True Fantasy Points: Week 9

It's getting close to the trade deadline in many leagues, so now is an ideal time to roll out True Fantasy Points (TFP) for the scarcest position in fantasy football: running backs (RBs). As always, the question TFP answers is "How much is a RB over- or underperforming his 'true' fantasy scoring ability?" Armed with this number, we then use our human reasoning skills to decide whether or not we should target that RB as a buy-low/sell-high opportunity. Pretty simple.

You know what's not simple? Actually calculating TFP for RBs, that's what. This is because (a) RBs derive a larger share of their fantasy points from a larger variety of categories than do other offensive positions, and (b) the various categories have different "units of observation." To show you what I mean, let's walk through the TFP14 calculation for DeMarco Murray.

Here are all of the pieces of information we need to know:¹

1. In 2014, Murray has 206 rush attempts, and has run 142 routes as a receiver.²
2. A run by Murray has averaged 5.12 yards per attempt (YPA) and resulted in a TD 3.40% of the time (TD%).
3. Murray's receiving routes have averaged 1.28 yards per route run (YPRR), resulted in a reception 18.3% of the time (RPRR), and ended with a TD 0.0% of the time (TDPRR).
4. In 2014, the league averages for these stats are 4.25 YPA, 2.89 TD%, 1.31 YPRR, .161 RPRR, and 0.0066 TDPRR.
5. It takes the following number of observations for each of these stats to stabilize:
• YPA = 1,978 carries
• TD% = 667 carries
• YPRR = 308 routes
• RPRR = 169 routes
• TDPRR = 1,063 routes

tfp through week 8: running backs

Now that I've showed you how the sausage is made, below is a table showing TFPs and +/- stats for Footballguys' Top 72 RBs Forward either by standard FBG scoring or by PPR scoring (sorted by +/- in PPR leagues):

Three weeks ago, fellow Footballguy Chase Stuart wrote about Antone Smith's historically unseen ability to conjur up 70-yard TDs every five or six touches. Well, not much has changed in the interim. Of the RBs in the table, Smith's .075 TDPRR, 7.42 YPA, and 10.5 TD% all rank second. Remember, the league averages for these stats are .0066 TDPRR, 4.25 YPA, and 2.89 TD%! Now, also remember that they're the three RB fantasy stats that take a really long time to stabilize. The result: Smith has the highest +/- in the league -- by a mile -- despite having yet to reach 60 combined carries and routes run. In short, TFP says that Smith's point total so far is 25% skill and 75% luck, which you don't need me to tell you is unsustainable. 

standard buy-low candidate #1: lesean mccoy

McCoy is Bizarro Antone Smith: By any of the metrics in the table, McCoy is by far the most underperforming fantasy RB this season. One obvious explanation -- especially in the passing game -- is the presence of Darren Sproles. However, even if we focus on TFP14 in a standard scoring system, McCoy's still underachieving by twice as many points per game than the next biggest underachiever. His 4.21 True YPA and 2.52 True TD% means his 138 carries this season should have produced 581.0 yards and 3.5 TDs, not 505 yards and 1 TD. That 22.6-point difference would still place him at the bottom of the table -- and it doesn't include receiving whatsoever! 

standard buy-low candidate #2: jerick mckinnon

This one's easy. McKinnon has 0 TDs in about 180 carries and routes run combined. However, based on 2014 league averages, his True TD% is 2.60% and is True TDPRR is .006, which translates to an expectation of 4.7 rush TDs and 1.1 receiving TDs. Given McKinnon's small stature, some of this is understandable; given his skill set, the rest of it isn't. 

standard sell-high candidate #1: arian foster

We can make the inverse argument for Foster: He's been a fantasy overachiever so far in 2014 due to an inflated TD% and an inflated TDPRR. Again focusing on TFP14 -- this time because of an offseason coaching overhaul in Houston -- Foster's 4.79% TD% and .013 TDPRR are well above his "true" ability in the context of 2014 only: 3.23% and .007, respectively. In addition to the high likelihood of regression to the mean (RTM) in his TD rates, Foster's 5.25 YPA is probably unsustainable as well. 

standard sell-high candidate #2: lamar miller

The case for selling Miller low is akin to the case against Foster. Miller's 4.29% TD% is a full percentage point higher than his True TD% for 2014 only (3.06%) and even more outlying if we consider all of his carries and routes run in the Dolphins offense (2.71%). It's also the case that Miller's 4.94 YPA this season is at least a half-yard higher than his True YPA, regardless of which version you look at.

Also worth noting: Miami has the second-toughest remaining schedule for fantasy RBs.

ppr buy-low candidate: frank gore

Get this. Gore's True RPRR over the past eight seasons in the 49ers offense is .131; his True RPRR in 2014 only is .117. Perhaps you think the latter's a better indicator given changes in San Francisco over time. Fine. His current RPRR is .041. At his current rate of 14 routes run per game, putting all of this together means that Gore should catch about 15 passes the rest of the season (i.e., nearly four times as many as he has so far).

His actual TD% through Week 8 (0.98%) is also well below his True TD%, which is 2.77% if we apply RTM to his 49ers career or 2.64% if we apply RTM to his 2014 stats only. Either way, he should have around 10 points more from rush TDs.

prr sell-high candidate: ahmad bradshaw

I sound like a broken record, but most of Bradshaw's overachievement is due to a .041 TDPRR, which is nearly seven times the league-average rate. Per his True TDPRR, he should have 1.6 TDs, not 6. To put this further into perspective, if Bradshaw were to not catch a TD in the second half of the Colts' season over the same number of routes run, he would still end the season having scored a TD three times as frequently as a league average RB. As he's scored nearly two-thirds of his total fantasy points via the passing game, this is a bigger deal for Bradshaw's future value than it is for most other RBs. 

¹ To calculate Murray's TFP, rather than using only his 2014 stats, we would use his aggregate stats in the Cowboys' offense alongside the league averages for all RBs since they've been in their current offenses.

² Routes run statistics provided by Pro Football Focus.

Follow @IR_DannyT