Swami Yards: Predicting 2018 Passing and Rushing Yards with PACR
Editor’s Note: This guest post was written by Peter Howard. You can follow him on Twitter @pahowdy and donate to his Patreon account.
To compensate for my nervousness submitting to the best two-quarterback and Superflex site out there, I did what I usually do. I came up with as many ideas as I could and collected as much data as I could handle. After a few false starts in different directions, I settled on a process for predicting next season’s passing yards using PACR (Passing Air Conversion Ratio). It’s an advanced efficiency metric from airyards.com (also available on FantasyADHD.com) that we can use to predict the passing yards of quarterbacks in 2018, and I will use it to predict what I call “Swami Yards” — a combination of passing and rushing yards.
My hope is to use PACR in conjunction with the dynasty rankings on this site to target value signal-callers in potential trades. But after some thought and conversation, I decided the process may need a separate article. No one wants constant segues about how the numbers are made in an article about why you should trade for Ryan Tannehill (yeah, I said it).
Let’s see how PACR can be used to project the yardage outcomes for quarterbacks this upcoming season…
Yards are more stable, predictive, and offer a more reasonable expectation of floor for dynasty quarterbacks. Understanding how they are made is therefore important for finding value and winning in league formats that add value to the quarterback position like 2QB and Superflex.
You see, when a daddy yard and a mommy yard love each other very much…wait… Yards are produced, in statistics, by a quarterback’s efficiency and how far he targets his receivers down the field. Yards after catch (YAC) plays an important role for wide receivers, and average depth of target (aDOT) is owned by the receiver, but PACR focuses on the efficiency of each QB with his offense. PACR is more stable year over year than any other efficiency metric. This suggests it’s a quality inherent to the player and not his supporting cast.
For just one minute, let’s forget about wide receivers and how awesome Golden Tate is in the open field. Instead, let’s focus on efficiency, the number of attempts, and a quarterback’s average depth of target.
The Who, What, When, and How
Using FantasyADHD.com, I compiled a list of statistics from every quarterback season across everything from attempts to ANY/A (and all the way back again to sacks and interceptions) since 2012. I then tabled all of the data into a career average. If you’re inclined, I made all of this available in a Google Sheet.
With this data, we can make a whole bunch of interesting findings, like comparing any quarterback’s efficiency to league average year over year:
I outlined the process on Twitter if you want more details about my process or the data.
From airyards.com and the esteemed Josh Hermsmeyer, I know that the formula for passing yards based on PACR is simple:
Attempts * aDOT * PACR = Passing Yards
The attempts portion of our formula is kinda/sorta broken. Despite how much sense some players like Drew Brees make, new potential starters are being docked for not having been starters before last year (or not starting very often). Let’s use Nick Foles as an example. His poor showing in the chart above isn’t about whether Nick Foles is a starting quarterback next season. The problem, if he is a starter, is that he’s doubtful to throw only for his career average of 225 attempts.
I reached out to Justin Freeman, and he was able to provide me with his weighted projections on passing attempts for 2018. He bases them on the tendencies of the offensive coordinator and head coach for the last three years. This was both quicker and better then I could have done by myself given a reasonable time frame. I highly recommend everyone check out Justin and his model. It’s a killer resource, a heck of a deal, and supports a good cause.
Quarterbacks have to suffer the “league average” number because many of the new offensive coordinators have no track record as play-callers. Still, the numbers are coming closer to accurate for 2018 projections.
Rushing to the finish
Next, I brought in rushing yards. In the chart above, Cam Newton looks a little low, as does Russell Wilson. They both add value to dynasty teams every year with their legs. A quick cry for help soon brought TwoQBs’ own Sean Slavin to my rescue. Thanks, Sean!
Don't have my computer with me, so don't have the numbers. But IIRC it is much more stable than any of the passing efficiency metrics.
— Sean (@Slavin22) March 14, 2018
Using Pro-Football-Reference.com, I ran down the average rushing numbers (attempts and yards per rushing attempt) for every quarterback who has attempted at least 30 passes since 2008. This gave me the chance to test the year over year stability (Sean was spot on, of course) and populated my Swami-like projections (passing yards + rushing yards) with player’s average rushing yards (AVG att. * Y/A).
So now the formula looks like this:
(“Predicted pass Attempts using OC/HC tendencies” * “AVG aDot” * “AVG Car.PACR since 2012”)
(“AVG Rushing Attempts” * “AVG Yards Per Rush”)
Swami Projection Yards
Below you will find Swami-like yards (passing yards + rushing yards) projections for every quarterback currently in the NFL. We can use this data to find value at the position in fantasy.
- If we don’t know who a team’s starter is, I’ve included the two most likely candidates.
- Remember, touchdowns are not stable and will get dumped on top of these numbers.
- Since aDOT is not specific to the quarterback, it can change with supporting cast, which would affect the projections. I’ve used the AVG since 2012 to keep the model simple.
- Before you look, say a quiet thank you to Sean Slavin for changing my simple screen-grab and spreadsheet into this wonder of modern interactivity.
A Few Observations
1. Don’t take this as an Endorsement of Deshaun Watson or Jimmy “FFL’s Handsomest Player Alive” Garoppolo.
Deshaun Watson’s 11.01 aDot is the living embodiment of variance bait. His fantasy production was largely based on individual games with four, five, three, and four touchdowns. Touchdowns are the window dressing they use to lure the suckers in. I’m saying this strongly to fight back the overwhelming acceptance he’s had in fantasy ranks. Slow your roll. We do not know where he will balance out.
The same goes for Jimmy GQ. Garoppolo played in only six games in 2017, starting five of them. He was nowhere near as touchdown-dependent as Watson, averaging close to one score per game. He did have well over 300 yards twice and had a consistent 308 yards per game average. It’s probably fair to say he was also playing with a low-level supporting cast. While I’m more willing to “buy-in” to his positive signs than Watson’s, Garoppolo is being valued very highly on a tiny sample size. It could all very easily swing back the other way with just one or two below-average games.
2. Drew Brees is the value you are looking for.
Is Drew Brees old? By NFL standards, yes. Is Drew Brees a top-five quarterback in the league? Again, yes. His PACR predicts him to have as many or more yards than everyone ranked above him (outside of Russell Wilson and a player whose sample size is so small we have to dismiss it, Deshaun Watson). Brees will be a value again in 2018. Who would I trade for him? Well, that’s a bigger conversation. But briefly, would I trade Kirk Cousins or Jared Goff for Drew Brees and another piece? Yes. Yes, I would.
3. Jameis Winston may be the young QB you are looking for in Dynasty.
I’ve been firmly on #TeamMariota since the duo went 1-2 in the 2015 draft. But after failing to finish in the top-12 for his first three years, and eating his fingers in front of a live camera, Winston’s value may have dropped further then it should. PACR loves him to rise into the top six in yards at the position. Plus, he’s only 24 years old and is ranked as QB11 on this site’s dynasty ranks.
4. Kirk Cousins is really high for a 29-year-old player (I love Stefon Diggs and Adam Thielen, too, but maybe too much).
Cousins is good, but in our rush to condemn Washington’s front office’s mishandling of him, we may have jumped the shark. You don’t have to regress his attempts very much (in fact, the weighted tendencies I’m using from Justin Freeman has him projected for 20 attempts over his 2017 total) to find that his yardage floor isn’t going to compete for top-five numbers. Could he do it with touchdowns? Sure. Is he a top-10 quarterback? Yes, I think so. However, he’s not in competition for top-five consideration, outside of heavy touchdown variance.
5. Jared Goff is now overrated based on his position rank.
We have paid for our sins, it’s okay. We don’t have to do this. Just because most of us completely gave up on Jared Goff after his rookie year, doesn’t mean we have to punish ourselves. Not to the tune of QB7, anyway. In his rookie year, Goff had a PACR of 0.77, and in 2017 it was a marvelous 1.01. While his rookie season probably drags his average down too far, the truth is probably somewhere in the middle. Compare him to Jameis Winston (age 24), Derek Carr (27), and Dak Prescott (24).
6. Tyrod Taylor could do well with the Browns.
Taylor is seen as a one-year stopgap starter while the Browns work out the shiny new quarterback they’ll most likely draft. But if Jarvis Landry and Tyrod’s PACR can do anything like they project for, it might be harder to get him off the field in 2018 than we think. What’s more, he wouldn’t have to do nearly half this well to guarantee himself future starts somewhere in the NFL. I like the idea of targeting Taylor at his one-year price, with the hope for more.
7. Teddy Bridgewater may be a slightly better QB option for the Jets.
Teddy’s PACR and aDOT project him for slightly more yards than Josh McCown. But given their recent trade with the Colts, it’s likely both will be fighting for a temp job ahead of the Jets’ new draft pick. Still, I remain, #TeamBridgewater.
Finally, notice that Anthony Spangler is good at rankings. His dynasty Top 200 rankings — while still putting an emphasis on youth — have an overall trend downward in yards. In other words, if you take the higher ranked quarterback per Anthony’s ranks, you will mostly get a good balance of yards and years. Nice. Otherwise, thanks for checking this out! You can find me on Twitter (@pahowdy) anytime, or comment down below and I’ll try to get back to you as soon as I can.