000 271 7 2.6 0.6 298 6 2.0 1.2 0.6
Elvis Grbac 2000 256 7 2.7 0.5 291 7 2.4 0.8 0.4
Kerry Collins 2000 254 7 2.8 0.5 275 6 2.2 1.1 0.6
Steve Beuerlein 2000 276 8 2.9 0.3 257 10 3.9 -0.6 -1.0
Brett Favre 2000 285 9 3.2 0.1 295 7 2.4 0.9 0.8
Peyton Manning 2000 280 9 3.2 0.0 291 6 2.1 1.2 1.2
Daunte Culpepper 2000 244 9 3.7 -0.5 230 7 3.0 0.2 0.7
Charlie Batch 2000 244 9 3.7 -0.5 168 6 3.6 -0.3 0.1
Jay Fiedler 2000 200 9 4.5 -1.3 157 5 3.2 0.1 1.3
Mark Brunell 2000 281 10 3.6 -0.3 231 4 1.7 1.5 1.8
Vinny Testaverde 2000 276 11 4.0 -0.8 314 14 4.5 -1.2 -0.5
Jake Plummer 2000 295 14 4.7 -1.5 180 7 3.9 -0.6 0.9
Brad Johnson 1999 259 3 1.2 2.2 260 10 3.8 -0.5 -2.7
Drew Bledsoe 1999 271 4 1.5 1.9 268 17 6.3 -3.0 -4.9
Jon Kitna 1999 191 4 2.1 1.3 304 12 3.9 -0.6 -1.9
Mark Brunell 1999 229 4 1.7 1.6 212 5 2.4 1.0 -0.7
Kurt Warner 1999 250 5 2.0 1.4 249 8 3.2 0.1 -1.3
Tim Couch 1999 196 5 2.6 0.8 203 8 3.9 -0.6 -1.4
Rich Gannon 1999 270 6 2.2 1.2 245 8 3.3 0.1 -1.1
Jeff Garcia 1999 178 6 3.4 0.0 197 5 2.5 0.8 0.8
Elvis Grbac 1999 219 7 3.2 0.2 280 8 2.9 0.5 0.3
Brian Griese 1999 254 8 3.1 0.2 198 6 3.0 0.3 0.1
Troy Aikman 1999 244 8 3.3 0.1 198 4 2.0 1.3 1.2
Jim Harbaugh 1999 164 8 4.9 -1.5 270 6 2.2 1.1 2.6
Peyton Manning 1999 276 9 3.3 0.1 257 6 2.3 1.0 0.9
Steve Beuerlein 1999 280 10 3.6 -0.2 291 5 1.7 1.6 1.8
Doug Flutie 1999 281 11 3.9 -0.5 197 5 2.5 0.8 1.3
Brett Favre 1999 287 13 4.5 -1.1 308 10 3.2 0.1 1.2
Jake Plummer 1999 181 14 7.7 -4.3 200 10 5.0 -1.7 2.7
Vinny Testaverde 1998 166 2 1.2 2.0 255 5 2.0 1.4 -0.6
Randall Cunningham 1998 187 3 1.6 1.6 238 7 2.9 0.4 -1.2
John Elway 1998 154 3 1.9 1.3 202 7 3.5 -0.1 -1.4
Trent Dilfer 1998 226 5 2.2 1.0 203 10 4.9 -1.6 -2.6
Steve Young 1998 309 6 1.9 1.3 208 6 2.9 0.5 -0.8
Steve McNair 1998 223 6 2.7 0.5 269 4 1.5 1.9 1.3
Drew Bledsoe 1998 250 6 2.4 0.8 231 8 3.5 -0.1 -0.9
Dan Marino 1998 247 7 2.8 0.4 290 8 2.8 0.6 0.2
Trent Green 1998 216 7 3.2 0.0 293 4 1.4 2.0 2.0
Kordell Stewart 1998 220 10 4.5 -1.3 238 8 3.4 0.0 1.3
Jake Plummer 1998 245 12 4.9 -1.7 302 8 2.6 0.7 2.4
Brett Favre 1998 280 15 5.4 -2.1 271 8 3.0 0.4 2.5
Peyton Manning 1998 292 16 5.5 -2.3 283 12 4.2 -0.9 1.4
Dan Marino 1997 288 3 1.0 1.9 260 8 3.1 0.0 -1.9
Troy Aikman 1997 254 3 1.2 1.8 264 9 3.4 -0.3 -2.1
Neil O'Donnell 1997 258 3 1.2 1.8 202 4 2.0 1.1 -0.6
Mark Brunell 1997 180 3 1.7 1.3 255 4 1.6 1.6 0.3
Steve Young 1997 164 3 1.8 1.1 192 3 1.6 1.6 0.4
Jeff George 1997 249 4 1.6 1.3 272 5 1.8 1.3 -0.1
Erik Kramer 1997 217 5 2.3 0.6 260 9 3.5 -0.3 -1.0
Scott Mitchell 1997 265 6 2.3 0.7 244 8 3.3 -0.2 -0.8
Trent Dilfer 1997 207 6 2.9 0.1 179 5 2.8 0.3 0.3
John Elway 1997 244 7 2.9 0.1 258 4 1.6 1.6 1.5
Tony Banks 1997 247 7 2.8 0.1 240 6 2.5 0.6 0.5
Steve McNair 1997 211 7 3.3 -0.4 204 6 2.9 0.2 0.5
Brett Favre 1997 274 8 2.9 0.0 239 8 3.3 -0.2 -0.3
Brad Johnson 1997 286 8 2.8 0.2 166 4 2.4 0.7 0.6
Drew Bledsoe 1997 277 9 3.2 -0.3 245 6 2.4 0.7 1.0
Vinny Testaverde 1997 306 9 2.9 0.0 164 6 3.7 -0.5 -0.5
Kordell Stewart 1997 199 9 4.5 -1.6 241 8 3.3 -0.2 1.4
Todd Collins 1997 204 9 4.4 -1.5 187 4 2.1 1.0 2.4
Warren Moon 1997 272 10 3.7 -0.7 256 6 2.3 0.8 1.5
Gus Frerotte 1997 246 10 4.1 -1.1 156 2 1.3 1.8 3.0
Kerry Collins 1997 163 10 6.1 -3.2 218 11 5.0 -1.9 1.3
Jim Harbaugh 1996 231 2 0.9 2.4 174 9 5.2 -1.7 -4.1
Stan Humphries 1996 216 3 1.4 1.9 200 10 5.0 -1.5 -3.4
Drew Bledsoe 1996 319 4 1.3 2.0 304 11 3.6 -0.1 -2.2
Gus Frerotte 1996 204 5 2.5 0.9 266 6 2.3 1.2 0.4
Troy Aikman 1996 248 5 2.0 1.3 217 8 3.7 -0.2 -1.5
Jeff Blake 1996 272 6 2.2 1.1 277 8 2.9 0.6 -0.5
Brett Favre 1996 271 6 2.2 1.1 272 7 2.6 0.9 -0.2
Mike Tomczak 1996 193 6 3.1 0.2 208 11 5.3 -1.8 -2.0
Kerry Collins 1996 171 6 3.5 -0.2 193 3 1.6 1.9 2.1
Steve Bono 1996 276 7 2.5 0.8 162 6 3.7 -0.2 -1.0
Tony Banks 1996 153 7 4.6 -1.3 215 8 3.7 -0.2 1.0
Jim Everett 1996 263 8 3.0 0.3 201 8 4.0 -0.5 -0.8
Vinny Testaverde 1996 288 9 3.1 0.2 261 10 3.8 -0.4 -0.5
Dave M. Brown 1996 214 9 4.2 -0.9 184 11 6.0 -2.5 -1.6
Bobby Hebert 1996 216 10 4.6 -1.3 272 15 5.5 -2.0 -0.7
John Elway 1996 273 10 3.7 -0.4 193 4 2.1 1.4 1.8
Jeff Hostetler 1996 205 10 4.9 -1.6 197 4 2.0 1.4 3.0
Scott Mitchell 1996 266 11 4.1 -0.8 171 6 3.5 0.0 0.8
Trent Dilfer 1996 245 12 4.9 -1.6 237 7 3.0 0.5 2.1
Jim Kelly 1996 199 12 6.0 -2.7 180 7 3.9 -0.4 2.3
Mark Brunell 1996 282 13 4.6 -1.3 275 7 2.5 0.9 2.2
Troy Aikman 1995 208 2 1.0 2.0 224 5 2.2 1.0 -1.0
Vinny Testaverde 1995 213 3 1.4 1.5 179 7 3.9 -0.7 -2.2
Erik Kramer 1995 265 4 1.5 1.4 257 6 2.3 0.9 -0.6
Steve Bono 1995 273 4 1.5 1.4 247 6 2.4 0.8 -0.7
Dan Marino 1995 186 4 2.2 0.8 296 11 3.7 -0.5 -1.3
Steve Young 1995 233 4 1.7 1.2 214 7 3.3 -0.1 -1.3
Scott Mitchell 1995 299 5 1.7 1.2 284 7 2.5 0.7 -0.5
John Elway 1995 303 5 1.7 1.3 239 9 3.8 -0.6 -1.8
Jim Kelly 1995 239 5 2.1 0.8 219 8 3.7 -0.5 -1.3
Drew Bledsoe 1995 336 6 1.8 1.1 300 10 3.3 -0.1 -1.3
Jeff Blake 1995 273 6 2.2 0.7 294 11 3.7 -0.6 -1.3
Dave M. Brown 1995 231 6 2.6 0.3 225 4 1.8 1.4 1.1
Kerry Collins 1995 165 6 3.6 -0.7 268 13 4.9 -1.7 -0.9
Jeff George 1995 287 7 2.4 0.5 270 4 1.5 1.7 1.2
Trent Dilfer 1995 203 7 3.4 -0.5 212 11 5.2 -2.0 -1.5
Stan Humphries 1995 237 8 3.4 -0.5 241 6 2.5 0.7 1.2
Boomer Esiason 1995 184 8 4.3 -1.4 205 7 3.4 -0.2 1.2
Warren Moon 1995 302 9 3.0 -0.1 304 5 1.6 1.5 1.6
Brett Favre 1995 304 9 3.0 0.0 266 4 1.5 1.7 1.7
Jim Everett 1995 283 9 3.2 -0.3 284 5 1.8 1.4 1.7
Dave Krieg 1995 245 13 5.3 -2.4 276 8 2.9 0.3 2.7
Rick Mirer 1995 215 14 6.5 -3.6 176 6 3.4 -0.2 3.4
Craig Erickson 1994 183 3 1.6 1.8 216 7 3.2 -0.4 -2.2
Stan Humphries 1994 210 5 2.4 1.1 243 7 2.9 -0.1 -1.1
Troy Aikman 1994 207 5 2.4 1.0 154 7 4.5 -1.7 -2.7
John Elway 1994 297 6 2.0 1.4 197 4 2.0 0.8 -0.6
Brett Favre 1994 284 7 2.5 1.0 298 7 2.3 0.5 -0.5
Jim Everett 1994 287 7 2.4 1.0 253 11 4.3 -1.5 -2.5
Joe Montana 1994 300 7 2.3 1.1 193 2 1.0 1.8 0.7
Steve Young 1994 234 7 3.0 0.4 227 3 1.3 1.5 1.1
Dan Marino 1994 303 8 2.6 0.8 312 9 2.9 -0.1 -0.9
Randall Cunningham 1994 275 8 2.9 0.5 215 5 2.3 0.5 0.0
Boomer Esiason 1994 199 8 4.0 -0.6 241 5 2.1 0.8 1.3
Jeff George 1994 271 9 3.3 0.1 253 9 3.6 -0.7 -0.8
Jeff Hostetler 1994 236 9 3.8 -0.4 219 7 3.2 -0.4 0.0
Jim Kelly 1994 241 9 3.7 -0.3 207 8 3.9 -1.0 -0.7
Warren Moon 1994 302 11 3.6 -0.2 299 8 2.7 0.2 0.4
Vinny Testaverde 1994 209 11 5.3 -1.8 167 7 4.2 -1.4 0.5
Drew Bledsoe 1994 358 14 3.9 -0.5 333 13 3.9 -1.1 -0.6
Troy Aikman 1993 219 2 0.9 2.6 173 4 2.3 0.7 -1.9
Neil O'Donnell 1993 222 4 1.8 1.7 264 3 1.1 1.9 0.2
John Elway 1993 286 5 1.7 1.7 265 5 1.9 1.1 -0.6
Bobby Hebert 1993 165 5 3.0 0.4 265 12 4.5 -1.5 -2.0
Boomer Esiason 1993 241 6 2.5 1.0 232 5 2.2 0.8 -0.1
Drew Bledsoe 1993 182 6 3.3 0.2 247 9 3.6 -0.6 -0.8
Jeff Hostetler 1993 176 6 3.4 0.1 243 4 1.6 1.4 1.3
Phil Simms 1993 205 7 3.4 0.0 195 2 1.0 2.0 1.9
Wade Wilson 1993 233 7 3.0 0.5 155 8 5.2 -2.2 -2.6
Jim Kelly 1993 243 8 3.3 0.2 227 10 4.4 -1.4 -1.6
Steve Young 1993 245 9 3.7 -0.2 217 7 3.2 -0.2 0.0
Steve Beuerlein 1993 215 9 4.2 -0.7 203 8 3.9 -0.9 -0.2
Rick Mirer 1993 247 11 4.5 -1.0 239 6 2.5 0.5 1.5
Craig Erickson 1993 203 12 5.9 -2.4 254 9 3.5 -0.5 1.9
Brett Favre 1993 246 13 5.3 -1.8 276 11 4.0 -1.0 0.8
Warren Moon 1993 281 14 5.0 -1.5 239 7 2.9 0.1 1.6
Brett Favre 1992 222 3 1.4 2.2 249 10 4.0 0.1 -2.1
Steve Young 1992 184 4 2.2 1.4 218 3 1.4 2.8 1.4
Chris Chandler 1992 221 5 2.3 1.3 192 10 5.2 -1.1 -2.4
Browning Nagle 1992 186 6 3.2 0.4 201 11 5.5 -1.3 -1.7
Randall Cunningham 1992 186 6 3.2 0.4 198 5 2.5 1.6 1.3
Dave Krieg 1992 207 7 3.4 0.2 206 5 2.4 1.7 1.5
Vinny Testaverde 1992 178 8 4.5 -0.9 180 8 4.4 -0.3 0.6
Dan Marino 1992 285 9 3.2 0.4 269 7 2.6 1.5 1.1
Mark Rypien 1992 231 9 3.9 -0.3 248 8 3.2 0.9 1.2
Jim Everett 1992 212 9 4.2 -0.7 263 9 3.4 0.7 1.4
Troy Aikman 1992 235 10 4.3 -0.7 238 4 1.7 2.5 3.1
Jim Kelly 1992 254 10 3.9 -0.4 208 9 4.3 -0.2 0.2
Bobby Hebert 1992 210 10 4.8 -1.2 212 6 2.8 1.3 2.5
Stan Humphries 1992 227 12 5.3 -1.7 227 6 2.6 1.5 3.2
Bernie Kosar 1991 240 0 0.0 3.6 254 9 3.5 -0.1 -3.7
John Elway 1991 235 2 0.9 2.7 216 10 4.6 -1.2 -4.0
Steve DeBerg 1991 203 4 2.0 1.6 231 10 4.3 -0.9 -2.6
Ken O'Brien 1991 227 5 2.2 1.4 262 6 2.3 1.1 -0.3
Jim McMahon 1991 153 5 3.3 0.3 158 6 3.8 -0.4 -0.7
Dan Marino 1991 260 6 2.3 1.3 289 7 2.4 1.0 -0.3
Jeff George 1991 259 6 2.3 1.3 226 6 2.7 0.7 -0.5
Mark Rypien 1991 192 7 3.6 0.0 229 4 1.7 1.7 1.7
Jim Everett 1991 219 8 3.7 -0.1 271 12 4.4 -1.0 -1.0
John Friesz 1991 259 8 3.1 0.5 228 7 3.1 0.3 -0.2
Jim Kelly 1991 238 8 3.4 0.2 236 9 3.8 -0.4 -0.6
Jim Harbaugh 1991 238 9 3.8 -0.2 240 7 2.9 0.5 0.7
Boomer Esiason 1991 201 10 5.0 -1.4 212 6 2.8 0.6 1.9
Warren Moon 1991 305 11 3.6 0.0 350 10 2.9 0.5 0.6
Chris Miller 1991 238 13 5.5 -1.9 175 5 2.9 0.5 2.4
Steve DeBerg 1990 232 3 1.3 2.2 212 1 0.5 3.1 0.8
Ken O'Brien 1990 224 4 1.8 1.7 187 6 3.2 0.3 -1.4
Vinny Testaverde 1990 162 4 2.5 1.1 203 14 6.9 -3.4 -4.4
Jay Schroeder 1990 175 4 2.3 1.2 159 5 3.1 0.4 -0.9
John Elway 1990 238 6 2.5 1.0 264 8 3.0 0.5 -0.5
Jim Everett 1990 276 7 2.5 1.0 278 10 3.6 -0.1 -1.1
Dan Marino 1990 245 7 2.9 0.7 286 4 1.4 2.1 1.5
Randall Cunningham 1990 259 7 2.7 0.8 206 6 2.9 0.6 -0.2
Timm Rosenbach 1990 209 8 3.8 -0.3 228 9 3.9 -0.4 -0.1
Bernie Kosar 1990 249 8 3.2 0.3 174 7 4.0 -0.5 -0.8
Billy Joe Tolliver 1990 194 8 4.1 -0.6 216 8 3.7 -0.2 0.4
Bubby Brister 1990 189 8 4.2 -0.7 198 6 3.0 0.5 1.2
Warren Moon 1990 331 9 2.7 0.8 253 4 1.6 1.9 1.1
Joe Montana 1990 306 9 2.9 0.6 214 7 3.3 0.3 -0.3
Dave Krieg 1990 230 11 4.8 -1.3 218 9 4.1 -0.6 0.7
Troy Aikman 1990 235 11 4.7 -1.1 164 7 4.3 -0.7 0.4
Rich Gannon 1990 162 11 6.8 -3.3 187 5 2.7 0.9 4.1
Boomer Esiason 1990 231 13 5.6 -2.1 171 9 5.3 -1.7 0.4
Chris Miller 1989 236 2 0.8 3.2 290 8 2.8 0.9 -2.3
Joe Montana 1989 185 3 1.6 2.5 201 5 2.5 1.2 -1.3
Bubby Brister 1989 164 3 1.8 2.3 178 7 3.9 -0.2 -2.5
Boomer Esiason 1989 212 4 1.9 2.2 243 7 2.9 0.8 -1.4
Warren Moon 1989 229 7 3.1 1.0 235 7 3.0 0.7 -0.3
Jim Kelly 1989 162 7 4.3 -0.2 229 11 4.8 -1.1 -0.9
Phil Simms 1989 190 8 4.2 -0.1 215 6 2.8 0.9 1.0
Randall Cunningham 1989 266 9 3.4 0.7 266 6 2.3 1.4 0.7
Jim Everett 1989 257 9 3.5 0.6 261 8 3.1 0.6 0.0
Bernie Kosar 1989 240 9 3.8 0.3 273 5 1.8 1.9 1.5
Mark Rypien 1989 281 9 3.2 0.9 195 4 2.1 1.6 0.7
Vinny Testaverde 1989 224 10 4.5 -0.4 256 12 4.7 -1.0 -0.6
Ken O'Brien 1989 275 10 3.6 0.5 202 8 4.0 -0.3 -0.7
Steve DeBerg 1989 156 10 6.4 -2.3 168 6 3.6 0.1 2.4
Dave Krieg 1989 279 11 3.9 0.1 220 9 4.1 -0.4 -0.6
John Elway 1989 232 11 4.7 -0.6 184 7 3.8 -0.1 0.5
Don Majkowski 1989 280 12 4.3 -0.2 319 8 2.5 1.2 1.4
Dan Marino 1989 289 16 5.5 -1.4 261 6 2.3 1.4 2.8
Doug Williams 1988 167 4 2.4 1.8 213 8 3.8 -0.2 -1.9
Jim Everett 1988 220 5 2.3 1.9 297 13 4.4 -0.8 -2.7
Randall Cunningham 1988 265 6 2.3 1.9 295 10 3.4 0.2 -1.7
Bobby Hebert 1988 237 6 2.5 1.7 241 9 3.7 -0.1 -1.8
Neil Lomax 1988 227 7 3.1 1.1 216 4 1.9 1.8 0.6
Phil Simms 1988 277 8 2.9 1.3 202 3 1.5 2.1 0.8
Joe Montana 1988 232 8 3.4 0.7 165 2 1.2 2.4 1.6
Jim Kelly 1988 241 9 3.7 0.5 211 8 3.8 -0.2 -0.6
Steve DeBerg 1988 205 10 4.9 -0.7 209 6 2.9 0.7 1.4
Boomer Esiason 1988 204 10 4.9 -0.7 184 4 2.2 1.4 2.1
John Elway 1988 227 11 4.8 -0.7 269 8 3.0 0.6 1.3
Steve Pelluer 1988 237 11 4.6 -0.5 198 8 4.0 -0.4 0.0
Dan Marino 1988 302 12 4.0 0.2 304 11 3.6 0.0 -0.2
Vinny Testaverde 1988 279 22 7.9 -3.7 187 13 7.0 -3.3 0.3
Tony Eason 1986 175 0 0.0 3.8 273 10 3.7 0.5 -3.3
Bernie Kosar 1986 245 4 1.6 2.2 286 6 2.1 2.1 -0.1
John Elway 1986 237 4 1.7 2.1 267 9 3.4 0.8 -1.3
Jay Schroeder 1986 238 6 2.5 1.3 303 16 5.3 -1.1 -2.4
Ken O'Brien 1986 213 6 2.8 1.0 269 14 5.2 -1.0 -2.0
Neil Lomax 1986 254 7 2.8 1.1 167 5 3.0 1.2 0.1
Dave Wilson 1986 156 7 4.5 -0.7 186 10 5.4 -1.2 -0.5
Jack Trudeau 1986 233 8 3.4 0.4 184 10 5.4 -1.3 -1.6
Dave Krieg 1986 219 8 3.7 0.2 156 3 1.9 2.2 2.1
Boomer Esiason 1986 233 9 3.9 0.0 236 8 3.4 0.8 0.8
Mark Malone 1986 178 10 5.6 -1.8 247 8 3.2 0.9 2.7
Phil Simms 1986 261 11 4.2 -0.4 207 11 5.3 -1.1 -0.7
Jim Kelly 1986 223 12 5.4 -1.6 257 5 1.9 2.2 3.8
Dan Marino 1986 299 13 4.3 -0.5 324 10 3.1 1.1 1.6
Randy Wright 1986 267 13 4.9 -1.1 225 10 4.4 -0.3 0.8
Warren Moon 1986 271 17 6.3 -2.5 217 9 4.1 0.0 2.5
Dan Fouts 1986 263 19 7.2 -3.4 167 3 1.8 2.4 5.8
Ron Jaworski 1985 178 3 1.7 2.6 306 17 5.6 -1.5 -4.1
Ken O'Brien 1985 222 5 2.3 2.1 266 3 1.1 2.9 0.8
Marc Wilson 1985 152 5 3.3 1.0 236 16 6.8 -2.8 -3.8
Joe Montana 1985 257 6 2.3 2.0 237 7 3.0 1.1 -0.9
Boomer Esiason 1985 214 7 3.3 1.0 217 5 2.3 1.7 0.7
Dan Fouts 1985 183 7 3.8 0.5 247 13 5.3 -1.2 -1.7
Eric Hipple 1985 189 7 3.7 0.6 217 11 5.1 -1.0 -1.6
Mike Pagel 1985 203 8 3.9 0.4 190 7 3.7 0.3 0.0
John Elway 1985 290 9 3.1 1.2 315 14 4.4 -0.4 -1.6
Neil Lomax 1985 254 9 3.5 0.8 217 3 1.4 2.6 1.9
Warren Moon 1985 190 9 4.7 -0.4 187 10 5.3 -1.3 -0.9
Dan Marino 1985 301 10 3.3 1.0 266 11 4.1 -0.1 -1.1
Phil Simms 1985 263 10 3.8 0.5 232 10 4.3 -0.3 -0.8
Dieter Brock 1985 213 10 4.7 -0.4 152 3 2.0 2.1 2.4
Dave Krieg 1985 269 12 4.5 -0.2 263 8 3.0 1.0 1.1
Danny White 1985 264 12 4.5 -0.2 186 5 2.7 1.3 1.6
Tommy Kramer 1985 283 13 4.6 -0.3 223 13 5.8 -1.8 -1.5
Tony Eason 1984 167 1 0.6 3.2 264 7 2.7 1.6 -1.6
Joe Montana 1984 200 3 1.5 2.3 232 7 3.0 1.2 -1.1
Dan Marino 1984 250 5 2.0 1.8 314 12 3.8 0.4 -1.4
Neil Lomax 1984 264 5 1.9 1.9 296 11 3.7 0.5 -1.4
Gary Danielson 1984 216 5 2.3 1.5 194 10 5.2 -0.9 -2.4
Phil Simms 1984 278 6 2.2 1.7 255 12 4.7 -0.5 -2.1
Joe Theismann 1984 214 6 2.8 1.0 263 7 2.7 1.6 0.6
Warren Moon 1984 233 6 2.6 1.3 217 8 3.7 0.6 -0.7
Steve DeBerg 1984 205 7 3.4 0.4 304 11 3.6 0.6 0.2
John Elway 1984 171 7 4.1 -0.3 209 8 3.8 0.4 0.7
Dan Fouts 1984 317 11 3.5 0.4 190 6 3.2 1.1 0.7
Dave Krieg 1984 220 12 5.5 -1.6 260 12 4.6 -0.4 1.3
Lynn Dickey 1984 212 12 5.7 -1.8 189 7 3.7 0.5 2.4
Paul McDonald 1984 291 14 4.8 -1.0 202 9 4.5 -0.2 0.8
Joe Theismann 1983 233 3 1.3 3.1 226 8 3.5 0.8 -2.3
Steve Bartkowski 1983 244 3 1.2 3.2 188 2 1.1 3.3 0.1
Joe Montana 1983 254 5 2.0 2.4 261 7 2.7 1.7 -0.8
Ron Jaworski 1983 205 5 2.4 2.0 241 13 5.4 -1.0 -3.0
Steve Dils 1983 200 7 3.5 0.9 244 9 3.7 0.7 -0.2
Cliff Stoudt 1983 186 8 4.3 0.1 195 13 6.7 -2.3 -2.4
Bill Kenney 1983 286 9 3.1 1.3 317 9 2.8 1.5 0.2
Joe Ferguson 1983 261 9 3.4 1.0 247 16 6.5 -2.1 -3.1
Eric Hipple 1983 212 10 4.7 -0.3 175 8 4.6 -0.2 0.1
Scott Brunner 1983 217 11 5.1 -0.7 169 11 6.5 -2.2 -1.5
Lynn Dickey 1983 239 12 5.0 -0.6 245 17 6.9 -2.6 -2.0
Danny White 1983 268 13 4.9 -0.4 265 10 3.8 0.6 1.0
Jim Plunkett 1983 175 13 7.4 -3.0 204 5 2.5 1.9 4.9
Richard Todd 1983 265 14 5.3 -0.9 253 12 4.7 -0.4 0.5
Vince Ferragamo 1983 242 14 5.8 -1.4 222 9 4.1 0.3 1.7
Jack Thompson 1983 204 14 6.9 -2.5 219 7 3.2 1.2 3.6
Brian Sipe 1983 295 17 5.8 -1.4 201 6 3.0 1.4 2.7
Ken Anderson 1981 242 4 1.7 2.3 237 6 2.5 2.0 -0.3
Joe Montana 1981 237 5 2.1 1.9 251 7 2.8 1.8 -0.1
Tommy Kramer 1981 258 7 2.7 1.3 335 17 5.1 -0.5 -1.8
Doug Williams 1981 262 7 2.7 1.3 209 7 3.3 1.2 -0.1
Danny White 1981 223 7 3.1 0.9 168 6 3.6 1.0 0.1
Craig Morton 1981 193 7 3.6 0.4 183 7 3.8 0.7 0.4
David Woodley 1981 180 7 3.9 0.1 186 6 3.2 1.3 1.2
Joe Ferguson 1981 261 8 3.1 0.9 237 12 5.1 -0.5 -1.4
Richard Todd 1981 245 8 3.3 0.7 252 5 2.0 2.6 1.8
Ron Jaworski 1981 222 8 3.6 0.4 239 12 5.0 -0.5 -0.9
Vince Evans 1981 249 8 3.2 0.8 187 12 6.4 -1.9 -2.7
Dan Fouts 1981 305 9 3.0 1.0 304 8 2.6 1.9 0.9
Brian Sipe 1981 316 9 2.8 1.2 251 16 6.4 -1.8 -3.0
Bert Jones 1981 217 10 4.6 -0.6 209 10 4.8 -0.2 0.4
Joe Theismann 1981 260 11 4.2 -0.2 236 9 3.8 0.7 1.0
Steve Bartkowski 1981 271 12 4.4 -0.4 262 11 4.2 0.4 0.8
Gary Danielson 1980 187 4 2.1 2.6 230 7 3.0 1.4 -1.2
Brian Sipe 1980 277 5 1.8 2.9 277 9 3.2 1.2 -1.8
Steve Bartkowski 1980 214 7 3.3 1.5 249 9 3.6 0.8 -0.7
Ron Jaworski 1980 206 7 3.4 1.4 245 5 2.0 2.4 1.0
Doug Williams 1980 228 8 3.5 1.2 293 8 2.7 1.7 0.4
Jim Zorn 1980 226 8 3.5 1.2 262 12 4.6 -0.2 -1.4
Joe Theismann 1980 242 8 3.3 1.4 212 8 3.8 0.6 -0.8
Danny White 1980 232 8 3.4 1.3 204 17 8.3 -3.9 -5.2
Vince Ferragamo 1980 184 8 4.3 0.4 220 11 5.0 -0.6 -1.0
Bert Jones 1980 241 9 3.7 1.0 205 12 5.9 -1.4 -2.5
Jim Hart 1980 242 9 3.7 1.0 183 11 6.0 -1.6 -2.6
Terry Bradshaw 1980 193 9 4.7 0.1 231 13 5.6 -1.2 -1.3
Archie Manning 1980 237 11 4.6 0.1 272 9 3.3 1.1 1.0
Joe Ferguson 1980 248 11 4.4 0.3 191 7 3.7 0.7 0.4
Phil Simms 1980 244 13 5.3 -0.6 158 6 3.8 0.6 1.2
Lynn Dickey 1980 234 14 6.0 -1.2 244 11 4.5 -0.1 1.1
Richard Todd 1980 258 15 5.8 -1.1 221 15 6.8 -2.4 -1.3
Dan Fouts 1980 291 16 5.5 -0.7 298 8 2.7 1.7 2.5
Ken Stabler 1980 239 16 6.7 -1.9 218 12 5.5 -1.1 0.8
Tommy Kramer 1980 260 17 6.5 -1.8 262 6 2.3 2.1 3.9
Joe Ferguson 1979 197 4 2.0 2.9 261 11 4.2 0.1 -2.8
Roger Staubach 1979 229 5 2.2 2.7 232 6 2.6 1.7 -1.0
Ken Anderson 1979 158 5 3.2 1.8 181 5 2.8 1.5 -0.2
Joe Theismann 1979 182 8 4.4 0.5 213 5 2.3 1.9 1.4
Ron Jaworski 1979 176 8 4.5 0.4 198 4 2.0 2.3 1.9
Dan Fouts 1979 253 9 3.6 1.4 277 15 5.4 -1.1 -2.5
Greg Landry 1979 218 9 4.1 0.8 239 6 2.5 1.8 1.0
Steve Grogan 1979 201 10 5.0 -0.1 222 10 4.5 -0.2 -0.2
Steve Bartkowski 1979 225 10 4.4 0.5 155 10 6.5 -2.2 -2.6
Jeff Komlo 1979 190 10 5.3 -0.3 178 13 7.3 -3.0 -2.7
Ken Stabler 1979 225 11 4.9 0.0 273 11 4.0 0.2 0.2
Terry Bradshaw 1979 226 11 4.9 0.1 246 14 5.7 -1.4 -1.5
Doug Williams 1979 196 11 5.6 -0.7 201 13 6.5 -2.2 -1.5
Jim Hart 1979 224 11 4.9 0.0 154 9 5.8 -1.6 -1.6
Steve DeBerg 1979 285 12 4.2 0.7 293 9 3.1 1.2 0.5
Jim Zorn 1979 252 12 4.8 0.2 253 6 2.4 1.9 1.7
Tommy Kramer 1979 252 13 5.2 -0.2 314 11 3.5 0.8 1.0
Dan Pastorini 1979 173 13 7.5 -2.6 151 5 3.3 1.0 3.6
Archie Manning 1979 228 15 6.6 -1.7 192 5 2.6 1.7 3.3
Brian Sipe 1979 266 16 6.0 -1.1 269 10 3.7 0.6 1.7
Ron Jaworski 1978 192 4 2.1 3.0 206 12 5.8 -0.2 -3.3
Joe Ferguson 1978 164 5 3.0 2.1 166 10 6.0 -0.4 -2.5
Jim Hart 1978 224 6 2.7 2.4 253 12 4.7 0.9 -1.6
Terry Bradshaw 1978 200 6 3.0 2.1 168 14 8.3 -2.7 -4.8
Dan Pastorini 1978 179 6 3.4 1.8 189 11 5.8 -0.2 -2.0
Archie Manning 1978 224 7 3.1 2.0 247 9 3.6 2.0 0.0
Brian Sipe 1978 214 10 4.7 0.4 185 5 2.7 2.9 2.5
Joe Theismann 1978 198 10 5.1 0.1 192 8 4.2 1.4 1.4
Pat Haden 1978 220 11 5.0 0.1 224 8 3.6 2.0 1.9
Roger Staubach 1978 246 11 4.5 0.6 167 5 3.0 2.6 2.0
Fran Tarkenton 1978 269 12 4.5 0.7 303 20 6.6 -1.0 -1.7
Jim Zorn 1978 215 12 5.6 -0.5 228 8 3.5 2.1 2.6
Dan Fouts 1978 192 12 6.3 -1.1 189 8 4.2 1.4 2.5
Steve Grogan 1978 210 15 7.1 -2.0 152 8 5.3 0.3 2.4
Ken Stabler 1978 216 19 8.8 -3.7 190 11 5.8 -0.2 3.5

Related posts:

    Similarity scores for 2009 teams, part II
    Similarity Scores for the 2009 teams, part III
    Best and worst divisions
    Greatest Super Bowl Teams II
    Random trivia on past Super Bowl Winners
ShareThis

This entry was posted on Monday, March 1st, 2010 at 8:08 am and is filed under Statgeekery. You can follow any responses to this entry through the RSS 2.0 feed. You can skip to the end and leave a response. Pinging is currently not allowed.

23 Responses to “More on interception rates”

  1. Joseph Says:
    March 1st, 2010 at 11:56 am

    Would have been nice to give us the top 20 or so, sorted by rate, change, etc.
    I'd also be curious to know WHY the correlations are low/non-existent. My theories are:
    1. INT's are dependent upon the defender's hands. IMO, the reason some CB's aren't WR's or Reggie Bush-type RB's is their LACK of good hands.
    2. Some INT's (I'd say about 1/3) are random--tipped ball, INT instead of INC on Hail Mary, etc. This drags down any correlation data.
    3. The long-term data on a single player IS significant. You mentioned the 1 INT per 200 throws for a less accurate QB vs. Drew Brees. Over a career, that will be 25/30 INT's. I would suspect that 60% comp.% QB vs 50% comp.% QB is greater.
    4. Why is it that certain top-tier QB's generally have less INT's vs. "One of the 5 worst starters in the league" type? For example, if you took bets on, "Who will have the least INT's PER PASS in 2010--Drew Brees, JaMarcus Russell, or David Garrard?" 90% would take Brees, and maybe 10% would take Garrard, and nobody except a couple of drunk guys in Oakland would take JMR. And probably the overwhelming majority would be right. Now, if INT's (or INT rate) is basically random, why can we predict this with almost 100% certainty?

  2. Chase Stuart Says:
    March 1st, 2010 at 1:14 pm

    Joseph,

    Regarding your first point, the table above is sortable.

    On your last point, I think that's a great idea for a PFR contest. Note, btw, that in 2008, David Garrard and JaMarcus Russell had lower INT rates than Drew Brees. Russell barely played in '07, but Garrard had a lower INT rate than Brees in '07 as well. Your post is a good example of hindsight bias, where things seem so obvious in retrospect when they really weren't clear-cut at all. Who would have guessed that Favre and Aaron Rodgers would have the two lowest INT rates in 2009? Or that Matt Schaub (along with Brady Quinn and Vince Young) would have a better INT rate than Peyton Manning?

  3. Jason Lisk Says:
    March 1st, 2010 at 1:38 pm

    I was going to say, Chase, that in 2008, the correct order was the exact opposite of what Joseph suggests, and the drunks would have gotten it right.

    I agree with Joseph's first two points. Both of those things are beyond the quarterback's control. I've seen enough passes that should have been intercepted that are dropped, and enough crazy plays like the one of Jason Witten's foot this year that are intercepted. Add in passes that bounce off receivers hands or chest and go the wrong way (versus those that bounce harmlessly to the ground) and add in the hail marys, and you are going to destroy alot of the correlation.

    I'll add a couple of more. First, game and team context. Interception rates are highest for qb's who are trailing close and late, and there actually higher for QB's who are within one score compared to say, someone down 20 in the fourth. QB's will also take more risks when their own defense stinks. If one QB faces more close game situations in one part of the year than the other, then that will increase or decrease his int rates.

    Second, self-regulation. If I have thrown alot of interceptions early, even if they were not all my fault, I may be more careful, particularly if my job is on the line. I may hold the ball longer and not make as many risky throws. People dont harp on things like completion percentage like they do interceptions. These QB's will hear about their int's, and it may be a behavior deterrent. I'd be interested to look at whether QB's who throw int's early take sacks late, and vice versa, to over-correct their early flaw.

    I am surprised though, that the correlation isn't a little higher within season with the same QB. The random bounces and drops thing will keep it down, but it is the same guy, and presumably, the same defense and team situation most of the time. I'm wondering if we have a little bit of survivor bias bringing it down a little. If you throw lots of ints early and continue throwing them late, you may not get to 150 more passes after game 8. Of the 24 quarterbacks who were 2.0% worse than average early at throwing interceptions, 12 were better than average, 11 were worse, and 1 was average. It looks like it was random, because we don't have the guys who got benched before 150 passes late. Compare that to early guys who avoided int's. Of the 43 who were 2.0% or better than average, 32 were better than average late also. Those guys were likely not in threat of losing their jobs, and they regressed, but most were still better than average because the team situations were similar.

  4. Chase Stuart Says:
    March 1st, 2010 at 1:47 pm

    Good points, Jason.

  5. Rodney Says:
    March 1st, 2010 at 1:53 pm

    Chase - I think part of the problem is you need to look at Ints alongside TDs. If a QB throws 3 Ints in a game, but also 5 TDs including the game winner, nobody's going to care about the 3 picks. If a QB throws no TDs and 2 Ints, including the one that lost the game, those 2 picks are a lot worse than the 3 picks. You also give poor examples for 2009 as far as perception of QB quality, because Favre is a good QB who is just inconsistent, Rodgers is a good QB, Schaub is an average QB throwing to Andre Johnson, and Vince Young is a good QB who has been inconsistent in the past, and also had Chris Johnson as a RB this past season. So yes, I would expect all of those guys to put up Peyton Manning-like numbers. As far as Brady Quinn, he may have had only 7 ints, but he also had only 8 TDs, so that's not very good. Let's look at the 2008 guys you mentioned: Russell had an Int-TD rate of 62% (8-13), Garrard 87% (13-15), and Brees 50% (17-34). So in that context, Ints can be a good indication of who is the better QB.

    I'd like to see what you just did with Int rate done with Int-TD rate. I normally use other stats to rate QBs, so I've always found a quick look at the Int-TD rate to be a good way to check if the other stats hold up. I think most of the time the good QBs have the lowest rates.

  6. BP Says:
    March 1st, 2010 at 1:57 pm

    When you split the season in half, you have to account for the weather variable. Is passing more difficult in the second half of the season for outdoor teams? What if you split the season into even/odd games, also ensuring equal number of home/road games in each split. Would there be more of a correlation?

  7. Chase Stuart Says:
    March 1st, 2010 at 2:11 pm

    BP,

    By adjusting for era within each half-season, you solve that problem. The league average rate for the second-half season is not the same as the first-half season. I simply compared the QB's performance relative to league average in each half-season in the table above.

  8. Jim A Says:
    March 2nd, 2010 at 11:04 am

    I agree with Jason about game context. We know that interception rates are higher (on a per pass basis) when a team is trailing and it needs to take more risks. And INT rates are lower when a team is leading and can play it safe. There is some predictability to which teams will be in these situations. I'll bet that if you were to adjust for game context, the QB INT rate correlations would almost completely disappear.

  9. Scott Kacsmar Says:
    March 2nd, 2010 at 12:49 pm

    One game can really screw your season up on INT numbers. One of those "it's just not your day" 4-6 pick games are hard to overcome no matter who you are. It happened to Kurt Warner this year when he threw 5 picks against the Panthers (14 total on the season).

  10. Scott Kacsmar Says:
    March 2nd, 2010 at 12:53 pm

    Expanding on above's post and the previous article...

    It'd be a lot more work, but interesting to see what happens if you remove the worst game of the season (either by total INTs or highest %) for the QB.

    I'd imagine you would rather have the QB that has one of those 4-6 pick games but stays pretty turnover-free the rest of the season compared to someone that's consistently throwing one every week. Because that just means you're (likely) sunk for that day as the same defense keeps owning you, but overall you're protecting the ball better on the season even if the final numbers say otherwise.

  11. Red Says:
    March 2nd, 2010 at 7:26 pm

    Assuming that INT rates are essentially random, why do your AY/A and ANY/A metrics carry such a heavy penalty (-45) for interceptions?

    I would argue that the -45 penalty is appropriate for TEAM statistics, but not for the QB individually, given that your research shows that a majority of INT's happen for reasons other than the QB (receivers, game situation, defense, etc).

  12. Jim A Says:
    March 3rd, 2010 at 10:36 am

    Red, I think it comes down to retrodictive vs. predictive stats. Interceptions certainly help explain why a game was won, and it's probably true that a QB is still more responsible for interceptions than any other single player. But interceptions tell you hardly anything about whether a QB is more likely to throw INTs in the future.

    Unfortunately, people tend to use AY/A and ANY/A (and for that matter passer rating) to measure a QB's past performance and future expectations (talent/ability) fairly interchangably. But clearly the interception penalty is only appropriate in the retrodictive usage of passing stats, and I think it's arguable whether that usage even has any practical value.

  13. Jim A Says:
    March 3rd, 2010 at 11:20 am

    Scott, I think you'd probably find that many of these high INT games are the result of a team playing from behind and the QB is forced to make higher risk passes. So maybe the first couple INTs are "legitimate" interceptions but the later ones are more dictated by the game situation and don't necessarily mean that the QB is having a bad day. But definitely these games can skew overall INT rates. And interceptions are so rare and unpredictable as it is, I'd guess you wouldn't find many meaningful differences in the game-to-game consistency of INT rates among QBs, either.

  14. Scott Kacsmar Says:
    March 3rd, 2010 at 11:39 am

    I think what happens a lot of times is the QB throws an early INT and there's just a snowball effect from there and he can't seem to recover. He's always playing an uphill battle and just takes more and more risks to make up for the early mistakes.

    I'd be curious to see what % of 4+ INT games saw the QB make all of those picks in the 2nd half.

  15. Jim A Says:
    March 3rd, 2010 at 12:42 pm

    Yeah, I've seen plenty of those games, and it does seem like some QBs have more trouble recovering than others. I'm just wondering whether that snowball effect results from some conscious change in effort on the part of the QB or rather he's just gone cold. Kind of a variation of the old problem of whether streaks are meaningful or just random sequences.

  16. Rodney Says:
    March 3rd, 2010 at 1:39 pm

    Does anyone else think TD/Int ratio is more meaningful than Ints on their own? No one responded to my initial reply (#5). If it is more meaningful, what exactly is the point of this blog entry?

  17. Scott Kacsmar Says:
    March 3rd, 2010 at 2:23 pm

    I don't think TDs have anything to do with interception rates here.

    What bothers me is how people never want to look at fumbles for QBs, as if INTs are the only kind of turnovers you can have. Is it because of the flukey nature of fumbling/fumble recoveries? Because I've seen some QB fumbles that are far worse plays by the QB compared to some INTs. They both have their flukes, whether it's (for INTs) tipped balls, deflected balls, hail mary's, or for fumbles, the RB doesn't get the handle and the QB is charged with the fumble, or the center's snap isn't too good and he gets hit with that one too.

    I came up with fumble rate a year ago or so and just looked at total fumbles since recoveries are pretty random and we don't have fumble lost data prior to 1991. Any time you cough the ball up there's going to be an opportunity for a turnover.

    Fumble rate = Total fumbles / (Pass Attempts + Sacks + Rushes)

    No one protected the ball better than Bert Jones, with a 0.53% fumble rate on his dropbacks. Peyton Manning is 2nd at 0.76%. Namath, Montana and Joey Harrington (!) round out the top 5. Technically Sid Luckman's up there but we know he fumbled more than his incomplete stats say he did. No one fumbled at a higher rate than Jack Trudeau (2.82%).

    Then I took that and added INTs to get a Turnover Rate, which may be more appropriately called Turnover Opportunity Rate since I am using fumbles that didn't always result in a turnover.

    Here's the full list, best to worst, for Turnover Rate for all 156 QBs with 1500+ career pass attempts. The two really old QBs (Baugh and Luckman) have incomplete fumble numbers, and quite a few QBs don't have complete sack numbers. So keep that in mind when you see them so low, knowing that they're already low to begin with because of the different style of game played then. They would all be a bit higher with complete data.

    1. David Garrard (3.03%)
    2. Joe Montana (3.12%)
    3. Neil O'Donnell (3.21%)
    4. Donovan McNabb (3.22%)
    5. Jeff Garcia (3.28%)
    6. Mark Brunell (3.31%)
    7. Peyton Manning (3.32%)
    8. Steve Young (3.35%)
    9. Tom Brady (3.54%)
    10. Bernie Kosar (3.61%)
    11. Matt Hasselbeck (3.64%)
    12. Rich Gannon (3.64%)
    13. Drew Brees (3.66%)
    14. Chad Pennington (3.74%)
    15. Troy Aikman (3.75%)
    16. Philip Rivers (3.76%)
    17. Jason Campbell (3.79%)
    18. Marc Bulger (3.81%)
    19. Jim McMahon (3.83%)
    20. Brad Johnson (3.83%)
    21. Bert Jones (3.86%)
    22. Joey Harrington (3.92%)
    23. Carson Palmer (3.95%)
    24. Steve McNair (3.99%)
    25. Steve Bono (4%)
    26. Ken Anderson (4.02%)
    27. Jim Harbaugh (4.05%)
    28. Dan Marino (4.05%)
    29. Jeff George (4.07%)
    30. Doug Flutie (4.08%)
    31. Kordell Stewart (4.09%)
    32. Neil Lomax (4.09%)
    33. Trent Green (4.1%)
    34. Tony Eason (4.11%)
    35. Ken O'Brien (4.12%)
    36. Elvis Grbac (4.13%)
    37. Jeff Hostetler (4.16%)
    38. Ben Roethlisberger (4.17%)
    39. Joe Theismann (4.19%)
    40. Byron Leftwich (4.24%)
    41. John Elway (4.25%)
    42. Jeff Blake (4.27%)
    43. Randall Cunningham (4.31%)
    44. Michael Vick (4.31%)
    45. Drew Bledsoe (4.35%)
    46. Brett Favre (4.37%)
    47. Aaron Brooks (4.38%)
    48. Jim Everett (4.4%)
    49. Stan Humphries (4.45%)
    50. Roger Staubach (4.46%)
    51. Bubby Brister (4.47%)
    52. Rick Mirer (4.47%)
    53. Eli Manning (4.56%)
    54. Phil Simms (4.57%)
    55. Fran Tarkenton (4.59%)
    56. Erik Kramer (4.62%)
    57. Jay Fiedler (4.63%)
    58. Jim Kelly (4.64%)
    59. Brian Griese (4.65%)
    60. Jake Plummer (4.66%)
    61. Mark Rypien (4.69%)
    62. Chris Miller (4.7%)
    63. Bill Kenney (4.7%)
    64. Greg Landry (4.73%)
    65. Dave Brown (4.73%)
    66. Jay Cutler (4.77%)
    67. Doug Williams (4.77%)
    68. Kerry Collins (4.83%)
    69. Wade Wilson (4.85%)
    70. Gus Frerotte (4.87%)
    71. Roman Gabriel (4.88%)
    72. Jake Delhomme (4.93%)
    73. David Carr (4.97%)
    74. Bobby Hebert (4.98%)
    75. Ron Jaworski (4.98%)
    76. Steve Beuerlein (4.99%)
    77. Warren Moon (5.04%)
    78. Bill Nelsen (5.06%)
    79. Vinny Testaverde (5.07%)
    80. Bill Munson (5.08%)
    81. Sonny Jurgensen (5.1%)
    82. Kurt Warner (5.11%)
    83. Chris Chandler (5.13%)
    84. Boomer Esiason (5.14%)
    85. Tony Banks (5.16%)
    86. Steve Bartkowski (5.16%)
    87. Gary Danielson (5.17%)
    88. Daunte Culpepper (5.19%)
    89. Tim Couch (5.19%)
    90. Steve Deberg (5.2%)
    91. Mike Pagel (5.23%)
    92. Tommy Kramer (5.27%)
    93. Jay Schroeder (5.28%)
    94. Jim Zorn (5.29%)
    95. Scott Mitchell (5.31%)
    96. Rodney Peete (5.31%)
    97. Billy Joe Tolliver (5.33%)
    98. Jon Kitna (5.41%)
    99. Brian Sipe (5.43%)
    100. Archie Manning (5.56%)
    101. Bart Starr (5.56%)
    102. Bob Griese (5.57%)
    103. Mike Tomczak (5.59%)
    104. John Brodie (5.59%)
    105. Joe Ferguson (5.61%)
    106. Jim Hart (5.66%)
    107. Dave Krieg (5.66%)
    108. Dan Fouts (5.66%)
    109. Trent Dilfer (5.73%)
    110. Billy Kilmer (5.74%)
    111. Craig Morton (5.79%)
    112. Daryle Lamonica (5.86%)
    113. Frank Ryan (5.9%)
    114. Jim Plunkett (5.9%)
    115. Mike Livingston (5.93%)
    116. Johnny Unitas (5.93%)
    117. Milt Plum (5.94%)
    118. Danny White (5.96%)
    119. Mark Malone (6.02%)
    120. Richard Todd (6.05%)
    121. Marc Wilson (6.09%)
    122. Tom Flores (6.09%)
    123. Charley Johnson (6.19%)
    124. Vince Ferragamo (6.22%)
    125. Len Dawson (6.28%)
    126. Terry Bradshaw (6.32%)
    127. Eric Hipple (6.32%)
    128. Joe Namath (6.35%)
    129. John Hadl (6.51%)
    130. Dan Pastorini (6.54%)
    131. Norm Snead (6.56%)
    132. Jack Trudeau (6.56%)
    133. Don Meredith (6.58%)
    134. Mike Phipps (6.61%)
    135. Billy Wade (6.61%)
    136. Y.A. Tittle (6.66%)
    137. Steve Grogan (6.74%)
    138. Earl Morrall (6.81%)
    139. Lynn Dickey (6.82%)
    140. Don Majkowski (6.86%)
    141. Ken Stabler (6.87%)
    142. Otto Graham (6.89%)
    143. Charlie Conerly (7.12%)
    144. Tobin Rote (7.28%)
    145. Norm Van Brocklin (7.44%)
    146. Jack Kemp (7.48%)
    147. Bobby Layne (7.49%)
    148. Sammy Baugh (7.53%)
    149. Sid Luckman (7.55%)
    150. Cotton Davidson (7.6%)
    151. Ed Brown (7.63%)
    152. Babe Parilli (7.77%)
    153. George Blanda (7.79%)
    154. Frank Tripucka (8.11%)
    155. Eddie LeBaron (9.22%)
    156. Bob Waterfield (9.69%)

    David Garrard really surprised me. He's elite in both fumble rate and INT%. I was praising him for it before the season, then of course he went on to lead the league in fumbles with 14 in 2009. Well he's still in possession of first place.

  18. Scott Kacsmar Says:
    March 3rd, 2010 at 2:36 pm

    Guess I left out Tony Romo. He'd be 51st, right behind Roger Staubach with a 4.46%

  19. Red Says:
    March 3rd, 2010 at 11:13 pm

    Jim A, thanks for the response, and I mostly agree with you. INT rates are definitely useless for predictive stats. Retrodictive stats can be trickier, because there's two things you can use them to measure: 1) what actually happened, and 2) how well the given player performed. Those are often interchangable, but not always. A fluky tipped-ball INT might determine the outcome of a game, but it doesn't have anything to do with how well the QB played. Also, on average, are attempts that result in INT's any worse of a throw/decision than throws that result in incompletions? I would argue not.

    Roger, I do think TD/INT is more accurate than INT's alone, at least over a full season. A QB with 40 TD's and 20 INT's is probably helping his team win more than a QB with 10 and 10. In fairness, though, I think both TD's and INT's are overrated stats.

  20. DSMok1 Says:
    March 4th, 2010 at 10:37 am

    Great work here... it looks like the predictive power of interceptions is just about dispelled.

    Are you going to do similar work with fumbles? That would be very interesting.

    Actually, split half correlations on EVERYTHING would be very informative for creating predictive metrics.

  21. Jim A Says:
    March 4th, 2010 at 12:04 pm

    We know that QB sack rates are highly predictable, so I'd guess that QB fumble rates are also fairly predictable at least on passing plays, though probably not so much on running plays or botched snaps. It would be interesting to see whether some QBs are more to prone to fumble on sacks after adjusting for sack rate. I'm not sure that combining fumble and interception rates tells you much about a QBs ability to avoid turnovers, though.

  22. Zach Says:
    March 5th, 2010 at 4:51 pm

    The average number of attempts by all the quarterbacks in the sample is 237. If you use Tangotiger's formula of (1-r)/r*n, it takes (1-.12)/.12*237 = 1736 attempts for regression to the mean to be 50%. My own research, looking at QB INTs year-to-year since 1980, says that number is about 1300. So, it takes about three to four full seasons for a QB's INT rate to regress 50% to the mean!

  23. Bob S. Says:
    June 23rd, 2010 at 4:52 pm

    Rodney #16, I agree with you 100% that you must consider TD passes along with Int's. The bigger the differential (TD passes minus Int's) the less impact the Int's had on a QB's team. When a QB throws an Int., his opponent has a chance to turn it into either 0, 3, or 6 points. But when a QB throws a TD Pass, it results in 6 points every time, so it is a counter to his Int's. Right now, nobody has a better differential than Brett Favre for their entire career (194 diff.), although Peyton manning is also at 194. In the regular season, Manning is 1st at 185 and Favre is 2nd at 180.

Leave a Reply

HTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

«
»