Writers of Pro Football Prospectus 2008

Most Recent FO Features


» 2017 Defensive Personnel Analysis

Defenses have taken a wide variety of responses to the rise of 11 personnel. Is any one system better than another? And how has the rise of the "moneybacker" changed defensive philosophy?

16 Dec 2015

Film Room: Khalil Mack, Blake Bortles

by Cian Fahey

Sunday was Khalil Mack's day.

The fifth overall pick of the 2014 NFL draft has had a quiet career so far. Despite being selected so high, Mack has spent the majority of his career being overlooked in favor of others who debuted last season. Mack himself hasn't really been the reason for that. His sack numbers were low as a rookie, but he was still consistently playing at a very high level. Sacks are very important for pass rushers, but they shouldn't be used alone to measure a player's quality. Mack was consistently close to bringing quarterbacks down, but the combination of the lack of pass rush around him and the lack of coverage behind him meant that the edge rusher's numbers stayed low. His disruption in the pocket was still very valuable, even if his defense as a whole couldn't capitalize on that value. Mack was also a dominant player against the run, so there was nothing about his skill set or performances that suggested he should have been left in the shadows.

Still, Mack was overlooked because of how he entered the NFL and by whom he was selected. Even though he was a top-five pick, he was widely seen as the boring pick, the small school player who was extremely well rounded but not a physical freak like Jadeveon Clowney. The Raiders themselves didn't prop him up because they had been a losing franchise for so long, and the defense in particular was lacking any kind of star power.

So it came down to Sunday. Even with the Raiders contending for a playoff spot and Mack notching nine sacks in 12 games, the defender still wasn't receiving much national recognition. He had two sacks against the Titans in Week 12 and two sacks against the Chiefs in Week 13, but those displays weren't enough to stand out. Ultimately, to get the recognition he deserved, Mack would need to do something few before him ever have. Before Sunday, only 14 players in the history of the NFL . had five or more sacks in a game, The only player still in the league to do it, Aldon Smith, also plays for the Raiders, though he is currently suspended.

Mack joined this exclusive club by harassing Denver Broncos quarterback Brock Osweiler. He finished the game with seven tackles, five sacks, and one forced fumble that led to a safety. Considering how tight the game was, (the Raiders won 15-12), those plays were not only timely but also hugely valuable.

Whenever one player gets five sacks, something has to have gone wrong on the offensive side. The Broncos clearly didn't pay enough attention to Mack in pass protection, but they were also hampered by their lack of talent on the offensive line. The Broncos are particularly weak at both tackle spots, having lost Ryan Clady before the year and Ty Sambrailo during the season. The line as a whole wasn't great entering the season, so those injuries really hurt.

Mack's success on Sunday was a great example of how it's not just about players being good or bad, but also about how they match up. The Raiders defensive end was able to exploit the Broncos tackles because his greatest strengths intersected with their greatest weaknesses. Whenever you are comparing two players who are about to match up to each other, it's not just about the success that they've had all season long. Some tackles are better against speed rushers than bull rushers, so if they had faced 10 speed rushers before coming up against Mack they would look like a tougher matchup for him than they actually are. That is a hypothetical and a line of thinking that should be applied to every position.

For this specific matchup, Mack was exploiting the weakest spot of a player who hasn't played well all season.

(Click here if you are having trouble loading the image.)

It was clear early on the game that Michael Schofield was going to struggle against Mack. In the above GIF, Mack doesn't get to the quarterback because the ball is released too quickly, but the disparity between the two players' level of talent can be seen. Mack effortlessly advances downfield past the tackle's outside shoulder. Schofield can't drop far enough to establish a comfortable position between him and his quarterback, so he timidly reaches his hands out in the hopes of engaging the defender. The speed and power of Mack's hands can be seen as he easily swats away the tackle's hands and knocks him out of his way in one motion. From there, he has a clean route to the quarterback as Schofield spins around and falls to the floor.

That play came in the second quarter. Mack had a relatively quiet first half, with all five of his sacks coming in the second half. For four of those sacks, he beat Schofield. Three of those four were very similar.

(Click here if you are having trouble loading the image.)

(Click here if you are having trouble loading the image.)

(Click here if you are having trouble loading the image.)

Dating back to his days in college playing for Buffalo, Mack has always been at his best when he is using his power to overwhelm opponents. A lot of power rushers in the NFL have been rigid movers, players who don't possess great bursts of speed or fluidity to adjust their body positions while advancing downfield. Mack is one of the more fluid pass rushers in the league and his burst is impressive for the type of player he is. More often than not, he uses his fluidity and his burst to set up his power rather than bend around the edge. Schofield struggled to match Mack's initial speed in his drop throughout this game, but when Mack engaged him his greatest weakness was shown off.

Mack has quick, violent, and powerful hands. He uses them to consistently concentrate his power and punch his way through pass protection. Because Schofield couldn't match his speed with his hands, he was constantly in a losing position. With Mack being a hugely powerful bull rusher when he is aggressive and Schofield proving to be extremely weak against bull rushes, it was easy to see how he managed to get to the quarterback so often from this side of the field.

There aren't many edge rushers in the NFL today who can run through offensive tackles so easily. Defenses are becoming more and more about speed and complexity, so edge rushers are asked to play in space more than ever. Mack is so well-rounded that he can play in space and dominate in tight. He showed off his power against the Broncos left tackle also.

(Click here if you are having trouble loading the image.)

On Sunday, the Broncos left tackle was Ryan Harris. Harris has been in the NFL for a long time, bouncing around the league while filling various roles. The journeyman tackle has endured a litany of significant injuries over the course of his career. He was once a third-round draft pick who showed promise early in his career, but his play tailored off over the years. Harris doesn't have any standout traits, and Mack was able to take advantage of his lesser athleticism for his fourth sack. He widened the tackle's stance by aggressively advancing downfield before planting his foot to cut back across his face. Mack didn't look to continue through to Harris' inside shoulder, instead squaring off against the off-balance lineman to drive his hands through his chest.

Harris had no chance. His feet weren't perfectly set and Mack exploded with greater power, having already built up speed with which to attack him. Harris was so badly beaten that he toppled over backwards as Mack wrenched him to the ground with his left arm.

Mack wasn't celebrated for his versatility as a pass rusher when he came out in last year's draft. He's still not so versatile that he can whip around offensive tackles like Robert Quinn, but he doesn't need to be at that level to be so productive. He is extremely difficult to stop because of how he uses his hands, feet, and fluidity to set up his power. Most tackles will struggle to contain Mack because of those traits, but they also can't completely ignore his ability to come off the edge with his speed. For good measure, Mack showed off his speed for his final sack against the Broncos.

(Click here if you are having trouble loading the image.)

It's true that Mack feasted on inadequate starters and an offense that was too stubborn to give them help in this game, but his career as a whole to this point hasn't been engulfed in enough fanfare that overcompensating now shouldn't be considered unfair, even if the reasons for the praise are somewhat misguided.

Blake Bortles

Blake Bortles appears to be on the Matthew Stafford path.

The second-year quarterback has been extremely productive over the past two weeks, throwing for eight touchdowns and zero interceptions while averaging more than 8 yards per attempt. Bortles' most recent game against the Indianapolis Colts is dividing opinion though. He completed 16-of-30 passes for 250 yards, three touchdowns and no interceptions with one rushing touchdown and three fumbles. Neil Paine of FiveThirtyEight.com stoked the flames of the fire by writing that Bortles' performance had little to do with the Jaguars scoring 50-plus points against the Colts. Paine based his analysis on the quarterback's awful QBR, a 3.8 out of a possible 100. Pro Football Focus added to the criticism, giving Bortles a minus-5.4 grade as the Jaguars' worst performer on offense. Our own Vincent Verhei noted in Quick Reads that according to FO stats, Bortles was just average overall, dreadful in the first half but tremendous in the second.

Pete Prisco of CBS Sports, a noted film watcher, contended that the analytical measurements were way off on Bortles' display. Football Outsiders alumnus and current Sports Illustrated writer Doug Farrar reluctantly agreed with Prisco.

One of the main points of conflict comes from Bortles' 80-yard touchdown pass to Allen Hurns. Some believe the quarterback was lucky to complete the pass and that it could have easily been intercepted, and at least should have been broken up. Others suggest that Bortles' aggressiveness and arm strength allowed him to fit the ball in to the talented receiver.

(Click here if you are having trouble loading the image.)

Analysis of individuals in football is often grey rather than black or white. This play is a perfect example. Both sides can make strong arguments to support their points. From the negative side, Bortles took his eyes away from a wide-open out route to force the ball in to double-coverage. The deep safety played the ball terribly, as did the safety running with Hurns underneath. Bortles was fortunate that the ball found its way to the receiver. From the positive side, Bortles put the ball in a spot where his receiver could catch it, and the first open receiver wasn't past the first down line. Obviously, the positive side can also point to the result of this play. Hurns was able to catch the ball and bounce off the deep safety to run downfield for a huge touchdown, a critical score at that stage of the game.

Statistical analysis and film study are often divided into the objective and the subjective. The subjective is presented as a negative because it creates these scenarios where people disagree. Each analyst can only offer his opinion, but it highlights the nature of football where the result doesn't always reflect the quality of play on either side.

In terms of evaluating Bortles' play for this game specifically, this isn't the play to do it. Before making this throw, Bortles had endured one of the worst halves of football you're likely to see. He was better in the second half as he executed the offense more efficiently even if not spectacularly. There are two ways to look at this game. You can value the production of the offense as a whole and Bortles' steadier play in the second half, or you can focus on the more extreme plays in the first half. Considering just how bad those plays in the first half were, it's tough to look past them completely.

(Click here if you are having trouble loading the image.)

Bortles' first throw of the game gained 17 yards, but it should have been intercepted. The Jaguars used play-action that sent Bortles into the right flat with three receivers running routes to that side of the field. He correctly held onto the ball as all his options were covered, but didn't look to throw the ball away. Instead he held the ball to buy his receivers more time. He had the space to do it, so it wasn't a bad decision. It was a bad decision, though, when he attempted to fit the ball into a tight window downfield. His pass was way off target, which may have actually worked in his favor because a better throw would have been an easier catch for D'Qwell Jackson. Jackson dove beneath the ball and should have caught it, but instead batted it into the air for Clay Harbor to make the catch. This is a play where it's very difficult to praise the quarterback despite the result. He was extremely fortunate to get way with a bad decision and a terrible throw.

Despite Bortles' overall production this season, his accuracy and decision-making have been problematic throughout the year. He is an inconsistent player who flashes outlandishly good plays but also does the complete opposite. This is where the Matthew Stafford comparison comes into play. Stafford is a very talented quarterback, but his accuracy and decision-making have remained inconsistent throughout his career. Playing with Calvin Johnson and other standout receivers has helped him stay somewhat productive, but his overall contributions to his team's successes have been lower than they could have and should have been.

For Bortles to avoid becoming the next Stafford, he has to flip the ratio of positive and negative plays. In this game, the negative plays were too prominent.

(Click here if you are having trouble loading the image.)

At the end of the first drive, the Colts don't mask their intentions to blitz Bortles on third down. The blitz is successful, but it uses a stunt so Bortles has enough time to cleanly get to the top of his drop. From there, he stares down his slot receiver, Rashad Greene, to the left. That receiver is running an out route just past the first down marker. Bortles has predetermined that this is where the ball is going as a hot route, even though it's a route that needs some time to develop. The cornerback covering Greene is sitting on this route from the beginning. Bortles releases the ball with anticipation to get it out cleanly before the rush arrives, but his pass isn't accurate.

Bortles was fortunate that his pass wasn't accurate, because if it had been, the defensive back would have had a relatively simple interception. Instead, the defender can only reach one hand up to try and snag the ball out of the air while Greene has no chance of getting near the ball.

One of the main concerns with Bortles is his accuracy. He is completing just 57.5 percent of his passes, one of only six current starters in the league under 60 percent, despite playing with two outstanding starting receivers who have excellent ball skills. Of those five current starters, three entered the season as backups. Bortles' ball placement is a constant issue because of his inconsistent footwork and elongated throwing motion. That motion creates more problems than just accuracy though. To begin his second drive of this game, Bortles had two consecutive plays where his motion created turnover potential. The first was a fumble that one of the Jaguars offensive linemen recovered. Bortles had the ball knocked out from behind as he dropped the ball to bring it around his shoulder like a windmill. A quicker, more compact release would have allowed him to avoid that fumble. Bortles had three fumbles in the game, but one was't actually his fault.

(Click here if you are having trouble loading the image.)

The other came when he stared down a receiver who was never open against a blitz in the second quarter. The second play at the start of the second drive was the most concerning though. On this occasion, Bortles had a completely clean pocket as he looked to make a quick throw. He had predetermined his throw before he ever touched the ball.

D'Qwell Jackson was sitting in zone coverage, reading Bortles' eyes from the beginning of the play. Bortles' eyes led Jackson to the football, and his elongated motion gave Jackson more time to get in position. On a quick throw like this, even the slightest unnecessary movement can make a huge difference. Jackson once again showed off poor hands. He should have caught a pass that arrived quickly, but just above his shoulder. A linebacker with better ball skills would not have even had to leave his feet to make the reception.

If Jackson had made that play, he would have had an easy touchdown going the other way.

This was Bortles' worst play of the day, but his first half was littered with similarly reckless, inaccurate actions that were largely left unpunished by a defense that lacks talent. The Colts are an average defense by DVOA and a below-average pass defense, but they have had a tendency of not taking advantage of opportunities during the Chuck Pagano era. This game was another great example of that. For Bortles, these plays didn't have a significant impact on his statistical output or the Jaguars' production, but to ignore them completely wouldn't be rational in evaluating the young quarterback.

Bortles has a lot of physical talent and he could develop into a good starting quarterback one day, but at this point his touchdown-to-interception ratio is flattering him. Even if you just look at his statistical output, he is still completing passes at a lower rate than the majority of the league (seven of his 13 starts have seen him complete 56 percent or fewer of his passes). His yards per attempt is 7.0, lower than 25 quarterbacks who have thrown at least 100 passes, and he is tied for the third-most interceptions in the league, 13, with Matthew Stafford. Whether it's raw numbers, analytics, or his film, there are as many reasons to be concerned about Blake Bortles as there are to be optimistic.

Posted by: Cian Fahey on 16 Dec 2015

29 comments, Last at 19 Dec 2015, 2:41am by Duff Soviet Union


by RickD :: Wed, 12/16/2015 - 3:06pm

One of the biggest things I dislike about QBR is that it is not, strictly speaking, a statistic. If a person has to argue about why a certain play was judged in a negative way for a given player, we've moved from the land of objective measurement to that of subjective judgment. And yes, there is not only a place for a subjective judgment, it is necessary to have a complete picture of any kind of situation. But it muddies the waters to pretend that a subjective judgment is an objective statistic.

by Fierydemise :: Wed, 12/16/2015 - 3:45pm

Unlike say PFF grading, which is really just judgement using numbers to appear more objective, QBR seems pretty solidly like a statistic. As I understand it under QBR Bortles gets less credit for that throw's yardage because so much of it was YAC. This can be objectively determined purely from game stats. Now we can argue about whether he should get credit for that play at all given the quality of throw but my understanding is QBR doesn't incorporate those kind of judgement calls.

by Eleutheria :: Wed, 12/16/2015 - 4:24pm

And this is where subjective analysis really falls apart in my opinion.

If Aaron Rodgers or Tom Brady made that throw, people would gawk at the play. They'd point out how tight the window was and claim he threw the ball perfectly between two defenders where only his receiver could get it. Clearly this was a perfect throw that exemplifies the elite quarterbacks hall of fame career.

But it wasn't a future hall of famer making the throw, it was Bortles, a QB who has a reputation for inaccurate throws. Clearly then, it was a poor decision to throw the ball into double coverage, and this is yet another example of Bortles making a bad read, and he was lucky to not be intercepted on the play.

The problem with subjective analysis is it's subjective and our evaluations are tainted too much by our biases and past expectations.

And I think QBR gives Bortles little-to-no credit for that 80-yard play. ESPN claims it uses EPA evaluate the success of plays, Bortles had an EPA of 10.02 on the game, which is pretty good, but ignoring that play, his EPA 3.30, which is below the league average QB of 4.61.

by Fierydemise :: Wed, 12/16/2015 - 5:51pm

But this isn't a subjective analysis at all. QBR has a discount for YAC because YAC doesn't appear particularly predictive. Brady has a low (for him) QBR this year partially because of his poor YAC, as was pointed out last week:

A global reduction on credit for YAC, or giving a player YAC based on historical average is the kind of exact, repeatable computation we want in our metrics.

by RickD :: Wed, 12/16/2015 - 6:06pm

It's silly to say that Brady should be dinged for YAC. Brady consistently gets good YAC from his receivers, no matter who they are. That's because his passes are accurate and well-placed. I don't care if YAC isn't "predictive" for the mean QB.

The "clutch weighting" nonsense is also silly.

by theslothook :: Wed, 12/16/2015 - 6:08pm

Actually, the article showed Brady specifically saw a dip in his YAC numbers when his receivers were hurt. This was also true in 2013 as well. If QB play really was a significant factor in YAC, it would show up in the data. According to the stuides that have looked into this, it has not.

by Hoodie_Sleeves :: Thu, 12/17/2015 - 12:44pm

Even if YAC was 100% dependent on the quarterback, you'd expect a dip when guys get hurt, for a couple of reasons:

YAC is very much affected by timing - if there's new WRs, the timing is going to be different.

New receivers are most likely to be less talented than the previous ones, which means they'll be less open, which minimizes the QB's chances to pick the best option.

Most of these studies are very poorly designed - they have the same sort of issues that studies on pitcher control of BABIP have - they assume that for a talent to exist, it must be normally distributed at various strength throughout the population - and this is hardly ever the case.

The other problem is that YAC is a bit like Yards Per Carry when looking at RB - it doesn't tell you a whole lot by itself.

On a 15 yard throw where there's a broken coverage or a DB falls down, the receiver may end up with 60 YAC simply because he's fast enough that he can't be run down by someone who he has a 10 yard head start on - it doesn't tell you anything about the QB or the receiver.

On the other side, on something like a slant, or a short crossing pattern, if the QB puts the ball where the receiver can catch it in stride, you may get 5-10 yards of YAC, whereas if he puts it on the backside, the WR gets tackled immediately, with no YAC. That's almost ALL on the QB.

Another type of play - like WR screens, or the little quick passes the Pats run where Brady basically stands up, and throws the ball parallel to the LOS to a WR, and that guy is expected to make the first defender miss and pick up 5 yards, those are all WR talent.

Lumping all these things together in one stat doesn't tell us anything - because they're incongruous - which is why the studies basically show noise.

by Eleutheria :: Thu, 12/17/2015 - 1:46pm

But there does come a point in a play, such as when the receiver dukes out three defenders on his way to the endzone, that's it clear the quarterback had no control over the final outcome of the play.

I agree ESPN is wrong for making the QB have 0 credit for what happens after-catch, but I do think the QB should get more credit for what happens before the catch then what happens after.

by Hoodie_Sleeves :: Thu, 12/17/2015 - 2:00pm

Of course.

Honestly, I think the YAC stats would be a lot more interesting looked at in buckets - I'd bet if you look at Brady his YAC (compared to the rest of the league) in short passing is fantastic, good in medium passing (especially over the middle) and terrible in long stuff.

Patriots long balls - the ones that are actually caught - seem to almost always end with the receiver either coming back for the ball and getting killed, or making a stumbling/diving catch with little YAC (and that's because Brady just isn't that accurate at range).

The short crossing stuff though - he's better than anyone at timing things - and looking off defenders to open windows on these plays. The entire Patriots offense is largely built on being able to turn 3 yard gains into 5-7 yard gains.

by theslothook :: Thu, 12/17/2015 - 4:39pm

I feel like somehow, by arguing yac is (mostly) a function of the receiver, its a subtle way for us to dig at Brady. I'm not wanting to get into that can of worms - I just think YAC is mostly a function of the depth of target and receiver - that once you control for those factors, the qb play becomes less significant.

I read the bryan burke piece. There was endogeneity problems with his regressions which could lead to a biased estimator, but generally speaking, even in that scenario, your first instinct should be to take a panoramic view of the results. At that point, the right takeaway isn't to say qb play is not important or not significant, but that its relatively speaking probably less important than other factors listed. Does that mean Brady is hurt by qbr unfairly? Probably some, but that doesn't mean the stat itself is garbage. It probably means its overcorrecting for the Yac effect, but that doesn't mean there shouldn't be a correction.

For instance, DVOA corrects for opponents, but probably overstates it(especially when opponent quality itself is evolving over time. Facing the panthers in week 1 is not the same as facing the panthers today). We still like that DVOA has opponent adjustments in place.

To that end - is it really a surprise that BRady and the offense are infinitely more successful with better receivers? Hasn't that been shown with Peyton manning and Aaron Rodgers and Brett favre..etc etc? One doesn't have to take the position that Yac is mostly a wr thing as an indictment that the qb is riding the cotails of the receivers.

by Fierydemise :: Wed, 12/16/2015 - 6:26pm

Wow I totally botched that comment but it looks like you got what I meant to say.

I agree the clutch weighting is pretty silly but a YAC discount seems quite logical.
From the ESPN article, "it's worth nothing that Brady's receivers were close to neutral in each of the previous two seasons (a total of 18 YAC+ in 2013 and 2014)" so it doesn't appear that Brady is consistently good at getting YAC from his receivers.

by Eleutheria :: Wed, 12/16/2015 - 8:29pm

Perhaps you misunderstood my argument.

I support breaking a pass down into before and after the catch, giving more credit to the pre-catch (though based on what I've read QBR doesn't give enough credit for the YAC).

My problem is ESPN claims it incorporates stuff like whether or not the receiver was hit in stride.

And that's what I'm complaining about. ESPN claims QBR is both simultaneously an objective statistical approach to evaluating QBs and is also derived from an analytical study of game film. It can't be both.

Either it's an analytical study of game film making it at least partially subjective like PFF grade. Or it's a completely objective evaluation of play-by-play data like DVOA. It can't be both.

There are benefits and drawbacks to both, but ESPN can't claim it carries the advantage of studying game film without also acknowledging that it carries the largest drawback of doing so: loss of objectivity.

And an objective study can incorporate measurements derived from game film. So long as you're measuring them objectively:
You can adjust for the time the offensive line gives the QB to throw, so long as you have a clearly defined objective way to measure pressure in the pocket. eg. PFF has a time-to-pressure stat, but I can't find an article defining the stat, so it's not clear how objective it is.

You can account for how accurate the throw is, so long as you use a clearly defined (most likely binary until we have better tracking data) measure of whether the incompletion was off-target or dropped (and even then, you should penalize the QB for a drop, just penalize them less then a throw that's way off target). eg. Sporting charts defines a dropped pass as a pass where the ball bounces off a part of the receiver. It therefore includes some clearly uncatchable throws while excluding some clearly catchable ones, but at least it's 100% objective.

Maybe you want to account for the defensive formation being used (eg. blitz vs non-blitz, cover-2 vs nickle/dime vs stacked box, etc.), so long as you're clearly defining types of formation and aren't taking into account subjective points like how open or covered the target receiver is.

The key here is too only include objective data. ESPN doesn't appear to do so, but I could be mistaken.

by Eleutheria :: Wed, 12/16/2015 - 3:45pm

I agree that QBR is too subjective for it to be considered an advanced statistic (as is PFF Grade).

I'm not against breaking down passes into various components an assigning QBs a different amount of credit for each of them, so long as you keep it objective.

So if you really wanted to include drops, keep it objective. Sporting charts has a relatively good definition of a dropped passes:
If the ball bounces off the receiver it's a drop. If it doesn't its a missed throw.

This is far from a perfect measure, it will include some clearly uncatachable balls (like when the ball is thrown behind the receiver and hits his back shoulder), and will exclude some catchable balls, but at least it's objective.

If QBR does something like that for drops, I'd be relatively ok with it (so long as a QB still gets a large deal of blame for drops, just not as much as other incompletions). If it instead makes people analyse the throw and subjectively determine whether it was catchable (like ESPN claims it does), I'd be against that.

And maybe I'd be less against QBR if they actually published how they calculate it, since it might be more objective then I claim, but at the moment it seems too subjective.

by deus01 :: Wed, 12/16/2015 - 3:54pm

If you really want to isolate the impact from the quarterback you're probably going to need some subjective measure. E.g. what if a receiver manages to just barely get a finger tip on it. Everyone would probably agree that it should be classified as an overthrow. A similar thing could be done for balls where the receiver makes a great play to turn a bad throw into a catch.

Maybe as the next gen stats get more advanced you would be able to determine if a throw was within a certain radius of a receiver in order to determine if it's catchable but I think we're still a long way from that.

QBR is a bit too subjective but what it's trying to do, separate the impact from the QB from all the other components of the team/game, is interesting and I think worthwhile.

by Eleutheria :: Wed, 12/16/2015 - 4:06pm

As I said, the objective definition of a catchable ball used by sporting charts isn't perfect, but at least it's objective.

In my opinion it's far better to use clearly objective definitions and measurements, rather then subjectively assign credit based on what the film makes it look like.

You're not going to get a 100% perfect measurement, I don't think me or any other supporter of statistical-focused analysis thinks that.

But we do think a good objective statistics-based approach is going to better then a subjective measurement.

by deus01 :: Wed, 12/16/2015 - 4:15pm

Sure. I agree that something that's purely objective is going to be better and maybe having a binary indicator as to whether the ball hit the receiver would be enough.

I think the subjective components of QBR are what make it more important for the underlying calculation to be opened compared to something like DYAR (though I would also like this to be open). At least with DYAR you can look at the play by play to see how it was determined. With QBR you need additional information that isn't available like the subjective grading of plays.

by Tomlin_Is_Infallible :: Wed, 12/16/2015 - 3:32pm

That looks like a terrible play/route design in the Bortles gif with the stack set left. A line call to short comeback or even a curl or sit down would be an easy completion to the trail receiver.

The standard is the standard!

by jtr :: Wed, 12/16/2015 - 4:28pm

Even that speed rush on the last sack by Mack was set up by his power. It looks like the tackle leans forward to try to brace himself when Mack gets even with him. The tackle was already late getting in front of Mack, and he puts himself off balance as well because he's expecting Mack to smack the crap out of him like he has so many times already. We're used to modern edge rushers using their speed to set up their other moves, it's fun to watch a guy whose default move is to knock the hell out of the opposing tackle and everything else works off of that.

by Jmf02f :: Wed, 12/16/2015 - 5:36pm

So the first part of the article is almost a love letter to Khalil Mack, where the Bortles portion is essentially a highlight of everything he did wrong with no analysis of his good plays, with a caveat that he MAY become good. But his stats at this point are mere chance-accumulated over 13 weeks. Also you point out that his comp % is on par with two other qbs who entered as back ups--- but you fail to mention that one of the other QBs with so low a completion percentage is you other crush Cam Newton. I guess mentioning that would be counter prodictive to the narrative you are pushing. Ok. So essentially Cian called his shot that Bridgewater is the best and Bortles sucks and any evidence to the contrary will be explained away as either bad defense or solely the product of his wide receivers. Got it.

by ChicagoRaider :: Wed, 12/16/2015 - 5:45pm

There are some competitions where it only matters how the best do against the best. But sometimes, it is just the job of the best guys to wipe out as many of the average guys on the other side as possible. It looks like Mack is a long way to contributing the second way. There are a lot of guys with weaknesses at the tackle positions. As he builds his skills sets, he will get scarier against more guys. I hope he is a film maven. Developing a book on how to beat each tackle in the league and what skills to develop to get the ones he cannot beat yet would look like a plan for someone with his versatility.

by ChicagoRaider :: Wed, 12/16/2015 - 5:45pm

There are some competitions where it only matters how the best do against the best. But sometimes, it is just the job of the best guys to wipe out as many of the average guys on the other side as possible. It looks like Mack is a long way to contributing the second way. There are a lot of guys with weaknesses at the tackle positions. As he builds his skills sets, he will get scarier against more guys. I hope he is a film maven. Developing a book on how to beat each tackle in the league and what skills to develop to get the ones he cannot beat yet would look like a plan for someone with his versatility.

by Perfundle :: Wed, 12/16/2015 - 6:06pm

But his stats at this point are mere chance-accumulated over 13 weeks.

In 2010, Matt Cassel had 27 TDs to 7 interceptions. Josh Freeman had 25 TDs to 6 interceptions. You could just look at the gaudy numbers, or you could watch the actual play and see the deficiencies hidden beneath.

but you fail to mention that one of the other QBs with so low a completion percentage is you other crush Cam Newton.

Cam Newton has a sparkling 7.8 yards per attempt to go with his low completion percentage. Blake Bortles has a low 7.0 YPA, and has receivers that got very high grades from PFF. Glad you got it.

by Jmf02f :: Fri, 12/18/2015 - 2:34am

So youre using another stat to prop up a bad stat, while also arguing that gaudy stats should be ignored. Well thats sound logic. Also calling 7.8 sparkling and 7.0 low tells me youre a mental midget that should be euthanized or in the very least sterilized. And i didnt say there were not deficiencies. Bortles has plenty. But if someone is going to write analysis and not merely a hit piece (as this is) then there should be analysis of his good plays as well. When one blatantly ignores any good traits in their analysis of a player and merely highlights the bad you have to question the motivation. Feel free to lick Cians candy-ass and defend him like youre his bottom bitch, a position you are probably very familiar with. However, i dont merely defer to his opinion because he has an all 22 subscription and holds himself out as an expert- despite no stated credentials for what qualifies him as an expert. I wouldve been more civil had you not been snarky in your comment to me, but since you were i feel no need to be polite to a bitch.

by theslothook :: Fri, 12/18/2015 - 12:06pm

You know, i actually agree with your premise, but that(hopefully alcohol fueled) diatribe is a good reason why this site needs a filter. Just cuz someone is snarky doesnt mean you have to reply like a cussing teenager.

by Duff Soviet Union :: Sat, 12/19/2015 - 2:41am

You seem reasonable, and obviously have no anger issues whatsoever.

by hrudey :: Thu, 12/17/2015 - 12:17pm

The easiest way to put some of the comparison with Bridgewater and Bortles into perspective is that Bortles has thrown 118 more passes than Bridgewater this year overall, but only 18 more passes 10 yards or less down the field (at least using ESPN's splits). Bortles has thrown 41.7% of his passes more than 10 yards downfield, while Bridgewater's at 28.5%.

Now, as a Jaguars fan and therefore an expert on god-awful QB play, I can say that he does have a point that at times, Bortles is inaccurate and is usually good for a few head-scratchers a game. On the other hand, he's also throwing the ball down the field a lot, and often extending plays to do so - which, if you look back at the play above where he's inaccurate high on a pass to the left, you see he's releasing it just before getting clobbered.

Fortunately (at least for me as a fan), I am not stuck with Bortles of 2014 or Bortles of the first half against Indy, and instead am forced to suffer with a QB who's already blown away our franchise record for TDs by seven, with three weeks to go in the season. Sure, we'd all rather have Captain Checkdown methodically plodding the team down the field to attempt another field goal (have you met Jason Myers?). We can't all be that lucky, though, and having the guy who's third in the league in TD passes is at least some small consolation while we weep and gnash our teeth about the sub-60 percent completion percentage.

Okay, the sarcasm got laid a little thick. The point is the guy is a work in progress (wasn't even supposed to play last season, re-did his mechanics this offseason), who is actually making progress. The stuff he's being nitpicked for now? That was 80% of the time last season, now it's maybe 25%. He's third in the league in INTs and third in the league in TD passes, and both are what he is - maybe a future article might actually look at how a player with such "serious holes in his game" can possibly have thrown for more TDs this season than anyone not named Tom Brady or Carson Palmer. Maybe call it "Any 13 Given Sundays: Blake Bortles somehow finds his own guy in the End Zone"?

by Tomlin_Is_Infallible :: Thu, 12/17/2015 - 1:12pm

"and often extending plays to do so - which, if you look back at the play above where he's inaccurate high on a pass to the left, you see he's releasing it just before getting clobbered. "

the thing is, like I posted above, the offense doesn't even need to make it that difficult for him. That is a terrible route combo out of that set vs that D alignment. I doubt he audibled into it, so blame the coaches for not putting him in the best position to succeed.

The standard is the standard!

by t.d. :: Fri, 12/18/2015 - 7:26pm

They doubled down on Carr being a mediocre prospect compared to Bridgewater in their early season breakdown, too. When they form an opinion prior to the draft, they have a hard time letting it go, and they've been skeptical of both guys all along (Bridgewater's fine, in my opinion, but the jury's still out, and while the Minnesota line sucks, the Jags' line matches it)

by hrudey :: Thu, 12/17/2015 - 12:32pm

"You can value the production of the offense as a whole and Bortles' steadier play in the second half, or you can focus on the more extreme plays in the first half. Considering just how bad those plays in the first half were, it's tough to look past them completely"

But it's apparently not tough to look past the fact that every single time the offense had the ball in the second half, they scored a TD. Every. Single. Drive. But that's dismissed as merely "steadier play" as if the baseline expectation of a second year QB is offensive success on every drive.