Guest Post: Introducing Marginal Yards After the Catch, Part I

Introducing Marginal YAC

A quarterback throws a two yard dump off pass to his running back, who proceeds to juke a couple defenders and run 78 yards into the endzone. Naturally, the quarterback deserves credit for an 80 yard pass. Wait, what? Sounds illogical, but that’s the way the NFL has been keeping records since 1932, when it first began recording individual player yardage totals. The inclusion of YAC — yards after the catch — in a quarterback’s passing yards total can really distort efficiency stats, which in turn may distort the way he is perceived.

In response, I created a metric called Marginal YAC (mYAC), which measures how much YAC a quarterback has benefited from compared to an average passer. Its calculation is very straightforward:

mYAC = (YAC/completion – LgAvg YAC/completion) * Completions

I have quarterback YAC data going back to 1992 for every quarterback season with at least 100 pass attempts.1 That gives us a healthy sample of 965 seasons to analyze, and includes the full careers of every contemporary quarterback. But first, let’s get a sense of what’s average here. The table below shows the league-wide YAC rates since 1992:

YearAttCompYardsYACYAC/CYAC %
20131813611102129177621715.648.1
20121778810833125951575715.3145.7
20111741010464125330581295.5646.4
20101726910491120964570475.4447.2
20091703310372118917564785.4547.5
20081652610081114766522465.1845.5
20071704510425116874510644.943.7
2006163899796112277503575.1444.9
2005164649790111721475544.8642.6
2004163549772115338472544.8441
2003164939695109467477624.9343.6
20021729210314116201506234.9143.6
2001161819542109639475564.9843.4
2000163229497110131477755.0343.4
1999167609567113254490825.1343.3
199815489876610608643872541.4
1997157298844105288439744.9741.8
1996159669198106661463975.0443.5
1995166999717113069492445.0743.6
1994150568739101884442705.0743.5
199314414835196490396804.7541.1
199213408770592011368454.7840

With the emphasis on the short passing game in recent years, it’s no surprise that YAC per completion has increased over the last decade. In 2013, yards after the catch represented an incredible 48.1% of gross passing yardage. It’ll be interesting to see if the trend continues and eventually pushes YAC above 50%.

So which quarterbacks have benefited the most and least from YAC? The table below shows the top 50 seasons of Marginal YAC since 1992. Daunte Culpepper, with 534 mYAC, is the single-season leader. In 2004, his targets gained 2,376 yards after the catch, producing 6.27 yards per completion after the catch. Since the average reception gained only 4.84 YAC in ’04, that means Culpepper’s Vikings gained 1.43 marginal YAC compared to league average. Multiply that by his 379 completions, and you get Culpepper’s mYAC number of 543.

#QuarterbackTeamYearmYACAttCompYardsYACYAC/CYAC %
1Daunte CulpepperMIN2004543548379471723766.2750.4
3Donovan McNabbPHI2006472316180264713977.7652.8
4Kurt WarnerSTL1999461499325435321286.5548.9
5Tony RomoDAL2009451550347448323406.7452.2
6Boomer EsiasonNYJ1993426473288342117956.2352.5
7Steve YoungSF1997424356241302916226.7353.5
8Matt CasselNE2008421516327369321166.4757.3
9Jake DelhommeCAR2005419435262342116926.4649.5
11Trent GreenKC2003412523330403920386.1850.5
12Trent GreenKC2002408470287369018176.3349.2
13Kurt WarnerSTL2000404347235342915866.7546.3
14Brett FavreGB2007387535356415521315.9951.3
15Jason CampbellWAS2009380507327361821616.6159.7
16Cam NewtonCAR2012360485280386918486.647.8
17Brett FavreGB2006358613343388521216.1854.6
18Steve YoungSF1994349461324396919906.1450.1
19Nick FolesPHI2013345317203289114827.351.3
20Chris ChandlerATL1998345327190315412966.8241.1
21Vinny TestaverdeCLE1994335376207257513846.6953.7
22Rich GannonOAK1999330515304384018906.2249.2
24Brett FavreGB1995315570359441321345.9448.4
25Matthew StaffordDET2013314634371465023926.4551.4
26Jeff GeorgeATL199531455733641432017648.7
27Rich GannonOAK2002309618418468923615.6550.4
28Donovan McNabbPHI2004304469300387517555.8545.3
29Jason CampbellOAK201030332919423871358756.9
30Brett FavreGB2001302510314392118675.9547.6
31Donovan McNabbPHI2009293443267355317476.5449.2
32Brandon WeedenCLE2012293517297338518716.355.3
33Steve YoungSF1993284462314402317765.6644.1
35David CarrHOU2004282466285353116605.8247
36Steve YoungSF1998279517322417018915.8745.3
37Matt SchaubHOU2011276292178247912657.1151
38Neil O'DonnellPIT1993274486270320815575.7748.5
39Bobby HebertNO1992273422249328714645.8844.5
40Gus FrerotteWAS1997265402204268212796.2747.7
41Jake PlummerDEN2004264521303408917295.7142.3
42Matt HasselbeckSEA2003262513313384118045.7647
43Ryan FitzpatrickBUF2012260505306340018866.1655.5
44Drew BreesNO2006260554356441820905.8747.3
45Aaron RodgersGB2009258541350443421646.1848.8
46Aaron RodgersGB2011258502343464321636.3146.6
47Drew BreesNO2008257635413506923985.8147.3
48Carson PalmerOAK2012255565345401820886.0552
50Trent GreenKC2005253507317401417935.6644.7

The list is primarily comprised of great quarterbacks playing with great receivers, and bad quarterbacks who threw a bunch of checkdowns. The dichotomy between those two groups makes further analysis a bit murky. How should we divvy up credit for YAC between the quarterback, his receivers, and the system he plays in? I’ve isolated a few cases that may shed some light on the situation.

Here are the Packers during the Mike McCarthy era, with six seasons from Aaron Rodgers and two from Brett Favre:

McCarthy PackersTeamYearmYACAttCompYardsYACYAC/CYAC %
Aaron RodgersGB2013242290193253613236.8552.2
Aaron RodgersGB2012164552371429521355.7549.7
Aaron RodgersGB2011258502343464321636.3146.6
Aaron RodgersGB2010180475312392218776.0247.9
Aaron RodgersGB2009258541350443421646.1848.8
Aaron RodgersGB2008-115536341403816524.8440.9
Brett FavreGB2007387535356415521315.9951.3
Brett FavreGB2006358613343388521216.1854.6

Playing in McCarthy’s system, Green Bay quarterbacks have benefited from positive mYAC in seven out of eight seasons, with an average of 216 mYAC per year. That’s very high, especially considering that such a level was sustained for nearly a decade. Rodgers and Favre are both future Hall-of-Famers, so it’s fair to hypothesize that they have certain skills that allow their receivers to gain a lot of YAC. However, the Packers have been widely recognized as having one of the best receiving corps in the league during that span, and McCarthy has a reputation as a sharp offensive mind. Given all of that, do Rodgers and Favre deserve credit for their receivers’ YAC, or are they beneficiaries of a strong supporting cast and an advantageous system? What do you guys think?

Now let’s take a look at the Patriots during the Bill Belichick era:

Belichick PatriotsTeamYearmYACAttCompYardsYACYAC/CYAC %
Matt CasselNE2008421516327369321166.4757.3
Drew BledsoeNE2000147531312329117175.552.2

This gets even more interesting. The Patriots have gained positive mYAC in 13 of 14 seasons under Belichick, including two seasons over 400. The Pats have been YAC monsters with a HoF quarterback, a mediocre quarterback, and a replacement level quarterback. They’ve done it with a revolving door of receiving targets, representing every conceivable type of pass catcher. They’ve done it with several offensive coordinators, applying conservative run-heavy schemes in the early years, and pass-wacky offenses in recent seasons. The only constant is Belichick. Does this mean Tom Brady is a product of his system? Would he be benefiting from so much YAC if he played for a different coach? The answer might be found in Matt Cassel and Drew Bledsoe:

Cassel with Belichick: +421 mYAC
Cassel without Belichick: -216 mYAC
Bledsoe with Belichick: +147 mYAC
Bledsoe without Belichick: -870 mYAC

The circumstantial evidence does not look good for Brady. Two passers who produce negative mYAC in other systems suddenly become YAC artists playing for The Hoodie. Furthermore, Brady has never been able to throw downfield consistently, so he seems like exactly the type of quarterback who would be dependent on Belichick’s YAC-heavy short passing system to cover his deficiencies.

One last table for the day – the 50 worst Marginal YAC seasons since 1992:

#QuarterbackTeamYearmYACAttCompYardsYACYAC/CYAC %
965Matt RyanATL2010-524571357370514173.9738.2
964Jon KitnaCIN2001-491581313321610693.4233.2
963Mike GlennonTB2013-45841624726089253.7435.5
962Peyton ManningIND2010-444679450470020034.4542.6
961Scott MitchellDET1996-38943725329178873.5130.4
960Drew BledsoeNE1994-359691400455516674.1736.6
959Jeff GeorgeIND1993-35440723425267583.2430
958Peyton ManningIND2006-353557362439715084.1734.3
957Drew BreesNO2010-339658448462020974.6845.4
956Mark BrunellJAC2001-336473289330911043.8233.4
955Hugh MillenNE1992-32420312412032692.1722.4
954Browning NagleNYJ1992-30738719222806113.1826.8
953Mark BrunellJAC1996-299557353436714824.233.9
952Ryan FitzpatrickCIN2008-29737222119058483.8444.5
951Jon KitnaSEA1999-296495270334610894.0332.5
950Peyton ManningIND2008-296555371400216274.3940.7
949Matt RyanATL2009-291451263291611414.3439.1
948Neil O'DonnellCIN1998-28434321222167773.6735.1
947Ryan TannehillMIA2013-284588355391317044.843.5
946Carson PalmerCIN2004-28443226328979883.7634.1
945Mark BrunellJAC2000-283512311364012824.1235.2
944Mark SanchezNYJ2012-281453246288310264.1735.6
943Jon KitnaCIN2002-281473294317811623.9536.6
942Chris ChandlerHOU1995-28035622524608603.8235
941Kordell StewartPIT1998-27945825225609823.938.4
940Eli ManningNYG2008-278479289323812204.2237.7
939Dan OrlovskyIND2011-27819312212014003.2833.3
938Byron LeftwichJAC2004-277441267294110143.834.5
937Mark BrunellJAC1999-274441259306010554.0734.5
936Troy AikmanDAL1996-269465296312612244.1439.2
935Steve McNairTEN2001-269431264335010473.9731.3
934Carson PalmerCIN2010-269586362397017004.742.8
933Carson PalmerCIN2005-268509345383614084.0836.7
932Matt HasselbeckSEA2006-26737121024428133.8733.3
931Matt MooreMIA2011-26634721024979014.2936.1
930Kerry CollinsTEN2010-26427816018236063.7933.2
929Matt SchaubHOU2013-26335821923109634.441.7
928Steve McNairTEN2002-262492301338712154.0435.9
927Mark BrunellJAC1995-25834620121687613.7935.1
926Warren MoonHOU1993-257520303348511833.933.9
925Jim KellyBUF1994-247448285311411974.238.4
924Scott MitchellDET1994-2462461191456357324.5
923Eli ManningNYG2012-245536321394814614.5537
921Matt CasselKC2009-244493271292412324.5542.1
920Jim EverettNO1996-243464267279711044.1339.5
919David GarrardJAC2008-242535335362014944.4641.3
918Peyton ManningIND2002-240591392420016844.340.1
917Andy DaltonCIN2011-239516300339814284.7642
916Colt McCoyCLE2011-233463265273312394.6845.3

There are a lot of good quarterbacks on this list, including Peyton Manning (x4), Carson Palmer (x3), Mark Brunell (x5), Matt Ryan (x2), Steve McNair (x2), Drew Brees, and Troy Aikman. Why did these guys lose out on so much YAC? Surely their offensive systems were a major factor, emphasizing intermediate sideline routes rather than short passes over the middle. But that begs another question: Do these quarterback’s deserve extra credit for making harder throws that are less reliant on the receiver creating YAC? I’m still torn on what this all means for evaluating quarterbacks.

Before we get to Part II, though, I’d love to hear feedback from you guys.

1. This data comes courtesy of sportingcharts.com. It’s obviously unofficial, but there doesn’t seem to be any noticeable biases from one team to another. Some unofficial stats, such as passes defensed or quarterback pressures, can vary wildly depending on the scorekeeper, but Sporting Charts’ YAC stats seem pretty fair, from what I can tell. Here is a link to the 2013 data. Chase note: I have not had the chance to compare these numbers to what is on NFLGSIS, but that’s a good idea. []
• Neat stuff! One interesting note on the trend in YAC%. It drops in 2004 and 2005. Possibly a blip, but perhaps more likely about the temporary greater enforcement of illegal contact on DBs (easier to complete downfield passes). Also interesting that Brady’s first great season (’04: 4th in DVOA w/ Branch and Givens) was his lowest in YAC% and negative in mYAC.

• Red

Thanks, Andrew! I noticed the same thing with YAC dropping steeply in 2004, and agree that the illegal contact rules were almost certainly the cause. I actually make reference to this in Part II, but you’re one step ahead of me 🙂

There are a number of QB’s whose best seasons featured their lowest YAC %, including Carson Palmer (05-06), Steve McNair (01-03), Mark Brunell (96-01), Scott Mitchell (94-96), Jeff Hostetler (93), Tom Brady (04,07), and Peyton Manning (03,06,08). These offenses all featured lethal downfield throwing, which essentially bypasses the need for YAC from short passes. At least that’s my theory.

• Chase Stuart

Interesting observation about the ’04 season; seems like a reasonable interpretation of the data. What about Philip Rivers? It feels like his best years came when he was throwing bombs to Jackson and Floyd.

• James

Which is why I’m fascinated that his two seasons with the most YAC are 2009 – one of the era’s greatest collections of offensive talent (Rivers, Gates, VJax, aging Tomlinson, Sproles, Floyd) – and 2010 when everything else fell apart (Tomlinson gone, Mathews, Floyd, and Gates got hurt after half a season, VJax held out for 10 games, at one point the primary skill players were Sproles, Tolbert, Legadu Naanee and Patrick Crayton!).

• Red

Of all the QB’s in my dataset, Rivers surprised me the most. I echo Chase’s sentiment that Rivers’ prime was all about hitting the deep ball, with relatively little YAC involved. But I was wrong. In reality, Rivers has been a deluxe version of Joe Flacco – a lot of deep balls, a lot of screen passes, and not much in between. He racked up a ton of yardage on screens to LdT and Sproles, which of course involved gobs of YAC. He also maximized the mismatch of Gates by feeding him plenty of short passes and letting Gates run after the catch.

There are a handful of QB’s who consistently produce both YAC and Air Yards above league average, and Philip Rivers is one of them, which makes him a special player and probably a Hall-of-Famer if he had better luck in the playoffs. Other QB’s of this unique mold include Steve Young, Kurt Warner, Trent Green, and Aaron Rodgers.

• Malene, CPH

This is really interesting, but you’re losing me when you want to use it to rate QBs. Why does it have to “not look good for Brady”? I think this type of analysis is way more interesting when used for descriptive rather than normative/prescriptive purposes.

• Chase Stuart

Malene,

One thing to keep in mind is that this is only Part I. I totally understand where you’re coming from about using this for descriptive rather than normative purposes, and I think that makes a lot of sense. But one thing we’re going to try to do is see if we can also tackle that other question. Not an easy one, that’s for sure.

Do you have any advice on how we might go about doing that?

• Red

Malene, my intention here is not to “rate” QB’s, but to explore the factors that are responsible for their success or failure. Let me explain why I said it doesn’t look good for Brady. Make no mistake, he was the leader of a juggernaut offense in four different seasons (07,10,11,12), which is inarguable and a big feather in his cap. During his `10-`12 campaigns, most of Brady’s marginal value as a passer came from producing exceptional amounts of YAC, whilst his Air Yards production was right around league average (as you’ll see in Part II). There’s nothing wrong with that. However, Matt Cassel and Drew Bledsoe ALSO produced big YAC numbers in Belichick’s offense, which caused me to question whether Brady’s YAC value was really about him or more a product of the system. The answer: I don’t know. But I do believe that the common narrative of “Brady has carried the Pats offense for 13 years” needs to be revised, because Belichick’s scheme, game planning, and drafting have all contributed a great deal as well, as evidenced by going 11-5 with Cassel in 2008.

Interesting Chase, thanks.

But if we really want to make progress towards separating quarterbacks from wide receivers you’re going to need to break out YAC by route type. For example, there isn’t much of a YAC expectation on a hitch or a comeback, so any YAC is credited to WR.

However, these beautifully timed ins and drags thrown by Peyton manning would get more YAC credit.

You’d need to create a YAC ratio for each route type, then adjust YAC accordingly.

• Chase Stuart

Thanks, Nick. If I recall correctly, Pro Football Focus publishes data on specific routes. That might be something we could try to incorporate here. Do you have any other thoughts on how to move towards Part II of this series?

• James

If you have routes types (or to simplify, average depth of target) you could control for how much YAC was expected based on routes or aDOT, and then see how much a team is above or below expectation. Then you could look over receivers and QBs careers to see if there is a consistent trend. I’m thinking YAC on short throws (under 5 yards) should go solely to the receiver, while the QB should get some credit for YAC on deep throws (although this is likely a very small percentage, and probably still depends on the receiver more).

Also, I wonder how different the data would look if the league average YAC was smoothed some. While there is a trend the league average YAC/C varies a lot, and it’s particularly wild in the 2005-2009 range. You could end up looking very positive twice, zero twice, and very negative once despite having the exact same 5.16 YAC/C every year. Maybe a 3 year rolling average or something else?

• Red

Nick and James, I don’t have YAC broken down by route, although I would love to have that data. The resulting evaluations would undoubtedly be more precise, and likely more predictive as well.

As far as using a 3 year average, I think it works in some instances but not in others. During a span of seasons where the rules stay the same, such as 2005-2009 that you mentioned, it probably would be more accurate to smooth out the league averages. However, 2003 and 2004 should NOT be lumped together since they were played under different rules (illegal contact enforcement). Same thing with 2010-2011 (defenseless receiver protection, helmet contact), and 1977-1978 (Mel Blount rule, offensive linemen use of hands).

• The Ancient Mariner

As much as I’ve seen in recent years about throwing receivers open and leading them with the ball and the like, it seems pretty clear that on some routes, at least, the QB’s accuracy and ball placement do play a significant role in making YAC possible (though what comes of that is still up to the receiver, of course). Are those situations just too small a percentage of completed passes, or is this a skill which is not sufficiently repeatable to make a significant difference?

• Red

That’s what I’m trying to figure out. I agree that on certain route types, the QB can markedly increase YAC by throwing an accurate, well-timed pass. The slant and shallow cross are the routes where YAC is likely most affected by QB accuracy. The problem is, there is a lot of noise in YAC, because the majority of its causes are outside the quarterback’s control. Even over a full season, a handful of fluky long YAC plays can really distort a QB’s numbers. Remember Victor Cruz turning a 10 yard pass into a 99 yard TD with one broken tackle? Those types of plays make a huge difference in the short run, but largely even out over the course of a career.

Subjectively, there are some instances where we can be pretty sure that YAC is a product of the quarterback’s skill. Steve Young is a great example, because he played in a West Coast Offense where timing and accuracy are paramount. Year after year, he was among the league leaders in both Comp % and Y/A, which signals to me that he was a major contributor in generating YAC for the 49ers offense.

• Jp

Great guest post.
We should get together on a Sunday and watch some games / geek out on stats. (I’m a numbers guy, too). I do a lot of spreadsheets, myself.
If you’re game, have Chase forward my email to you.
Apologies for using your comment box as a personal message, Chase.
Keep up the great work.

• Last year’s YAC 58752. Eli’s 2011 season had 2089 YAC, sportingcharts lists him as having 2175. I’m not sure where they are getting their numbers from but they are a little off.

Two types of passing plays generate a disproportionate amount of YAC: screen passes and really long passing plays.

• Chris,

The data on the site comes from STATS. Where are you seeing that Eli had 2089 YAC?

Cheers,
SportingCharts

• Chase asked me about Eli’s 2011 YAC last year so I had his numbers readily available. The YAC numbers for each play are in the pbps link on the nflgsis.com website.

• Kibbles

Right. Mike Clay was talking about that earlier this offseason, that YAC does not change linearly, but rather exhibits a parabolic distribution. The shortest and the longest passes generate the most YAC, while the mediumest passes generate the least.

That’s also a possible explanation for Payton Manning’s low mYAC totals. It seems like he’s a guy who has really made his living on those intermediate routes.

• Red

Kibbles, this is definitely true. Screen passes and dumpoffs generate YAC because the receiver has room to run in front 0f the defense, while deep passes generate YAC because the receiver has room to run behind the defense. Meanwhile, passes in the 10-20 yard range are usually into the teeth of the defense or near the sideline, which hinders YAC. Obviously, an offense can still be very successful without the aid of YAC, as Peyton Manning has repeatedly shown.

• Bill

Perhaps this analysis would also benefit from evaluating the YAC stats of the WRs/TEs/RBs that the QBs were throwing to? Many WRs have caught passes from many different QBs, so you could analyze their YAC with different QBs and try to analyze the effect of the QB on his receivers. Then again, the team system could also have a strong effect on this analysis.

• naturallythe1

Favre took more shots down the field and played with more heart than Rodgers. Rodgers is a little girl who cries when he is knocked down and calls for flags all the time. He never takes blame when his receivers on don’t catch the ball and he completes a huge amount of dump off passes and screen passes for big YAC, that it’s crazy. He could throw 30 times a game and 80% of the passes would be short passes under 5 yards that go for 5-10 yard plus YAC. Plus his O line gives him almost 5+ seconds to throw the ball. Overrated.