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.
20 Sep 2010
The Blindside Project
Ben Alamar, who helps with the season simulation that you see in each year's Football Outsiders Almanac, will be tracking the time in the pocket for quarterbacks on every play in Jets and Giants games this season. He'll also be looking at which linemen lose their blocks, and how often that leads to a sack. The goal? An attempt to figure out the value of individual linemen.
20 comments, Last at 22 Sep 2010, 3:59pm by Ben Alamar
Most Recent FO Features
» 2011 Defeats
Who had the most Defeats in 2012? Well, nearly setting the all-time sack record puts you in a pretty good position to claim that crown.
Comments
More data, please.
A noble venture. Keep us posted.
Re: The Blindside Project
Yeah, this the Holy Grail of pass blocking evaluation, and FO should be congratulated on taking up this immense task, if only for two teams. If I wasn't too damned lazy, I'd offer to do the same for the Vikings.
Re: The Blindside Project
This is super cool. Ben, my hat is off to you.
I am really curious, already, as to how Eli's line did last night. Collinsworth's counting aside, it seemed that he had no time at all. Pretty sure Eli would not be surprised if Freeney popped out of his underwear drawer this morning, he was riding his ass so much last night. "Man, these are nice and fresh! You use fabric softener on these?"
Come to think of it, how do you measure the terminal "tick" as it were--when he throws? Or if he's rushed into a crappy throw, or a throw-away at 2.5 seconds, is that your count, or what the rusher would have reached him....? Clearly if there is a sack or QB hit, the time the pass rusher makes contact seems reasonable. But if there's not a definitve hit, or if (like Brady on a coule throws yesterday) he stands flat-footed for 5-6 seconds, when do you stop counting? When his pizza is finally delivered?
An awesome task that I hope yields some useful data.
Re: The Blindside Project
There are so many different ways to define what the time in the pocket is, but for my purposes (evaluating offensive linemen) we start the clock on the snap and stop the clock when one of the offensive linemen (or other blockers) fails, ie: the clock stops when the pocket is no longer a safe haven for the QB.
Re: The Blindside Project
Thanks. That's logical, but a pretty subjective sounding measure. As long as you (and Keith) are the only ones doing it, and are consistent game to game, it works for me. (You can call it the CYP threshold--the pocket looks good, pocket looks okay, still holding up... oh damn! At what point do you crap your pants with concern for the QB? That's the CYP line.)
Re: The Blindside Project
Sounds like there's lots of different variables to track. Obviously you're going to want to track number of pass rushers and number of pass blockers. And I suppose you need to decide how to count backs and ends who provide quick blocks or chip help before slipping out as receivers. I imagine also you need to code for screens and play action, the latter of which could be tricky. I remember seeing Manning gesture a desultory kind of play action at the start of nearly every play.
Re: The Blindside Project
One answer to a question Ben brings up in his piece can be found in the quality and depth of the QBs in the NFL. Okay, so some teams might pay more for their LT than their QB... why? Because their QB is replacement level, but the best they have right now, but their LT is the foundation for the future and inplace to protect that franchise QB, or lure him there via FA, when the occasion arises. Not many teams have QBs who are far and away the most valuable player--maybe ten? Maybe fewer. (The Colts had young bookend tackles BEFORE Manning was drafted.)
The Brady/Cassell year of 2008 suggests that the blind side theory is irrelevant, that a replacement player can come in and win 11 games when a future HOFer goes down. (Obviously you need a very solid supporting cast and staff.) Fans of the Colts, who pee their pants at the very mention of Curtis Painter, would beg to differ on that point--we see a 10-12 game dropoff if #18 goes down. Oh we know our LTs are mediocre at best, but replacing the big guy is not an option.
Peyton Manning, in his own way, is doing everything he can to mootify (a new verb) the blind side theory, or the theory that competent OL play is required at all.
Those two very specific circumstances aside, my money is on the blind side theory holding water. But the measurables probably go beyond sacks--a QB who has complete faith in his BS protector has one less thing to worry about, can relax more and focus more downfield--in short, he performs better.
Good luck with this--a very worth project.
Re: The Blindside Project
Bob, I think the Saints pretty much have the same circumstances at the tackle position--great QB, no-namers at T (although RT Stinchcomb finally made the Pro-Bowl). However, at least the G's are elite--and C Goodwin seems at least competent.
Re: The Blindside Project
From what little I've seen Curtis Painter is significantly worse than Cassell, and that's saying something.
He must be the greatest clip board holder in history to have a job in the NFL.
Re: The Blindside Project
I wouldn't be too quick to dismiss Painter. Aaron Rodgers looked horrible his first two years. I doubt Cassell would have looked half as good had he been needed as a earlier. Even most players who started in their first two years (P Manning, Favre, Brady) had some growing pains. Not saying Painter will be great. But he could be at least replacement level ala Sorgi.
Re: The Blindside Project
I really doubt it. Even in college he never looked that great to me.
Re: The Blindside Project
"These results suggest that there is a significant difference between failures and sack totals, but does that translate into an effect on completion percentage? There the data appears to suggest that no, failure is failure when it comes to completing passes, so the higher salaries may be attributable solely to the insurance aspect: we don't want our QB getting hit too much."
Why use completion percentage, rather than something like NY/A or even DVOA? Completion percentage is not a great measure of passing game performance.
Also, another part of blindside theory, I believe, is that a quarterback is more likely to fumble on a sack where he does not know he is going to be hit. The difference in negative offensive value between a sack and a strip-sack is really quite large. I appreciate that there could be sample size issues here, given the comparative rarity of fumbles, but it seems to me that this would also be an important thing to track.
Finally, a player's compensation is determined not only by his value to his team in the year in question but also by his expected value to the team in other years of the contract (when they may hope to have a better quarterback) and his hypothetical value to other teams at the point when his contract is negotiated (or by his draft position, if he is playing on his rookie contract - and truly elite left tackles, more than any other position, are found almost exclusively at the very top of the draft).
Re: The Blindside Project
Mr Shush,
yes on the future-looking thing. Say your team has a warmed up corpse at RT and an inanimate carbon rod at LT but plenty of money to spend on a crop of FA QBs whose contracts roll this year.... which team is more likely to sign the most in-demand QB? Yours, or the team with a stud LT? If that's the only difference (not money, not O coordinators, not WRs) it seems to me the smart QB goes with the solid LT. Playoffs, fame, glory ensue.
Re: The Blindside Project
Not too often a stud QB hits free agency, but it happens occasionally, and perhaps just as importantly, if you draft a promising rookie, your already in-place quality left tackle greatly improves his chances of not getting killed or irrevocably learning to feel phantom pressure while he's finding his feet in the pros.
Re: The Blindside Project
Good luck with this. Seems like a massive undertaking.
Re: The Blindside Project
Good luck Ben!
Re: The Blindside Project
Damn, what I wanted for years.
Something along time in the pocket x (Nb of Defenders rushing the passer/Nb of Offensive players protecting the passer) but more clever...
Re: The Blindside Project
Good luck! I'll be keeping this in mind throughout the season.
Re: The Blindside Project
Is this large and worthy undertaking in any way Collinsworth-inspired?
Re: The Blindside Project
Just added some preliminary results from the original data set. Current data is coming in now so should be posting some results this weekend at www.analyticfootball.com
Post new comment