Stat Analysis

Advanced analytics on player and team performance

Adjusted Interceptions 2017

by Vincent Verhei

EDITOR'S NOTE: After publication, we realized our totals for pass attempts for all players had been miscalculated, and as a result player's interception rates were also incorrect. The data table now has been fixed, and the text referencing those errors has been corrected where applicable.

When is an interception not an interception? Typically, when a defender drops a pass that he should have caught, or when a wide receiver makes a big play to turn what should have been a turnover into an incompletion instead. On the other hand, sometimes quarterbacks are charged with interceptions that aren't really their fault -- Hail Mary passes, for example, or those that bounce off a receiver's hands and into a defender's.

After each season, we go back and account for these discrepancies and account for each quarterback's adjusted interceptions. Here's the process:

  • We start with each player's actual interception total. DeShone Kizer led the NFL in 2017 with 22 interceptions.
  • We then add plays where the quarterback threw a ball that could have or should have been intercepted but was not, either because the defender outright dropped the ball (which we have been tracking in game charting since 2007), or he had it knocked out of his hands by an offensive receiver (a "defensed interception," which we have been tracking since 2012). These are listed as "Drop/Def INT" in the table at the end of this page. Kirk Cousins and Matthew Stafford tied for the league lead with nine such passes each. Aaron Rodgers was the only qualifying passer to not benefit from a dropped interception.
  • Next, we subtract those interceptions that were tipped by receivers into the hands of defenders. Matt Ryan threw five interceptions that should have been caught by his own teammates. That wasn't just the most in the league, it was the most for any quarterback in a single season since Eli Manning also had five in 2010. We also subtract passes that are tipped by receivers but then dropped by defenders to make sure they are not double-counted. C.J. Beathard, Mitchell Trubisky, and Andy Dalton each had one play like that last season.
  • Finally, we subtract Hail Mary interceptions, as well as interceptions thrown in desperation on fourth down in the final two minutes of a game. (We're sometimes flexible on these definitions, but not in 2017 -- there just weren't any gray area interceptions last season.) These are listed as "HM/End Q4" in the table. Kizer and his homonymic peer Deshaun Watson each had two such interceptions last year; nobody else had more than one.

When we started running these numbers, we had to get the data from our own in-house volunteer game charters. For the last three seasons, we've had access to data from Sports Info Solutions. Determining whether or not a defender should be charged with a dropped interception will always be subjective on some plays, but you can rest assured that all the obvious calls have been counted here.

Enough with the minutia -- let's get to the big table of numbers!


Adjusted Interceptions, 2017
Name Team INT HM/
End Q4
Drop/
Def INT
Tip INT Tip AND
Drop INT
Adj INT Pass Att
(no DPI)
INT
Rate
Adj
Rate
7-D.Kizer CLE 22 2 6 3 0 23 514 4.3% 4.5%
13-T.Siemian DEN 14 0 6 0 0 20 382 3.7% 5.2%
8-K.Cousins WAS 13 0 9 2 0 20 586 2.2% 3.4%
3-J.Winston TB 11 0 8 0 0 19 474 2.3% 4.0%
4-D.Carr OAK 13 1 8 1 0 19 534 2.4% 3.6%
7-B.Roethlisberger PIT 14 0 4 0 0 18 580 2.4% 3.1%
9-M.Stafford DET 10 0 9 1 0 18 613 1.6% 2.9%
6-J.Cutler MIA 14 0 4 1 0 17 451 3.1% 3.8%
8-M.Mariota TEN 15 0 2 0 0 17 479 3.1% 3.5%
4-D.Prescott DAL 13 1 8 3 0 17 520 2.5% 3.3%
5-J.Flacco BAL 13 0 8 4 0 17 574 2.3% 3.0%
10-E.Manning NYG 13 1 5 0 0 17 602 2.2% 2.8%
1-C.Newton CAR 16 1 3 2 0 16 525 3.0% 3.0%
14-A.Dalton CIN 12 0 7 2 1 16 536 2.2% 3.0%
5-B.Bortles JAC 13 0 5 2 0 16 547 2.4% 2.9%
17-P.Rivers LACH 10 1 7 0 0 16 594 1.7% 2.7%
3-R.Wilson SEA 11 0 5 1 0 15 598 1.8% 2.5%
3-C.Palmer ARI 7 0 7 0 0 14 289 2.4% 4.8%
Name Team INT HM/
End Q4
Drop/
Def INT
Tip INT Tip AND
Drop INT
Adj INT Pass Att
(no DPI)
INT
Rate
Adj
Rate
15-J.McCown NYJ 9 0 5 0 0 14 437 2.1% 3.2%
11-C.Wentz PHI 7 0 6 0 0 13 471 1.5% 2.8%
12-T.Brady NE 8 0 5 0 0 13 617 1.3% 2.1%
7-B.Hundley GB 12 0 1 1 0 12 347 3.5% 3.5%
9-D.Brees NO 8 0 5 1 0 12 555 1.4% 2.2%
7-J.Brissett IND 7 0 5 1 0 11 518 1.4% 2.1%
16-J.Goff LARM 7 0 4 1 0 10 500 1.4% 2.0%
7-C.Keenum MIN 7 0 5 2 0 10 503 1.4% 2.0%
11-A.Smith KC 5 0 5 0 0 10 541 0.9% 1.8%
7-B.Gabbert ARI 6 0 3 0 0 9 194 3.1% 4.6%
3-T.Savage HOU 6 0 3 0 0 9 243 2.5% 3.7%
2-M.Ryan ATL 12 0 2 5 0 9 552 2.2% 1.6%
4-D.Watson HOU 8 2 2 0 0 8 221 3.6% 3.6%
2-B.Hoyer SF 4 0 4 0 0 8 228 1.8% 3.5%
10-M.Trubisky CHI 7 1 4 1 1 8 362 1.9% 2.2%
12-A.Rodgers GB 6 0 0 0 0 6 259 2.3% 2.3%
3-C.Beathard SF 6 1 1 1 1 4 243 2.5% 1.6%
5-T.Taylor BUF 4 0 2 2 0 4 467 0.9% 0.9%
Minimum 200 passes, plus Blaine Gabbert, because we forgot to eliminate him after we re-calculated pass attempts.

After all that work, Kizer still leads the league in adjusted interceptions, but the gap between him and the rest of the league has shrunk considerably. It's certainly not a good thing that Kizer threw 23 adjusted interceptions, but it's hardly unprecedented -- Ben Roethlisberger and Eli Manning both threw more than that last year. Kizer didn't get a lot of help from Lady Luck -- defenders only dropped six of his passes, while five came on either Hail Marys or drops from his receivers. Quarterbacks typically finish with about three-fourths (77 percent, to be more precise) as many actual interceptions as adjusted interceptions. Since Kizer had 23 adjusted interceptions, we would have expected him to throw 17 or 18 actual picks, not the 22 you'll find on his page at NFL.com.

The three quarterbacks who threw 20 or more adjusted interceptions in 2017 will all play for new teams this year. Kizer was traded to Green Bay, while Trevor Siemian and Kirk Cousins both moved to Minnesota, Siemian via trade and Cousins in free agency. Siemian only threw 406 passes last year and actually had the league's highest rate of adjusted interceptions, but then he's going to be a backup for the Vikings, and you'd expect backup quarterbacks to struggle with ball security. Cousins, however, was briefly the highest-paid player in the league by average salary. He has since been passed in that category by Matt Ryan, but that doesn't mean Minnesota's expectations for him have lowered. If his luck comes back to earth, you can expect your Vikings fan friends to be extra grouchy.

If we go by rate of adjusted interceptions instead of totals, we find Siemian first at 4.9 percent 5.2 percent, followed by a pair of Cardinals: Carson Palmer (4.8 percent) and Blaine Gabbert (4.3 percent 4.6 percent). Between Palmer, Gabbert, and Drew Stanton, the Cardinals threw 28 adjusted interceptions, which is why none of the three will be Cardinals in 2018. Kizer, meanwhile, combined with Kevin Hogan and Cody Kessler to throw 31 adjusted interceptions in Cleveland, which is why none of them will be with the Browns in 2018. Instead, Cleveland will (probably) be quarterbacked by Tyrod Taylor, who had the lowest adjusted interception rate in the league, followed by San Francisco's C.J. Beathard and Atlanta's Matt Ryan.

If those last two names are surprising for you, well, that's the whole point of adjusted interceptions. We mentioned Ryan's terrible luck with tipped interceptions last year, and that most quarterbacks end up with significantly more adjusted interceptions than actual interceptions. Well Ryan's luck was so bad that he actually threw three fewer adjusted interceptions than actual interceptions, the first quarterback to do that since Tom Brady in 2013. Beathard was right behind him, with two more actual interceptions (six) than adjusted interceptions (four). And Beathard threw less than half as many passes as Ryan did!

Beathard won't be starting in San Francisco this year, of course; Jimmy Garoppolo will. For all his success last season, Garoppolo threw five interceptions in 188 passes, a higher rate than most quarterbacks. And he had three dropped interceptions for a total of eight adjusted interceptions, a rate of 4.3 percent that would have ranked right behind Siemian, Palmer, and Gabbert among the highest in that category.

Here's a look at adjusted interceptions in prior seasons.

Comments

29 comments, Last at 04 Jun 2018, 9:46pm

1 Re: Adjusted Interceptions 2017

by dbostedo // May 30, 2018 - 3:49pm

How well do adjusted interceptions correlate with future actual interception numbers? Is this really a predictive thing? Or just a look at who's been luckiest?

Login or Register to post comments

2 Re: Adjusted Interceptions 2017

by Aaron Brooks Good Twin // May 30, 2018 - 4:10pm

It seems to suggest who is profligate with the ball and who is stingy.

Taylor and Brady have been on the low end of adjusted rates for years. High guys seem to skew high. Some, like Eli, seem to wander.

Login or Register to post comments

3 Re: Adjusted Interceptions 2017

by Aaron Schatz // May 30, 2018 - 5:24pm

At some point we did look at it, and the adjusted interceptions correlated better with interceptions the following year, but I can't remember exactly what the numbers were. Sorry about that.

Login or Register to post comments

4 Re: Adjusted Interceptions 2017

by Vincent Verhei // May 30, 2018 - 8:04pm

This is a tiny little sample size but it's data I collected for this and didn't end up using.

53 QBs had 200 pass attempts in 2015/2016 or 2016/2017. The year-to-year correlation rates were:

INT rate, Year N, to INT rate, Year N+1: 0.212
INT rate, Year N, to ADJ INT rate, Year N+1: 0.190
ADJ INT rate, Year N, to INT rate, Year N+1: 0.396
ADJ INT rate, Year N, to ADJ INT rate, Year N+1: 0.308

So adjusted interceptions seem to have more predictive value than un-adjusted interceptions, at least in the last two years.

Login or Register to post comments

6 Re: Adjusted Interceptions 2017

by jeffd // May 31, 2018 - 10:43am

"So adjusted interceptions seem to have more predictive value than un-adjusted interceptions, at least in the last two years."

This doesn't surprise me. The people dropping interceptions varies from week to week so I'd expect it to be much closer to a random event, so you'd expect mean regression.

Login or Register to post comments

8 Re: Adjusted Interceptions 2017

by JoeyHarringtonsPiano // May 31, 2018 - 11:53am

I see this as roughly analagous to total fumbles vs fumbles recovered by the defense. Both a fumble, and a bad pass that's catch-able by a defender are potential turnovers. Whether they actually become a turnover comes down to luck. These numbers suggest the only difference is that with average luck, fumbles are closer to 50-50, while bad passes are more 75/25.

Login or Register to post comments

5 Re: Adjusted Interceptions 2017

by andrew // May 31, 2018 - 9:46am

d

Login or Register to post comments

7 Re: Adjusted Interceptions 2017

by Led // May 31, 2018 - 10:59am

-fense

Login or Register to post comments

10 Re: Adjusted Interceptions 2017

by Mountain Time ---- formerly Ninjalectual // May 31, 2018 - 2:44pm

+1 white picket fence

Login or Register to post comments

9 Re: Adjusted Interceptions 2017

by Sixknots // May 31, 2018 - 12:26pm

Russell Wilson had the most pass attempts in the NFL last season!? In a Pete Carroll offense!? No wonder the Seahawks brought in run blocking linemen and TEs and drafted a RB in the 1st round.

Login or Register to post comments

12 Re: Adjusted Interceptions 2017

by Mountain Time ---- formerly Ninjalectual // May 31, 2018 - 2:48pm

"No wonder" could be interpreted sarcastically (I suspect this is what you meant), but also seriously, as in what the Seahawks did would indeed theoretically allow them to call more and more effective run plays next season.

Login or Register to post comments

15 Re: Adjusted Interceptions 2017

by Aaron Schatz // May 31, 2018 - 5:35pm

See above. This was a weird miscalculation error. We've fixed the numbers.

Login or Register to post comments

17 Re: Adjusted Interceptions 2017

by Mountain Time ---- formerly Ninjalectual // May 31, 2018 - 7:25pm

Ah, it seemed reasonable enough that Wilson led the NFL in attempts. Apparently he was 28 off the mark in 6th

Login or Register to post comments

18 Re: Adjusted Interceptions 2017

by Sixknots // May 31, 2018 - 11:24pm

OK, but the "fix" only drops him to fourth behind Brady, Stafford and Manning. That's not a Pete Carroll "pass off the run" offense.

Login or Register to post comments

11 Re: Adjusted Interceptions 2017

by andrew // May 31, 2018 - 2:47pm

There is always Kyle Sloter in Minnesota.

Call me crazy but I think he could end up being the long term answer....

Login or Register to post comments

14 Re: Adjusted Interceptions 2017

by Mountain Time ---- formerly Ninjalectual // May 31, 2018 - 3:01pm

You're crazy.

Remember how everyone is reluctant to guve Keenum the benefit of the doubt because of his "ideal situation"? Well, that's what a motivated Cousins has to work with now. And Seimein is really a victim of bad coaching asking him to have skills he doesn't. He's limited as a starting NFL QB; with Kubiak he was slightly below average (22nd/21st DYAR/DVOA in 2016). As a backup? That's better than solid, that suggests he's a potential starter somewhere with a more flexible coaching staff.

Don't get me wrong, I think Sloter has a chance of being on a roster as a backup, probably not this year, but who knows. I think Minnesota has an excellent stable of QBs, if not the embarrassment of riches they had last season.

Login or Register to post comments

13 Re: Adjusted Interceptions 2017

by andrew // May 31, 2018 - 2:50pm

So A-Rod is the only QB (qualifying) without a single dropped INT?

Just shows how catchable his throws are, I guess.

Login or Register to post comments

16 Re: Adjusted Interceptions 2017

by MC2 // May 31, 2018 - 6:42pm

The article notes how rare, and how unlucky, it is for a QB's actual INT total to be higher than his adjusted INT total. Yet it has happened to Matt Ryan 3 times in the last 7 years (2011, 2013, and 2017).

I've heard quite a few people here in the Atlanta area say that Ryan "isn't as good" as his stats would indicate. I guess they're right; he's better than his stats indicate.

Login or Register to post comments

19 Re: Adjusted Interceptions 2017

by dbostedo // Jun 01, 2018 - 9:23am

"I've heard quite a few people here in the Atlanta area say that Ryan "isn't as good" as his stats would indicate."

I'm guessing that's really just short hand for "the team hasn't won enough playoff games or Super Bowls".

I have heard that about Dan Marino, too, for instance... i.e. people who say things like "All those stats are meaningless - he couldn't even get to a second Super Bowl."

Login or Register to post comments

22 Re: Adjusted Interceptions 2017

by MC2 // Jun 01, 2018 - 6:58pm

That's certainly part of it, but even beyond that, the local fan base seems to lack an appreciation for how good Ryan has been, even compared to previous QBs that had even less postseason success.

For example, I imagine that if you did a poll of Falcon fans, and asked them who is the greatest QB in franchise history, Michael Vick would finish first, and I doubt that it would even be close. Even now, he's beloved by many people around here in a way that Ryan has never been, and probably never will be. In fact, in that hypothetical poll, Ryan might not finish all that far ahead of Steve Bartkowski for second place.

Login or Register to post comments

24 Re: Adjusted Interceptions 2017

by Mountain Time ---- formerly Ninjalectual // Jun 01, 2018 - 11:49pm

Local fanbases can be collectively stupid like that, but if Atlanta is like Denver, they'll figure it out around 2028 or so. And by then they'll always have been a big Ryan fan.

I remember for YEARS being the only guy in Denver who thought Jake Plummer was doing fine. "He throws too many interceptions" was the refrain about him, though it doesn't actually matter what specific issue the local fans latch on to, they're all cliches that only vaguely describe anything. "Not as good as his stats" is exactly the same.

Then a couple years ago, the Denver Post had a poll asking fans their favorite QB in between Elway and Manning. Who won? Jake Plummer. I couldn't even be happy people agreed with me finally, I'm still mad about the ridicule and dismissals from 2006!

Login or Register to post comments

25 Re: Adjusted Interceptions 2017

by dbostedo // Jun 02, 2018 - 10:20pm

Considering that the QBs between Elway and Manning were pretty much Brian Griese, Jake Plummer, Kyle Orton, and Tim Tebow, it's completely possible to think Plummer wasn't any good, and still have him be your favorite of those four.

Login or Register to post comments

20 Re: Adjusted Interceptions 2017

by Aaron Brooks Good Twin // Jun 01, 2018 - 3:46pm

Just for comedy purposes, what were Peterman's numbers?

Login or Register to post comments

21 Re: Adjusted Interceptions 2017

by justanothersteve // Jun 01, 2018 - 5:01pm

I just literally LOL'ed

Login or Register to post comments

23 Re: Adjusted Interceptions 2017

by Vincent Verhei // Jun 01, 2018 - 8:07pm

50 passes
5 actual interceptions
1 dropped/defensed
1 tipped
5 adjusted interceptions

Login or Register to post comments

26 Re: Adjusted Interceptions 2017

by T0r0 // Jun 03, 2018 - 11:48am

I think Wentz's numbers are wrong.
His first Int was a pick 6 in week 1 against the Redskins that was tipped at the line and picked off by ryan kerrigan.

finally his last Int of the season Was in the first drive of the Rams game that was a short out route to nelson agalor that bounced off agalors arms to the defender.

so that is 2 tipped picks not counted for Wentz

Login or Register to post comments

27 Re: Adjusted Interceptions 2017

by Vincent Verhei // Jun 03, 2018 - 3:39pm

The "Tip INT" column is for passes that should have been caught by receivers but were instead tipped into the air and then intercepted by a defender. It is not for passes that were tipped at the line of scrimmage. Those are on the quarterback.

I just watched the Agholor play against the Rams. In slow-mo you can see that Agholor didn't drop the ball, it was punched out by Nickell Robey-Coleman (https://twitter.com/FO_VVerhei/status/1003360002147414017). So this doesn't count either. I realize Joe Buck twice said that the ball bounced out of Agholor's hands, but it's important to realize that Joe Buck is a complete fool who is very bad at his job.

Login or Register to post comments

28 Re: Adjusted Interceptions 2017

by andrew // Jun 04, 2018 - 11:19am

The chart says "minimum 200 passes".

Yet it includes Blaine Gabbert, who only threw 194.

I must conclude the only reason for his inclusion is to further humiliate poor Blaine Gabbert.

Login or Register to post comments

29 Re: Adjusted Interceptions 2017

by Vincent Verhei // Jun 04, 2018 - 9:46pm

Heh. You're right, he should have been removed after we re-calculated pass attempts. I'll make a note.

Login or Register to post comments