Analysis Notebook: Week 2

Sam Monson takes a look at Ryan Tannehill and explains why his PFF grade is so much higher than his raw statistics suggest it should be.

| 3 years ago

Analysis Notebook: Week 2

analysis notebook copySo just what the hell is with PFF’s quarterback rankings? Ryan Tannehill is No. 3 despite a passer rating in the 70’s while Peyton Manning (with a league-leading 126.9 passer rating, six touchdowns and no interceptions) is down in 15th. Clearly PFF has been sniffing glue, right?

Well, not exactly. The PFF grades and thus the rankings aren’t a metric. They’re not some complex formula taking into account fifteen various statistics but rather a play by play grading of that player. The grades are not founded on or dependent upon statistics, and that is because stats – especially for quarterbacks – lie all the time.

Let’s talk hypotheticals for a moment. Three quarterbacks can each complete a pass for 50 yards and a touchdown. In the box score each guy will read 1-for-1 for 50 yards, 1 TD and 0 INTs.

Player A looks off the safety, splits him and the corner and lands the ball in his receiver’s hands at the 5-yard line without the need to break stride. It’s a picture perfect pass, and to make it even better he did it with pressure steaming towards him.

Player B hits his best playmaker across the middle 5 yards downfield against soft coverage on 3rd-and-long. That athletic receiver breaks a tackle, jukes the safety out of his shoes and outruns the defense the rest of the way for six.

Player C misreads the coverage entirely, throws it straight to a linebacker who displays the kind of hands that got him put on defense in the first place, deflects the ball over his head where it lands in the hands of an uncovered receiver who has a quick sprint to the end zone for the score.

Remember – each guy has exactly the same statistics from the throw, but those passes span the spectrum from NFL Films highlight reel material to something that would show up on ESPN’s C’Mon Man. One is a fantastic throw, the other is a dumpoff that a receiver did the heavy lifting on, and the final throw is a terrible play that got lucky.

Passer rating, passer yards, touchdowns, interceptions – they are all, in essence, measures of how the passing offense as a whole did, not the quarterback. PFF’s grades are different. We grade throws, runs, plays the quarterbacks makes regardless of the outcome. If a quarterback fires a perfect pass that hits his receiver in the hands for what should be a first down, but that receiver drops it – we give the quarterback credit for that throw – statistics don’t.

Tannehill has suffered from seven dropped passes already this season (only Joe Flacco has experienced more). He has lost 113 yards of passes due to drops (at a minimum – that only counts the yards in the air the ball traveled, not the potential yards after the catch the receiver lost by dropping the ball). That is more than any other passer in the league, including Flacco. If you add those plays into his official stats his passer rating jumps twenty points into the 90s. Sure, you could do that for every quarterback, but as we have already pointed out, Tanneill has suffered more than most from his receivers not helping him out.

Let’s take a look at one such great play from Tannehill that went unrewarded by statistics but not by PFF’s grading.

New England @ Miami | Q4, 5:51

The Dolphins won this game comfortably in the end, but with 5:51 remaining in the game Miami was up by just three. With a first down at New England’s 28-yard line they decided it was time for a shot into the end zone, and they had the perfect play drawn up for it.


After a play action fake TE Dion Sims had been able to work his way across traffic and then turn up field towards the end zone. He had Tannehill2athletic linebacker Jamie Collins trailing him, but there was space there to make the throw. Tannehill let fly from the 38-yard line and dropped the pass perfectly into the waiting hands of Sims, who couldn’t bring it in before being taken to ground by Collins.

This was a touchdown pass, and a fantastic throw, that statistics don’t give Tannehill credit for. This pass alone would push Tannehill’s passer rating for the season into the 80s and make his raw statistics look healthier.

How about the deep pass to Mike Wallace against Darrelle Revis that Wallace couldn’t get both feet in before stepping out of bounds? At this point in the game the Dolphins were actually trailing by 10 points and Tannehill flicked the ball almost 50-yards in the air, over a trailing Revis and into the back of the end zone.

2014-09-17 11_21_50

It’s important to make the point, however, that we’re not trying to say Tannehill has been fantastic. Tannehill has also had some very bad passes that have received heavy downgrades. Also against the Patriots he had a chance to hit another touchdown and instead threw an interception, handing the ball to the Patriots to begin a drive they later scored from.

New England @ Miami | Q2, 13:06

After a play-action fake Tannehill looked to his right, but Revis had blanketed his intended receiver, Brian Hartline, on the play up the right sideline. Tannehill came to his second read and saw Mike Wallace break free on a post pattern because the safety had abandoned the middle of the field while reading his eyes initially.


This is a relatively routine throw. As you can see, he has a large area of real estate into which he can put the ball and allow Wallace to run away from the coverage, but instead he underthrows it badly, allowing the corner Alfonzo Dennard to intercept it unchallenged.

There’s no denying this is a terrible pass, even with late pressure from Donta Hightower closing from his right, but he was heavily downgraded for it and yet still grades pretty well overall thanks to the balance of positive plays he has made this season.

After two weeks of the season the rankings are what they are – vulnerable to being influenced by a couple of big plays either way, or simply a run of good or bad form over two games that a player will not sustain over sixteen. The difference between being No. 3 in the rankings and No. 15 are probably three or four throws – we are dealing with very small sample sizes here, but we shouldn’t dismiss the play of Tannehill just because the raw numbers don’t match. Are we saying Ryan Tannehill is better than Peyton Manning because he has graded higher over two games? Of course not.

But Tannehill has made tougher throws more frequently and runs an offense far less predicated on three-step drop quick passes. While Manning has averaged less than two seconds per pass, the quickest in the league, Tannehill has the ball in his hands on average half a second longer per passing attempt. That’s a middle of the pack figure but more importantly it’s extra time that pass rush has to get to him and make his life tougher. Manning has felt pressure on just 10 snaps this season while Tannehill has had 25 plays where he has been pressured.

The point here isn’t to try and make out that Ryan Tannehill is in some way better than Peyton Manning. Nobody at PFF believes that and if you made each staff member GM for the day and asked them to pick one of those quarterbacks to win you a game, each and every one would choose Manning. The point is to illustrate that over two games there are reasons why Tannehill has received a higher grade than Manning and a higher grade than his passer rating suggests.

Football is a complex game, with a lot of moving parts and interdependent pieces all trying to come together with the same goal. The Miami passing attack hasn’t been firing on all cylinders – there have been miscues across the board that have resulted in some pretty pedestrian passing statistics for Tannehill, but it’s important that we realize how deceptive those statistics can be and actually look beyond them – to the tape.

When you grade Tannehill throw by throw and then compare him to the rest of the league’s quarterbacks by the same measure, he looks a lot better. Maybe he won’t sustain that over sixteen games, but let’s not dismiss the man’s performance because of his passer rating.


Follow Sam on Twitter: @PFF_Sam



| Senior Analyst

Sam is a Senior Analyst at Pro Football Focus, as well as a contributor to ESPN and NBCSports.

  • 561sportsfan

    Absolutely fantastic write up, Dolphins beat writers were killing you guys all week, glad you showed them up

  • Hmm

    No mention of the Buffalo game where you had him at a +3.0 and then changed it to +1.0? He was very inaccurate all game long. Yes he had a few drops by receivers but he also had some very poor throws. Including one to Gibson that would have been a TD in the redzone.

    • PFFSamMonson

      Which is why his grade ended up around average in that game. The change is down to the weekly A-22 film review process we run on each and every game.

    • Ethan

      For what its worth that would not have been a touchdown, but yes you are correct there were some botched throws.

  • Chris

    I couldn’t find a Dolphins fan who thought he was good last week. For whatever that’s worth.

    • Pygskyn

      Heres one (So now you can’t say that). He had a rough week, but he wasn’t the cause of that loss by any stretch, Philbins ultra conservative mistakes, key drops by recievers, horrific special teams play, and the running games inability to get rolling were the cause of that loss.

      • Chris

        Fair enough. I know fans are often more down on their team after a loss and vice versa after a win, but when I can’t find one of them who thinks their QB was decent and can’t find an analyst (non-PFF football guys) defending him, I’m a bit puzzled when I see his great grade on here.

        I’ll give PFF the benefit of the doubt and I’m glad they’re explaining the grade, but I don’t think it’s hard to see why I (and others) were very surprised to see that grade.

        • Pygskyn

          Oh yeah, I totally get that. I understand the difficulty because on an initial eyeball test, it looked ugly out there. Sadly most analysts are pretty much garbage nowadays too, making it even harder for casual fans or fans from other teams that don’t really study the players of a specific team to filter through all the noise and get the straight unbiased (relatively) details.
          When watching the game the first time I was tossing my hands up and cursing Tannys name, but forcing myself to sit through a repeat with my critical cap on though definitely showed a different story.

          • Randy Richbourg

            Also, keep in mind that analyst today, are nothing more than unemployed football players who were given a title to speak their opinion for competing networks. Nothing more.

  • Eclectic Misfit

    How about taking into account that “it” factor of a QB? Specifically that of an elite franchise QB. Peyton may not face too much pressure because the man is a genius running his offense and making adjustments. Tannehill has not only been inaccurate but has been inaccurate in crucial parts of games. Especially in the Buffalo game. An elite franchise QB would have made those critical plays and kept their team in the game. Tannehill gets criticized because it’s now becoming a pattern in regards to his inaccuracy and bad pocket awareness.

    • Ajit

      This is really hard to parse through. Absolutely true that Mannning’s quick release and decision making help explain why he isn’t pressured. At the same time, his receivers still need to get open for him to make that quick throw. It’s really hard to tell and Pff admits that their grades are doing the best they can. It’s a piece of the picture, not the entirety of it.

    • Finsfan22

      Dude Brady missed a bunch of throws late vs miami. It happens when you get sacked like 4 times. (Fins fan).

    • Harold

      Aah, the magical “it” factor. When you can come up with a way to measure, and grade this “it” factor, let us know. Until then it shouldn’t factor into the ratings. And if you say that Tannehill has been inaccurate at crucial times of the game, then it would be safe to say that the dropped passes would have been at crucial times as well.

      • Jose Burgos

        Those “it” factors you are talking about are called intangibles. I for one think Tannehill played decent against New England and subpar against Buffalo. Some of the sacks he took last week were on him not getting rid of ball faster or having pocket awareness to scramble. He needs to get in the face of his receivers/running backs when they drop passes as well to let them know that is unacceptable! Those in my opinion are what are intangibles.

    • Gregory Hutchinson

      Not always, I & the rest of the world has seen Manning, Brady, Brees & Rodgers miss some wide open guys this yr. already. It happens to all QB’s, but at the same time I agree he could be more accurate. Every QB could be a lil better at something, Denver would like Manning to be faster, Washington would like RG111 to be a bit tougher, the Saints would like Brees to be a bit taller etc. EVERY QB has some flaws, some worse than others, but all have parts of their game they wished was better.

    • bobrulz

      And I’m sure Team A beat Team B because they “wanted it more” right?

  • Hal

    Hard to evaluate QBs without evaluating their protection calls. Tannehill’s protection issues have spanned across multiple coordinators and talent on the OL.

  • Fintasy

    I love PFF, but I hope they consider subtleties in T-Hill’s play a little bit better. Tannehill has missed deep full-stride connections with Wallace going all the way back to last season. Often because Wallace is usually thrown the ball split seconds too late. That slight mis-timing is not always obvious to the viewer, but it’s important to the success of the connection, and the accuracy of it’s subsequent analysis.

    • Gregory Hutchinson

      They take all passes into account(the good & the bad). I have been telling people that Tannehill has played better than his statistics say he has. Now he has threw about 5 or so bad passes & that is a fact, but you must remember he has already thrown the ball close to 80x. And here’s an example, say you take some interceptions that may happen through out the yr.(lets say he throws 12 of them in the reg. season) & he is credited with them 12, but what the QB ranking stat’s doesn’t tell you is that 4 of them ints. were last play of the game desperate throws because they were behind in the game. And another 2ints. were right off his WR’s hands & another 2 was because the WR fell down as the ball was in the air & on target but being he fell the CB caught the ball instead. See, right there is 8ints. that PFF takes into account about what REALLY HAPPEN ON THAT PLAY & not just spit out the stats. It’s no different than a RB running for a 113yds. on 27 carries & you think he had a great game because the stats say he avg. more than 4yds. a carry. What the stats don’t tell you is he really had a awful game. If it weren’t for that one run that netted him 64yds. his stats would be 26 carries for for 49yds., wich is less than 2yds. carry. See, the stats have told you what a great game he had & the truth is that couldn’t be further from the truth. That is what is not good, but great about the PFF stats as compared to all others & I mean all others. They literally take every play & circumstance into account !!!!!!

  • Gui

    I love how PFF’s always willing to go against the same old clichés. Keep up the good work!

  • michigan fin fan

    The problem with Tennehill this year has been his ball placement. Even when he is completing passes they have been poorly placed eliminating any chance of YAC.

    • Pygskyn

      Which if you note is something they take into account. Both of the plays they listed above were well placed balls that would have added about 60 yards and 2 TD’s to Tanny’s numbers had they been caught. Both right over the shoulder into the breadbasket. Nobody was placing those passes any better. He has shown some issues on crossing routes, but as they note, PFF takes that into account when they grade.

  • Maine Fin Phan

    How about that bomb that Hartline watched slip through both arms at the end of the Buffalo game? That would have really padded the QB stats too. Looked like it would have been a TD. But then again Hartline probably would have fallen down as soon as he caught it. He usually does.

    • DolphinsRuleAll

      LOL Yeah he so does. Almost always. I have no idea what the stats are, but if you told me he was among the worst in the league in YAC I’d believe you.

  • Matteo ‘go Fins!’

    Always useful pont of view by PFF! But what about the inaccurate pass that should fall incomplete and is good thanks to the the receiver? Tannehill had many. Maybe more than 113 yards. Are they subtracted as much as the dropped pass are added? Tannehill was quite inaccurate. That made our passing game predictable and helped defenses to fill the pack and stop the running game. I wouldn’t choose this Tannehill above at least 25 other starting QBs. I hope to see another Tannehill next week.