Photo: Imagn Images

Is PPR scoring suitable for dynasty leagues? It seems like an awfully odd question to ask. What makes a scoring system “suitable” or “unsuitable”? And who am I to serve as the final arbiter of suitability, anyway?

I guess a better way to phrase the question is “does PPR accomplish our scoring goals better than comparable scoring systems?” That is an easier question to answer empirically. It’s easy to forget that our scoring system has a purpose other than merely serving as a medium in which we play fantasy football, but I’d say it does, whether explicitly or implicitly.

For instance, we all want to use a scoring system where better players are more valuable than worse players, to whatever degree possible. Obviously no system will be perfect: sometimes a 3-13 team throws 150 passes to a converted safety while one of the best receivers of the generation languishes unused on the worst passing offense in the league.

But by and large, fantasy scoring gives points for good things, subtracts points for bad things, and produces a reasonably sorted list of the best players in the NFL. And we like it that way, which is why we subtract points for interceptions and don’t add any points for dropped passes.

The second implicit goal of fantasy scoring systems is to achieve some degree of parity of value across different positions. In fact, PPR is itself a perfect example of this goal. In the early 2000s, standard scoring was so biased towards running backs that in some leagues over a dozen would be drafted before the first receiver went off the board.

Tired of seeing fantasy leagues value average NFL running backs more than All Pro receivers, owners started awarding a point per reception to achieve a closer parity in positional value. Add in a flex position, which was usually dominated by receivers, and suddenly things became much more even.

This seems like a good thing, so why do I question whether it has a place in dynasty? It’s simple: PPR was designed to make wide receivers as valuable as running backs from year to year. Wide receivers have longer careers than running backs. If they’re as valuable from year-to-year, then they’re more valuable over their entire careers.

In short, common sense tells us that a fix designed to achieve player balance in redraft would naturally throw off player balance in dynasty.

Confirming Common Sense

Common sense and I have a rocky history. In my so-called “Paean to Uncommon Sense” last year, I wrote:

There are few things as dangerous as a plausible narrative. Plausibility is a drug that lulls us into complacency. It causes us to accept without verifying. Some of the greatest fantasy malpractice will be committed in the name of “common sense”. Instead, I would encourage everyone to develop a little bit of uncommon sense. We must start questioning things that sound intuitively right with the same tenacity we typically reserve for questioning things that sound intuitively wrong.

So let’s do that. I have a very plausible theory that a scoring system designed to achieve positional balance in redraft should reduce positional balance in dynasty. Let’s put it to the test.

In the last few days, I’ve written about improving how we calculate value and changing how we measure baselines. I just happen to have built a database of every fantasy season since 1985, and using the approaches I outlined there, I’ve calculated career fantasy value in terms of Estimated Value over Baseline, or EVoB.

My hypothesis is two-fold. First, I hypothesize that running back and wide receiver value will be more balanced in standard scoring than it is in PPR. Second, I hypothesize that fantasy value in standard scoring better represents a player’s “true quality”, (however we measure that), than fantasy value in PPR.

Hypotheses stated, it’s time to gather data and test.

For starters, I ranked the top 500 fantasy players of the last thirty years and broke them down by position. Here's how they stacked up in standard, non-PPR scoring:

(The top 500 threshold gives us pretty much everyone who was even barely fantasy relevant. In that range you find players like Chad Pennington, L.J. Smith, Ernest Graham, and Eddie Royal.)

In a vacuum, that looks a little bit running back heavy at the top, but pretty balanced otherwise. But we don’t play fantasy football in a vacuum, which is a good thing because the noise alone would be unbearable. Let’s see how PPR compares.

Well, that proves it. PPR is just as biased towards WRs at the top as standard scoring is to RBs, but it remains biased the whole way down. What’s worse, while it mildly increases tight end value, PPR achieves this at the expense of quarterbacks, who already get the short end of the stick in most fantasy leagues.

So in terms of achieving positional balance, PPR clearly gets a failing grade. Nothing more to see here, we can all pack up and go home.

Except… well… there’s that pesky second half of the hypothesis.

Does PPR Reward Player Quality?

Are players of similar fantasy value also of similar NFL quality? There’s no perfect way to test this, but let’s start with just an anecdotal comparison.

In standard scoring, the players who rank 48th-51st are Larry Fitzgerald, Ricky Williams, Warrick Dunn, and Reggie Wayne. Now, Warrick Dunn was a spectacularly underrated running back. He is one of just 15 backs to top 15,000 career yards from scrimmage, finishing right between Hall of Famers Eric Dickerson and Jerome Bettis. But was he as good as Reggie Wayne, who ranks 8th in career receiving yards and made three more pro bowls and three more 1st or 2nd team All Pro squads?

Ricky Williams was a terrific back who stood out even in the height of the workhorse era. But is he as good as Larry Fitzgerald, who has a similar career yardage total, fifteen more touchdowns, and a whopping seven more pro bowl appearances? Fitzgerald holds the career receptions record through age 22, 23, 24, 25, 27, 28, 29, 30, 31, and, (barring injury), should easily get it through age 32 and 33, to boot.

Down in the 100 range, we get players like Charlie Garner juxtaposed against players like Derrick Mason. Garner was another very good back, and his 2002 season, (where he had 900 yards rushing and 900 yards receiving), is one of the most underrated years in recent memory. But Derrick Mason had four 1,000 yard seasons in Tennessee, then switched teams and had another four in Baltimore. Were they really comparable as players?

The lopsided comparisons continue all the way down the chart. T.J. Houshmandzadeh vs. Joseph Addai. Mark Duper vs. Ahmad Bradshaw. Chris Chambers vs. Antowain Smith. Mike Wallace vs. Mike Alstott. Ed McCaffrey vs. Marion Barber. In every instance, it seems to me that the wide receiver was simply a better player than the comparably-valuable running back.

How do things look on the PPR side? I mentioned that there were 27 wide receivers in the top 50 compared to just 13 running backs. But that leaves us comparing receivers like Herman Moore against runners like Steven Jackson. Was Jackson a better player than Herman Moore? Was Matt Forte better than Roddy White? Was Ahman Green better than Derrick Mason? Here, I’m not so sure.

Around the 100-player mark, we get Santana Moss vs. Chris Johnson. On the surface, that feels like Chris Johnson was the better player. But in Johnson’s career to this point, he has 10,782 yards and 60 touchdowns. Moss has 10,566 and 66.

Near the 200-player mark, we get an interesting comparison in Julio Jones and DeMarco Murray. Both players emerged from the 2011 draft. Is this an example of an inferior receiver ranking above a superior back because of a biased scoring system? Personally, I don’t think so. It seems like a pretty fair comparison. If anything, it might be the receiver who is underrated.

Enough with the anecdotes, however. Anecdotes are useful illustrations, but they’re no substitute for hard data. While there’s no perfect measure of player quality, Pro Football Reference keeps a statistic called “AV”, or Approximate Value, that I think does a pretty decent job.

You can read a full writeup of AV here, but the short summary is this: AV calculates how much value an offense or defense produced, and then attempts to apportion credit or blame for that production to the individual players. A player who was a very large part of a very good offense earns a lot of AV. A player who was a very small part of a very bad offense earns very little AV.

AV isn’t perfect, (it’s right there in the title: approximate value), but over large samples, it’s going to do a pretty good job at differentiating players. Someone with 151 career VBD may or may not be more valuable than someone with 149, but they’re both very likely to be more valuable than someone with 115.

I pulled up AV totals since 1985 and added them to my list of fantasy values. Immediately, an interesting pattern jumped off the page. Among the top 50 overall players in standard scoring, running backs averaged a career AV of 111.6, while wide receivers averaged a career AV of 142.9. The most valuable receivers were simply much better NFL players than their fantasy peers at running back.

In the 51-100 range, running backs averaged 74.5 AV, while wide receivers averaged 109.1 AV. Not only were the receivers much more valuable than the running backs again, but receivers in the 51-100 range were, on average, about as valuable as running backs in the top 50!

Switching to PPR scoring didn't just reduce this disparity, it eliminated it entirely. Running backs among the top 50 fantasy players since 1985 averaged 126.5 career AV. Wide receivers among the top 50 players averaged 126.2 AV. In the 51-100 range, running backs averaged 85.2 AV while wide receivers averaged 91.4 AV. That’s as close as anyone could reasonably expect, and as much as it differs, it suggests that receivers might actually remain the tiniest hair undervalued.

Drilling down into the data further, I ran a few quick regressions. The correlation between career fantasy value and career AV was remarkably high, ranging from 0.87 to 0.95 depending on position. A strong relationship is unsurprising since both fantasy value and AV are ways of measuring a player’s production, but a relationship this strong is still very impressive.

In standard scoring, the top 100 wide receivers averaged 7.11 more "Estimated Value over Baseline", or EVoB, for every additional point of AV. The top 100 running backs, on the other hand, gained 10.15 EVoB for every point of AV. Even in dynasty, standard scoring assigns far more value to running backs than to comparably talented wide receivers.

In PPR, top 100 receivers gained 11.07 EVoB per AV while top 100 running backs gained 11.14 EVoB per AV. I came into this article intending to argue that PPR scoring was biased towards receivers. You can imagine how shocked I was when I discovered that, instead, it was almost perfectly balanced between the positions.

So is PPR Scoring Suitable for Dynasty Leagues?

I don't mean to suggest that PPR scoring is perfect. Indeed, for all the gains it makes in balancing running backs and receivers, the damage it does to quarterbacks is borderline criminal.

Quarterbacks in both scoring systems gain just 5.58 EVoB per every point of AV. That’s a low enough conversion rate when running backs and wide receivers are at 7-10 points. When they are both at 11+ points of EVoB for every point of AV, it’s a deathblow. The idea that quarterbacks account for just 11% of the most valuable fantasy players over the last 30 years should be a huge red flag.

But my original point was that common sense tells us PPR is unfair to running backs in dynasty, and on that the data couldn’t possibly be any clearer. Once again, common sense is wrong and uncommon sense carries the day.

If you find yourself deciding between a running back and a wide receiver in a PPR dynasty league, ask yourself “which one of these is a better player?” Odds are great that that’s the guy who is going to have more fantasy value.

And if you find yourself starting a dynasty league and wondering what a good scoring system might be, PPR makes a lot of sense. Just be sure to do something to help make quarterbacks more valuable, too, unless you like the idea of Tom Brady being less useful to a fantasy team than Keyshawn Johnson.

Follow @AdamHarstad