RSS Feed
PDF NOW DISCOUNTED 50%. FIND OUT WHAT YOU MISSED BEFORE THE SEASON.
Click here to buy PDF version
.Click here to buy print version.
STILL AVAILABLE
Click here to buy PDF version
.Click here to buy print version.
28 Jul 2009
FO Mailbag: Dropped Interceptions
I knew I had been forgetting something lately... printing reader questions and FO's answers in Extra Points! Feel free to send your KUBIAK and FOA 2009 questions to mailbag-at-footballoutsiders.com.
Dan: The FOA appendix lists the defenders with the most dropped interceptions. Do you have a list of the quarterbacks who threw the most dropped interceptions? I'm especially interested in Cutler's numbers, given the criticism that he's been getting from K.C. Joyner.
Aaron: Sure, we can easily do that list. Here are the quarterbacks listed with at least five dropped interceptions according to 2008 game charting:
9: T.Thigpen
8: M.Bulger, A.Rodgers
7: M.Cassel
6: J.Campbell, J.Cutler, J.Delhomme, R.Fitzpatrick, E.Manning, D.McNabb, K.Warner
5: K.Orton, T.Romo, J.Russell
What's interesting is the missing name: Brett Favre, who led the league in interceptions last year. He had only four dropped interceptions.
Here's an "adjusted interceptions" top ten, adding in dropped interceptions and removing Hail Mary attempts:
1) B.Favre, NYJ: 25
2) J.Cutler, DEN: 24
3) A.Rodgers, GB: 21
4) D.Brees, NO: 20
4) M.Bulger, STL: 20
4) T.Thigpen, KC: 20
7) T.Romo, DAL: 19
8) K.Warner, ARI: 18
8) J.Delhomme, CAR: 18
8) M.Cassel, NE: 18
While we're on the subject of game charting... We've had a couple of requests from people who wanted the charting data for just one team. We're actually planning on introducing that product in the FO store sometime over the next couple weeks; you'll be able to buy three years of game charting data for a single team for $25.
38 comments, Last at 30 Jul 2009, 12:58pm by Dales
Most Recent FO Features
» Super Bowl XLVI Preview
Aaron Schatz pores over all the FO stats to break down Sunday's matchups. Do they agree with the oddsmakers and lean Patriots, or do they agree with the general public and lean Giants?
Most Recent Extra Points
Comments
Re: FO MAILBAG: Dropped Interceptions
Any statistical analysis regarding interceptions in 2008 which does not account for the singular awfulness of Gus Frerotte is obviously flawed. Back to the drawing board, Schatz!!
Re: FO Mailbag: Dropped Interceptions
Shouldn't the review of most interceptions for last year include a comparison to how many attempts each QB threw?
Re: FO Mailbag: Dropped Interceptions
Thanks, Aaron!
This looks good for Cutler. His adjusted interception rate (adjusted ints divided by attempts) was 3.9%, which was not as bad as Favre (4.8%), Thigpen (4.8%), Bulger (4.5%), Delhomme (4.3%), Romo (4.2%), or Rodgers (3.9%). A little bit higher than average, just like his interception rate was, but nothing terrible.
KC Joyner watches game tape and counts "bad decisions" (mistakes that lead to turnovers or near turnovers), and he has said that Cutler's bad decision rate of 4.6% was the worst in the league. I guess you have different definitions for what counts as a near interception. Cutler did have a ton of passes defensed according to the game charting data in FOA (2nd in the league in passes defensed per attempt), so maybe Joyner is counting more of those as near interceptions.
Re: FO Mailbag: Dropped Interceptions
I don't mean to start off another firestorm, but these numbers also imply that all those people out there who were loudly proclaiming "How could anyone ever think that Matt Cassel is remotely as good as Jay Cutler" were maybe exaggerating a bit. I'm not saying that Cassel was better than Cutler, but looking at these adjusted interception numbers certainly implies that Cassel made decisions at least as good, if not a little better, than Cutler.
What might be really interesting is to compare Cutler's interception rate and yards/INT ratio to Kyle Orton and Rex Grossman, to gauge what might happen in Denver and Chicago this year. Aaron, do you have the adjusted INT numbers for these guys?
Re: FO Mailbag: Dropped Interceptions
It either implies that Cassel made decisions at least as good, if not a little better, than Cutler... or else it implies that Cassel's defense allowed 9 fewer points per game than Cutler's.
The gap in their "decision-making" wasn't that big (3.49% for Cassel, 3.90% for Cutler), but the gap between their defenses was huge (over 21% DVOA). After watching the unbelievable transformation of Jake Plummer from Denver to Arizona I've gained new respect for the idea that there's a causal relationship between bad defenses and bad decisions. Plummer had a 4.1% int rate in Arizona (on bad teams that lost a lot and gave up a lot of points), and a 2.9% int rate in Denver (on good teams that won a lot and gave up much fewer points). If Cutler led the league in "bad decisions" in a season where his team didn't have the second worst defense in the history of DVOA, then I'd give a lot more credence to this theory that he's an incurable gambler who can't help but give the game away.
The fact that Mike Shanahan, probably the most run-loving coach of the past decade that didn't reside in Pittsburgh, the guy who called more runs than passes for four consecutive seasons from 2003-2006, oversaw an offense that called 620 passes to 387 runs should serve as a big indicator that we were hardly dealing with normal circumstances. In fact, prior to 2007 (when Denver's defense became truly terrible), Shanahan had never called fewer than 457 runs. Last year, he called 387. In addition to the abysmal defense, that sort of reliance on the pass also greatly contributed to Cutler's INT totals (since the defense was playing pass with far more frequency and commitment).
It also didn't help that Denver lost 7 RBs for the season and was forced to start the punchline that was Tatum Bell (the guy who stole Rudi Johnson's luggage and who no team that hadn't lost 7 RBs for the season even bothered to bring in for a tryout).
It's hard to exaggerate just how good Cutler was last season, because his circumstances need no exaggeration. His defense was the second worst defense of the last 15 years. More RBs were lost for the season on his team alone than most divisions combined for. Despite his high "bad decision" rate, he finished 7th in DVOA (and remember, DVOA penalizes SEVERELY for turnovers). Cassel, on the other hand, stepped into the greatest offense in NFL history and finished 20th in DVOA. Context is everything.
Re: FO Mailbag: Dropped Interceptions
Context is everything ... Denver had the best rushing DVOA in the league. So your running back argument doesn't mean much.
Clearly, the bad decision rate stuff is only a slice of the big picture.
But many things provided to support Cutler could be used to turn it around -- their defense sucked, so he was playing from behind against (often) softer zones (see rushing DVOA above) etc.
As a Charger fan who doesn't like Cutler's attitude, I'd take him over Cassel any day. I'd take him over all but about 7 or so other QB's.
Funny how that intuition lines up with the DVOA ranking.
Re: FO Mailbag: Dropped Interceptions
Much as I love DVOA, and much as I think that Denver's rushing attack remained potent despite all the injuries (in large part due to that ridiculously underrated line), I feel obligated to point out that DVOA is an efficiency stat, and offenses are going to be a heck of a lot more efficient doing something the defense doesn't expect them to do. To pull an example from the college ranks- according to the most recent Varsity Numbers article, Texas Tech had the third best rushing offense last year. Do you think that's because Tech is a hot destination for quality RB prospects, or because Tech is the first choice of elite OLine prospects from the state of Texas... or do you think that's because Mike Leach is a master of game theory and because opposing defenses are always shocked to see Tech run?
Denver had the most EFFICIENT rushing offense in the league last year, but if they really had the BEST rushing offense in the league last year, they likely would have run more than 357 times.
In the same way, playing from behind helps a QB compile cumulative statistics, but actually hurts the QB in terms of rate stats, because the defense knows that he's going to be passing and defends accordingly. It's fair to say that Cutler's yardage total was probably unnaturally inflated because he was playing with a historically awful defense, but it's not fair to say that playing with an awful defense should have given Cutler an advantage in terms of avoiding INTs. If anything, playing with a bad defense should have INCREASED Cutler's interceptions, because as your odds of winning decrease you're forced to take bigger and bigger risks, and also because as passing becomes your only option opposing defenses are going to devote more and more resources to shutting down that option.
To borrow a bit of context from www.advancednflstats.com... Advanced NFL Stats has a "win probability" (WP) stat that's designed to calculate the average chance a given team has of winning the football game at any particular point in time. The average WP of a team immediately before its QB threw an interception was .41, which makes sense (a QB is more likely to throw an INT when his team has a below average chance to win because he's forcing things). The average WP of the Denver Broncos immediately before Jay Cutler threw an INT last year was .40. Essentially, Jay Cutler doesn't throw more INTs than the average QB, he's just in more situations likely to result in an INT than the average QB.
Re: FO Mailbag: Dropped Interceptions
"To borrow a bit of context from www.advancednflstats.com... Advanced NFL Stats has a "win probability" (WP) stat that's designed to calculate the average chance a given team has of winning the football game at any particular point in time. The average WP of a team immediately before its QB threw an interception was .41, which makes sense (a QB is more likely to throw an INT when his team has a below average chance to win because he's forcing things). The average WP of the Denver Broncos immediately before Jay Cutler threw an INT last year was .40. Essentially, Jay Cutler doesn't throw more INTs than the average QB, he's just in more situations likely to result in an INT than the average QB."
There are a couple reasons why the average WP of a team immediately before its QB threw an interception was .41
A) The teams with the best qb's on average win more games. The alternative is that the average "worse" qb is trying to come from behind and since he isn't as good he throws more interceptions
2)Inflated extremes. A qb that is doing a 2 min drill but behind by 9+ points is gonna take a million chances. the win percentage average in a situation similar to this has got to be close to the 5%-0% range. Alternatively no team is gonna be throwing that late in a game with a 100%-95% chance WP. Low extremes inflate the importance of an already unimportant stat
Re: FO Mailbag: Dropped Interceptions
"To borrow a bit of context from www.advancednflstats.com... Advanced NFL Stats has a "win probability" (WP) stat that's designed to calculate the average chance a given team has of winning the football game at any particular point in time. The average WP of a team immediately before its QB threw an interception was .41, which makes sense (a QB is more likely to throw an INT when his team has a below average chance to win because he's forcing things). The average WP of the Denver Broncos immediately before Jay Cutler threw an INT last year was .40. Essentially, Jay Cutler doesn't throw more INTs than the average QB, he's just in more situations likely to result in an INT than the average QB."
There are a couple reasons why the average WP of a team immediately before its QB threw an interception was .41
A) The teams with the best qb's on average win more games. The alternative is that the average "worse" qb is trying to come from behind and since he isn't as good he throws more interceptions
2)Inflated extremes. A qb that is doing a 2 min drill but behind by 9+ points is gonna take a million chances. the win percentage average in a situation similar to this has got to be close to the 5%-0% range. Alternatively no team is gonna be throwing that late in a game with a 100%-95% chance WP. Low extremes inflate the importance of an already unimportant stat
Re: FO Mailbag: Dropped Interceptions
I can't speak for all QB's, but just dividing Aaron's numbers for the ten QB's above by number of attempts gives this list, sorted from worst to best:
Name Adj INT Attempts Rate
Favre 25 522 4.79%
Thigpen 20 420 4.76%
Bulger 20 440 4.55%
Delhomme 18 414 4.35%
Romo 19 450 4.22%
Rodgers 21 536 3.92%
Cutler 24 616 3.90%
Cassel 18 516 3.49%
Brees 20 635 3.15%
Warner 18 598 3.01%
Unsurprisingly, when you adjust for number of attempts, two QB's thought of as "good" last year (Brees, Warner) drop to the bottom of the list. Then you have Cassel, who was a good bit worse than these good QB's, then Cutler and Rodgers who were a good bit worse than Cassel, then a cluster of Romo, Delhomme, and Bulger (a trio of former rejects). Thigpen and Favre were the worst.
If you prefer to adjust for yards instead of attempts (to try to compare the ratio of something bad, like INT's, to something good, like yards) you get this list:
Name Ajd INT per 100 yards
Thigpen 0.77
Bulger 0.74
Favre 0.72
Romo 0.55
Delhomme 0.55
Cutler 0.53
Rodgers 0.52
Cassel 0.49
Brees 0.39
Warner 0.39
Inverting those numbers to get hundreds of yards per interception (or dropped interception) means that Thigpen, Bulger, and Favre were averaging an INT or dropped INT every 130 yards of passing offense or so. On the other hand, Warner and Brees were averaging about 255 yards between INT's or dropped INT's. All the other QB's were grouped in the middle.
This really makes 2008 Favre seem like the worst Pro-Bowl selection ever.
Re: FO Mailbag: Dropped Interceptions
Even worse when you consider Rivers wasn't selected to the Pro-Bowl while leading the league in passer rating (a stat all fans see), and Favre in instead.
Re: FO Mailbag: Dropped Interceptions
Deleted. Sorry, double post.
Re: FO Mailbag: Dropped Interceptions
The write-up on Eli Manning in the stats appendix for FOA states that Eli was "third in the league in dropped interceptions." The data above indicate that this is both (i) false and (ii) misleading. First, he was tied for 5th, not 3rd. Second, he was tied with 6 other QBs, which basically means he was tied for spots 5 through 11. Third, he isn't in the top 10 in ajusted interceptions or adjusted interception rate.
I love FOA and you guys do great work ... but in this case there appears to be some bias in the presentation of the stats.
Re: FO Mailbag: Dropped Interceptions
Sometimes what looks like bias is simply an honest error.
Re: FO Mailbag: Dropped Interceptions
Fair enough, though Eli has gotten more than his share of criticism on this site, which is why I thought bias (even sub-consciously). Either way, I'm glad this Extra Point debunks a myth I've never believed - that Eli is lucky in having an inordinate # of his potential INTs dropped.
Re: FO Mailbag: Dropped Interceptions
Aaron, leave aside (tied for) 5th versus 3rd. If I am remembering right (and I am half a continent away from last year's Prospectus so I cannot check my memory), Eli was near the top of the league in dropped interceptions the previous season as well.
Am I remembering right? How has he fared in 'dropped interception percentage' (dropped INTs / attempts) over his career? Is he normal or abnormal in this regard?
Just curious.
Re: FO Mailbag: Dropped Interceptions
This doesn't even add in the duck Favre threw up that turned into a TD though by all means it should have been picked.
Anyway, the next question I would ask is if QB dropped INTs have any year-to-year correlation? That is, is it a skill? For example, let's say because Orton is weak-armed, defenders are more likely to get to and hold on to his errant throws, whereas because Cutler has a gun, defenders have a harder time doing so. My guess as of right now is the "skill" is minimal. Even if my example is actually true, a rocket a defender might not handle from Cutler could still bound high in the air for anyone to catch. I'm saying much like fumble recovery, dropped INTs for QBs is largely just a luck factor.
Rodgers doesn't look as hot anymore. Go Bears.
Re: FO Mailbag: Dropped Interceptions
Which one? The one in week 2 to David Clowney I think it was? Because any throw for forty yards into the end zone on 4th-and-13 gets a free pass. You're actually better off there with an interception than with an incomplete pass.
I'm saying much like fumble recovery, dropped INTs for QBs is largely just a luck factor.
I think that's the point. This is a luck factor, and listing quarteback interceptions in gross as conventional statistics do should correct for it.
Re: FO Mailbag: Dropped Interceptions
I actually wouldn't be surprised if there were some correlation between certain QB's and dropped INT's. First, certain QB's (Favre) tend to throw more interceptable balls in the first place, so even if "drops" happened randomly some fraction of the time, those QB's would tend to have higher numbers of dropped INTs from year to year than the Peyton Mannings of the world.
However, I think the question you were asking is if dropped INT rate (dropped INT/adjusted INTs, where adjusted INTs = dropped INTs + INTs - intercepted hail mary's) is likely to correlate to particular QB's.
Again, I wouldn't be surprised if there were some correlation, for the reaso you mentioned, PROVIDED that you have a big enough sample size. And there's the rub. Interceptions happen so rarely, and a large number of them happen or don't happen because of dumb luck (i.e. the potential INT happened to be close to 2007 Brandon Merriweather versus Ed Reed, or the ball gets batted up into the air and falls randomly, or David Tyree happened to have gum stuck to his helmet :-). So any actual correlation is likely to get swamped in the noise. Also, I can see a counterargument to your "cannon arm" argument that would tend to offset it somewhat. Even a cannon-armed QB has to loft some passes in a nice parabola, any time he's throwing deep or trying to throw to a reciever who's gotten by his defender. Those passes are, by definition, going to be just as soft regardless of who throws them. On the other hand, the "bullet" passes are generally quick passes to guys in front of or beside their defender at short range. Cannon-armed QB's get them there so fast that I bet defenders rarely have time to react and get their hands on it, so maybe you wouldn't get too many dropped INTs from such passes even on errant throws. And weak armed QB's tend to not attempt those kind of throws...if the O.C. has any skill, he designs plays that don't require weak armed QB's to attempt many throws like that.
Re: FO Mailbag: Dropped Interceptions
Speaking of fumble recoveries being random, I've always wondered about defenses that swarm. When the Bears D was at its best this decade (Lovie stressed 11 men to the ball), it always seemed half the players were around the ball carrier, gang tackling. If a ball popped out, they obviously were more likely than not to come away with it. Seems like fumble recovery isn't a random, 50/50 proposition at that point.
Re: FO Mailbag: Dropped Interceptions
Lovie's first year as head coach of the Bears was 2004. When did he start teaching this? Was it in 2004 when they were a little above average on recovering fumbles, in 2005 when they were well below average at recovering fumbles, in 2006 when they were well above average at recovering fumbles, or in 2007 when they were average at recovering fumbles?
Re: FO Mailbag: Dropped Interceptions
Mostly I was thinking about the 2006 Super Bowl season and their D that year, which as you said was well above average.
Re: FO Mailbag: Dropped Interceptions
So Smith taught how to swarm and recover that year only?
Re: FO Mailbag: Dropped Interceptions
Is that a serious question or you just being a smarmy douche?
Of course it doesn't mean that's the only season he taught it. The lesson was the same, but many players were at their individual bests that year. Harris (both), Urlacher, Briggs, Brown (both), Vasher, Tillman, Wale, Tank all at their best. Injuries and/or decline have prevented them from being collectively that good other years. Plus 2006 had an offense that let them actually rest. In that perfect storm, they were able to get a group of guys to gang tackle with regularity at which point one would expect them to have an advantage when a ball did pop free.
2007 Brown's hurt, Vasher's hurt, Tommie's hurt, etc. Well, despite the same lesson being in place, everyone's not beating their individual matchup as regularly and getting to the ball as before, hence no gang tackle, hence not quite so good at the fumble recovery.
Perhaps the game charting should start tracking (maybe they already are) on all tackles the number of tacklers; it's not just solos and halves you know. You could also define some proximity to the ball carrier, like within 3, maybe 5, yards and capture how many defenders, even if not making the play are there anyway. This could have all kinds of implications. Does a certain defense make a lot of solo tackles? Do they routinely have 4 guys near a ball carrier? How does tackling distribution typically differ at the line, in the flat, in the secondary? On runs, on passes? And yes, if when a ball pops does the number of defenders "near" it change the likelihood of recovering it?
Re: FO Mailbag: Dropped Interceptions
Clearly you have never tried charting a game. It is impossible to get an accurate measure of details so granular with TV coverage.
Re: FO Mailbag: Dropped Interceptions
I'll second Vince's comment about the frequent impossibility of tracking tacklers with that level of granularity.
I'd also remind you that Aaron is trying to come up with things that explain (a) the entire NFL (b) over a span of multiple years. Pardon me if I don't buy "the 2005 team was tired" argument-they faced 11 more plays, over the course of a year, than the 2006 squad. For your story to be true, Lovie would really have had to start teaching it in the offseason of 2006, because the defensive players (who were about as good in 2006 in 2005, and not the injured/less bad unit of 2007) suddenly went from "really bad" to "quite good" at recovering fumbles. Can you find a story from the media, Chicago or elsewhere, that talks about Lovie teaching a different strategy that helped them recover fumbles much better? No Bear fan (and I've asked maybe a score) has proffered any evidence of such.
Oh, and if you want the results to be expandable beyond particular qualities of Lovie (and by swarming tackling, it sounds like you're talking about a teachable skill not inherently specific to Lovie), you also need to explain why the 2008 Lions were better at recovering forced fumbles than the Steelers, Ravens, Giants, Vikings, Titans, and Bucs.
I know you like the story you're telling, and you want to believe it's true, but it really sounds like a Just So Story.
Re: FO Mailbag: Dropped Interceptions
When Smith became the Bears coach reporters talked about two differences in drills he was running. The first was that every ball that touched the ground in training camp was treated as a fumble (and with Grossman taking most of the snaps there must have been a lot of practice). The second would be a drill that is designed to inmprove pursuit in the Tampa2 scheme where at the end of a play whenever the coaches blew a whistle the whole defense had to sprint to the corner of the endzone (the plays were run from the ten yard line). I can see how both drills would help fumble recovery but I doubt that they are specific to the Bears and as you point out they don't seem to have been all that effective. They are the only things I can think of that might fit with the kind of stuff you are talking about though and I am not sure there is much in it.
Re: FO Mailbag: Dropped Interceptions
In fairness to the Not Yet Verified Unverified Anonymous poster, if ever there was a defensive teacher who I thought could be said to defy the odds of "fumble luck", it was Lovie Smith. Not so much for what he did with the Bears, but for what he did with the 2001-2003 Rams. Watching those games, it did truly seem as if there were always more Rams defenders around the ball than seemed to be the case for other teams, and iirc his Rams always did post an impeccable record in terms of fumble recoveries. The 2003 Rams, in particular, stand out in my mind as perhaps the most swarming, opportunistic defense I have ever seen.
Speaking of the 2003 Rams, I can't let an opportunity pass without mentioning that they won my favorite NFL game ever- a 33-22 "shootout" with the Baltimore Ravens- a defensive shootout, if you will. St. Louis put up 33 points despite only posting 7 first downs and 121 yards of total offense, and despite turning the ball over 4 times. Marshall Faulk, at one point during the game, had two TDs... and negative rushing yardage. Of the game's other 3 TDs, two were scored by Adam Archuletta and Johnathon Ogden. Really just an incredible all-around game of football, and my counter to any claims that defensive football is no fun to watch.
Re: FO Mailbag: Dropped Interceptions
I agree that the Rams performance is a more compelling argument about Smith's tendencies as a coach. His teams have ended up among the very best in the league more often the would seem probable if there was simply a random distribution.
Re: FO Mailbag: Dropped Interceptions
My point was never that Lovie was a particular genius (though yes I do know of stories of him emphasizing this) or that others didn't teach having as many defenders near the ball as possible or that said "strategy" was anything profound. The point was when a team can do this (swarm, gang tackle) with regularity as I remember (yes remember, I didn't watch game film of every play) the better Bears defenses of the decade doing, it would only make sense they had an advantage in recovery if/when a ball hit the ground, since they had a numbers advantage in people around that ball. This would make fumble recoveries not 100% random. I know game charting isn't easy, but I was under the impression you got the 22 man angle, not the TV tape, not that it would make charting defenders near the ball that much easier. All I was saying if you could chart column A (number of defenders "near" the ball) versus column B (a binary 1 or 0 for recovery or not), maybe you'd get more than a random scatter plot.
There are lots of defenders who get knocked to the ground give little effort to get up and pursue "just in case." With Lovie (and look I have no particular love for him if you think I'm just trying to tell a glowing story), this was not tolerated. Guys played 100% to the whistle. You would think everyone on every team does that. They don't. Major leaguers don't sprint to first on every easy fly ball and not all NFL defenders give 100% every instant of every play, especially if they feel they are out of the play. In general, the lesson is never give up. Maybe a guy breaks a tackle and so you should be there in case. Maybe the ball pops loose and being there improves the odds of recovery. It's a Pop Warner lesson, for sure, but it doesn't mean all teams do it.
Look at the Lovie Rams as has been suggested. Look at the Ravens for most of the decade. I don't even know if they were good at recovery and that's not what I'm saying. I'm saying look for all teams that happened to gang tackle (whatever definition you come up with for that) and see if that has any correlation to fumble recovery. Ok, so the data's not available, but for as random as fumble recovery might be, I still don't think it's 100% random. Bad and lazy teams won't typically have as many guys around a ball, whereas good and hustling teams would. Maybe it makes a 5% difference in recovery ability, but maybe it's a 20% or more difference, but I have a very hard time believing it's precisely 0%. But I guess we don't have granularity to test the theory.
Re: FO Mailbag: Dropped Interceptions
I wonder how much tipped interceptions would do to a quarterbacks average. Tips would probably have to do somewhat with the accuracy of the quarterback but if you have a number of interceptions that are tipped up into the air for a defender to get then I wonder, well what ever that would mean?
Re: FO Mailbag: Dropped Interceptions
Is there a list of DBs who dropped these interceptions? Im very curious about that.
Re: FO Mailbag: Dropped Interceptions
There should be a table at the back of the book.
Re: FO Mailbag: Dropped Interceptions
See, this is such a slippery slope though. "Bad Decisions" could include throwing into double/triple coverage but your WR out-jumps them, or the throw is so under-thrown that the DB's overrun the play.... See Delhomme, Jake to Smith, Steven 2008-2009 season.
Re: FO Mailbag: Dropped Interceptions
On its own dropped interceptions makes for an interesting stat.
I don't think however that you can then legitimately combine that data with actual interceptions to give adjusted interceptions, unless you account for whether the two things happened on different drives.
In the real world, if player drops an interception and then the very next play the pass is intercepted your QB isn't liable for two interceptions; but with that stat he could be.
Re: FO Mailbag: Dropped Interceptions
This is true, but I think in practice the effect is relatively small--it has to be pretty rare for a QB to have an INT dropped and then immediately throw another one (or even throw one on the same drive). Based on the numbers above, an INT or dropped INT happen on average for every 150-250 yards of passing offense a team accrues. Since the field is only 100 yards long, this implies that, on average, a QB can't average multiple INT's or dropped INT's in the same drive. Furthermore, since an INT terminates a drive, the amount of passing yards between INT's is even larger than the average implies. So likely an INT or a dropped int usually happens no more frequently than once every three or four drives, at the most.
On top of that, a QB that almost throws an INT is probably going to be a little more careful immediately following it.
Re: FO Mailbag: Dropped Interceptions
In the real world, if player drops an interception and then the very next play the pass is intercepted your QB isn't liable for two interceptions; but with that stat he could be.
If my QB throws interceptions (real or otherwise) on consecutive plays, I want him to be penalized for it.
Re: FO Mailbag: Dropped Interceptions
Actually when Lovie Smith was hired in 2004 he stressed at his press conference that turnovers ( interceptions and fumbles) would be critical for his defenses sucess. During the Dick Jauron era the Bears were pretty pedestrian in forcing turnovers. In Lovie's first training camp, the Bears
defensive players were instructed to pick up any balls that hit the ground and return them for TD's. That was regardless of why the ball hit the ground. ( dropped passes,
incomplete passes etc ) It just got the Bears D in the mindset that any ball could be live and in play and that the defense could score at anytime. The first two years in particular, the Bears would often have one guy stand a running back or wide out up while another defender would make a concerted effort to rip the ball out of the offensive players hands. I have had several friends in other NFL cities comment on the Bears skills in doing just this.
Post new comment