Pairing a QB with his Receiver(s)

rotogrinders_user_19856

One of the most popular tactics in daily fantasy football is of course “stacking.” Technically speaking, stacking is using a whole bunch of players from the same team – most popular in baseball – but the term has been loosened to include quarterback/wide receiver pairings.

Editor’s Note: RotoAcademy has added NEW NFL courses. Click here to browse through and improve your DFS play!

Naturally, we kind of know that pairing a wide receiver with his quarterback is going to increase upside and risk at the same time since the duo is dependent on one another for production. If you look at the numbers, you see a ridiculously strong relationship between quarterback and wide receiver fantasy points. From Fantasy Football (and Baseball) for Smart People:

When playing daily fantasy football, you can increase upside by pairing your quarterback with one of his receivers. If the quarterback has a big day, which is pretty much a prerequisite for taking down a tourney, it’s highly likely that your pass-catcher will produce as well. Take a look at the strong correlation between quarterback points and team wide receiver points.

When a quarterback has at least 30 points on DraftKings, there’s roughly a 91.7 percent chance that his wide receivers will combine for 30 or more points, an 83.3 percent chance of them checking in above 40 points, a two-in-three probability of 50-plus points, and incredibly a one-in-three chance of at least 70 combined points.

That last tidbit is really surprising – a one-in-three chance of 70 combined receiver points when a quarterback posts 30? That’s ridiculous.

So to say stacking in daily fantasy football tournaments is a smart strategy is an understatement; pretty much every top player pairs a quarterback with at least one of his wide receivers in every large league.

In cash games, though, you’re not concerned solely with upside, but also consistency. We know that stacking is volatile, but can the upside make up for the inconsistency?

The Relationship Between QB and WR Production

There are a handful of ways to determine if pairing a quarterback and wide receiver on the same team is “worth it” in cash games. I chose to look at how often receivers kill it when their quarterbacks do the same versus how frequently the two “tank” together.

Looking at quarterbacks who played full seasons over the past two years, I broke down their production into buckets, analyzing their top-four and top-eight games. Here’s how the production for their WR1 and WR2 looks in those buckets.

If a wide receiver’s play weren’t tied to that of his quarterback, we’d expect a wide receiver to have a top-four fantasy game—that is, one of his best four games on the year – 25 percent of the time when his quarterback does the same. So we’re looking at a quarterback’s top-four (or top-eight) fantasy games, then determining how frequently his wide receivers turn in the same sort of performance.

Well, both WR1s and WR2s produce top-four and top-eight games at a rate greater than what would be produced from chance alone, which is what we’d expect. When a quarterback has a top-four performance, there’s a 40 percent chance that his No. 1 wide receiver also has a top-four performance and a 33 percent chance that it happens for his No. 2 wide receiver.

Not surprisingly, WR1 production is more closely linked to his quarterback than WR2 production. That means that pairing a top receiver with his quarterback possesses more upside than doing the same with a No. 2 receiver and his quarterback. But is it much riskier, too?

Here, I’m looking at the same thing as the first graph, except with bottom-four/eight performances. So if a quarterback lays a dud (one of his worst four/eight games), how likely is it that his wide receivers did the same?

Again, the results exceed the random expectation. But there’s actually a stronger correlation than with the top performances; WR1s have a 50 percent chance of having one of their worst four performances of the year when their quarterback does the same. The rate is 38 percent for WR2s.

Now let’s compare that risk with the upside from the initial graph.

What we want to see here is that the “top-four” correlations outweigh the “bottom-four.” They don’t. For both WR1s and WR2s, there’s more risk in playing with their quarterback on fantasy football teams than there is upside.

Note that the effect switches with “top-eight” performances, but just slightly. Those results aren’t significant, while the gap in the “top-four” category is meaningful.

When to Stack

One of the important things to remember in cash games is that consistency can matter as much as total points. We always want players projected to score a lot of points, but it’s preferable to create a team that has a high floor from week to week.

Stacking a wide receiver with his quarterback decreases your floor, and it doesn’t seem that your team’s week-to-week ceiling will rise enough to compensate for the downside. You’re kind of creating an asymmetrical team – which is actually the goal – but in the wrong direction.

So should you ever stack? You should do it all the time if you need upside, like in GPPs. When a wide receiver is almost the sole focus of his quarterback – like the Stafford-to-Johnson connection in Detroit – it creates a more volatile relationship.

It really comes down to targets. When a wide receiver sees a lot of targets – as in a WR1 – he’s less likely to “get lucky” in regards to efficiency. As sample size increases, numbers regress toward the mean. Compare that to a WR2 who sees fewer overall targets, and thus a more volatile distribution from week to week. He’s more likely to be able to turn in one of his better games, even if his quarterback isn’t on point, because he doesn’t need a truly elite game (like a WR1) to have a top-four performance.

Another huge factor at play is the cost. Spending top-dollar on a quarterback and receiver who play on the same team is expensive and risky. You don’t want to bypass teammates if both are of obvious value, but when things are close, it’s best to diversify in a head-to-head format.

And finally, note that this general heuristic might not extend to all sports. I’m going to do some more research on this topic for my next book on daily fantasy baseball, but I think there’s good evidence to suggest stacking isn’t a poor strategy in MLB cash games.

About the Author

  • Jon Bales (JonBales)

  • Jonathan Bales is the founder of RotoAcademy and author of the Fantasy Sports for Smart People book series.

Comments

  • daveinchi1975

    2012 DSBC Finalist

    Good stuff – of course, the hard part is predicting WHICH QB-WR combo will go off – did ANYONE have R.Wilson to L.Wilson this week against Arizona??? If they did, they made bank even if they missed on their RB picks. A bit easier to pick this week? Eli to Beckham

  • MRCOOKONUTS

    Russell Wilson Luke Wilson combo QB/TE COMBO
    who da thunk it?

  • grux

    I agree that the R.Wilson-L.Wilson stack was tough to predict but my mistake not putting R.Wilson on my radar for cash games.

  • johnnyj580

    Excellent article. This is something I’ve been thinking about over the past 3-4 weeks. Greatly appreciate you doing the legwork to datamine & analyze for the benefit of everyone.

  • WinkingRaccoon

    I did Wilson Baldwin and it cashed pretty damn good so far… Might lose a couple bucks waiting for Monday night game to end.

  • FBZombies

    Russell Wilson seems to always be overlooked in cash games. I’m guilty of that.

  • Esotericpc

    Awesome insight!! I used Rogers this week on DK and paired him with Jordy Nelson and Randall Cobb and got the following:

    Rogers: 18.72
    Nelson: 29.3
    Cobb: 27.1

    Mediocre day for Rogers but they combined for 75.12 total points. Helpful in positioning me to cash something in this weird week. It is rare for me to be able to do such a stack but I found value in other places (Tre Mason, Lamar Miller, Nick Toon).

  • jcleary1

    Wilson is probably the “safest” GPP QB out there that you don’t have to stack. When rostering Wilson, I almost view as rostering two players for the price of one; a QB and a RB. Not many players you can say that about. His 80 yrds/game and 1+ rushing TD upside is like tacking on an additional 3.5 passing TD’s to his total. UNREAL. IMO he’s the safest from week to week to reach his 30+ point ceiling and his price never seems to kill the rest of your LU.

    Stacking is an overrated practice IMO. I find that it often limits your upside by limiting your scope of possibilities. For example, last weekend, if your rostering Rodgers/Nelson, then you likely to have rostered a value WR that blew up your chances of big money (if Rodgers didn’t already do that). Where as you might have been open minded if you rostered a naked QB and ignored their expensive WR1. I’m torn on the subject because I found lots of success with this strategy early in the season, but eventually the whole idea fell flat and hurt me more than it helped.

  • OUBrent1

    There are a ton of examples where stacking worked, and where it didn’t. I agree with guys like Wilson being great GPP players where you don’t stack, but at the same time, none of the players you would stack him with cost much. I mean Baldwin is maybe a mid 6k player, richardson, helfet and wilson are cheaper than that. You have to evaluate the matchups/lines/over unders for the players you are stacking. I’m not going to automatically assume stacking a qb and wr1 are going to be a great play, if they are playing defenses that give up minimal points to WRs. It’s all about understanding the matchups for each player.

  • OUBrent1

    You also have to be careful with some QBs that don’t really have a clear WR1. Rodger is one of them. Yes, nelson is listed as the WR1, but there’s been many weeks where Cobb has out targeted Nelson. GB, NO, Chicago(before marshall), TB, and Houston are all good examples where you have 2 guys that could be considered WR1 guys. You have to be weary of that. If you are stacking guys like Ben/Antonio, Romo/Dez, Luck/Hilton the QB/WR1 distinction is a little easier.

  • dwoods5

    @jcleary1: I think you’re right about stacking being overrated. Seems like you lower overall EV to create “upside,” which is not a statistical term. Creating a higher variance? Yes. Creating more ROI? Don’t think so.

    @JonBales: I apologize for coming on all three of your posts with criticism. It’s not personal, it’s just that I’ve been seriously pondering these issues since I started dff this season, and I’ve read a lot of “generally accepted” advice that just seems to fly in the face of everything I’ve learned about probability and statistics. I’ve actually gotten a lot out of your writings and I’d count myself lucky to become half as successful in the industry as you.

  • daveinchi1975

    2012 DSBC Finalist

    In football, stacking pays off well if you vary your selections – I did quite well this season in NFL and all I did was stack gpp’s each week. Whether its an Eli-Randle-ODB stack or a Rodgers-Nelson-Cobb stack, when those guys go off for 75 points and you stacked them 10 times in a $2 gpp, the odds are good that your variant selections at the other positions (where you take 5 different sub 5K WR’s on 5 different entries for your WR3, etc.) well net you a high cash payout that pays for all of your misses. For example, I liked to focus on teams with 2 similarly priced cheap WR’s and use them both on different entries – so, when Steve Smith goes 4-35 with no TD’s, and Torrey Smith goes 7-107 with 2 TD’s, you have at least 1 entry with your stack’s 75 points paired with Torrey Smith’s 34 points – with 109 and your RB’s-TE-DEF-Flex not accounted for yet, you are cashing well even if you completely whiff on all those other picks.

  • daveinchi1975

    2012 DSBC Finalist

    Volatility will always be a factor with stacking – ask anyone who “safely” stacked Peyton-Sanders-Thomas this week with multiple entries – none of those entries paid because of the amount spent on those 3 and their failure to get close to the 75-point 3-stack barrier. That said (I’m just guestimating here), Rodgers-Nelson-Cobb stacks all season probably hit 75 in over half their games, which means that if you varied your other selections up a bit, your stacking of them was much less volatile than if you went contrarian every week with NJY, Tenn, Arizona stacks each week, where you hit the 75 barrier maybe 3 times all season. Sure, when you hit that 75 with Lindley-Floyd-Brown, your salary cap hit is so low, you can afford to go with Antonio Brown and an ace RB with that stack, it is tough to stomach going 5 weeks in a row without a cash, before you get that one week where the stars align and you cash for 1K. I prefer to grind it out with the usual suspects than take contrarian shots in the dark with my stacking in NFL, but really, it depends upon your tolerance for volatility (and the size of your bankroll).

    As for stacking H2H, or small leagues, it didn’t work back in the baseball team stacking days, so I don’t see how it would work in NFL, even if you only stack all-world QB’s in gravy matchups each week.

  • daveinchi1975

    2012 DSBC Finalist

    When you are stacking a gpp with Rodgers, you aren’t stacking to win it (too many players will own Rodgers independently of stacks and someone will luckbox into some weird combo of Rodgers-Decker-Antonio Brown to beat you anyways). Instead, you have to view your stack as a “cash” stack, where you are seeking a mid-level cash that will pay for all of your stacks that miss – for that reason, you’ll simply stack Rodgers-Cobb and Nelson, so you have the 2 TD combo that appears (one week, its Nelson, the next, its Cobb – like this past week against Detroit) – either way, you’re good – you won’t win the gpp when Nelson goes for 2 TD’s and Cobb doesn’t get one (since someone out there has that weird combo anyways), but you will be well-positioned for a solid cash that pays out 20X what you paid for it. Every now and again, you’ll place a top 50 when something like Eli-Randle-ODB happens (i.e. the wealth is spread between the top 2 WR’s), but really, you are shooting for a 200-1000 place finish when your triple stack hits for a solid 75.

  • whataboutjj

    Really great information in this thread. Thanks for the data analysis Jon & thanks to everyone who has contributed in the comments. I have been stacking QB & WR1 or WR2 but never thought through the QB – WR1-WR2 stack theory.
    J

  • justjeff

    Best article I’ve read on rotogrinders. Well written, direct, and concise. Thanx

  • justjeff

    Same to you daveinchi1975. Great thoughts. My strategy this nfl year will be a qb and two of his targets ten times or so in gpps

  • X Unread Thread
  • X Thread with New Replies*
  • *Jumps to your first unread reply

New to DFS?

Be sure to click through our links and use our exclusive promo codes to receive the industry's best sign-up bonuses, including free access to our premium content.

RotoGrinders.com is the home of the daily fantasy sports community. Our content, rankings, member blogs, promotions and forum discussion all cater to the players that like to create a new fantasy team every day of the week.

Bet with your head, not over it!
Gambling Problem? Call 1-800-Gambler