«

»

Aug 21

Wide Receiver Draft Strategy: controlling for outliers when assessing value

We spent some time last week looking at how outliers – in this case, big statistical outputs that stand far apart from a players average (or mean) performance – should impact our assessment of Running Backs. Today it’s time for the wideouts as an addition to articles we’ve already shared about assessing TD Ratio and YPRR/Catch Rate in relation to a pass catcher’s fantasy value.

We’ve been fortunate enough to find a kind soul to help with the data, improving some of last week’s figures and sharing updates in a detailed excel file. A big thanks to Ryley Henry who has shared a very thorough and well reasoned assessment of said data on his own site; Not So “Fantasy” Football.

Having a high number of outlier performances in one season isn’t necessarily predictive of the following year’s outcomes, and for some owners it’s so nice to have a boon week that they can tolerate a handful of low scoring affairs mixed in. For me when choosing a top 12 wideout – we’ve included our consensus top 12 WRs in the data – I’m looking for a consistent week in and week out player. An anchor pick, if you will, who I can count on for solid production each time he’s in the lineup, rather than knowing that a 12.0 FPPG average may manifest itself as 20 points some weeks and four the next. With that in mind we’ll take a look at the outliers and how they should influence our assessment of a given player as part of your wide receiver draft strategy.

Falcons Receivers Roddy White and Julio Jones finished back to back in fantasy scoring last year. Which would you rather own? (Photo: Streeter Lecka/Getty Images).

Falcons Receivers Roddy White and Julio Jones finished back to back in fantasy scoring last year. Which would you rather own? (Photo: Streeter Lecka/Getty Images).

Last season both Roddy White and Julio Jones had impressive fake (and real) seasons for the Atlanta Falcons; with the latter posting 11.4 fantasy points per contest and the former coming in right behind him at 11.1. If you owned either, you were happy. However, when assessing last season’s impact one must consider how players accumulated their points and not just how many of them there were. White gave owners six games with less than six fantasy points but helped his average with two performances well above his season average (300+ yards and four scores during weeks four and 16). Jones had five games with a similarly low output but just one outlier performance. The reality is, with the low scoring games, both players tend to cancel each out a bit as when one has a big week the other may not (they had just two games where both scored fewer than six despite 11 total). Who would you rather own? The contrast isn’t glaring as they finished back to back in overall scoring but for me I’d prefer Jones’ 2012 stats because he fluctuated less from week to week. Our outlier elimination process is explained in detail on the data sheet, but, what we’ve done is replace any performance more than 10 points above their season average with their average per game performance to calculate a new figure. When adjusting for these games that deviate from the mean the gap between the players broadens significantly, with Jones scoring 10.7 points per game and White posting just 8.9.

Now that we’ve demonstrated the reasons for removing these big figures, we’ll take a brief look at the takeaways, though I leave the individual interpretations to you.

Share your observations in the comments below. Think we missed something? Tired of math class? Rethinking draft rankings? Let us know.

  • Calvin Johnson remains awesome. He gave us a couple of low output games but because a) he set his per game average so high and b) he didn’t have a ton of touchdowns ,he avoids a big outlier penalty (with just one such effort). 
  • Seeing Dez Bryant in preseason I’m so excited that I could make a reasonable argument for ranking him at 1b, and I’m certainly feeling validated in placing him as my WR2 in our consensus rankingsbut it should be noted that he does not benefit from this particular lens. Last year Dez had three really unimpressive games, being held under two fantasy points in each of them and he posted two massive games. When we adjust for these his per game average drops a full two points.
  • AJ Green was consistent. He had two outlier performances early (the last one in week six) and two ‘dud’ performances late (week 14 and 17). It’s worth noting, though, that that trend is in the wrong direction for a player coming to the end of his second season.
  • Brandon Marshall and Demaryius Thomas were remarkably consistent, the former having one outlier and the latter none and each posting 10 double digit fantasy point outputs. Given the minimal number of outliers their averages don’t change a great deal – which is really what you’re looking for in an early round receiver.
  • Victor Cruz had two 23 point efforts that affected his average significantly when removed. He’s a hard player to project this season – but the ‘outlier’ performance, the occasional monster – has been a staple of his game the last few seasons.
  • Larry Fitzgerald‘s numbers can safely be ignored.
  • Randall Cobb was not hindered by the adjustment greatly… which is a bit of a surprise given his role as a slot receiver. You expect guys who rely on speed and agility to have more big games but also more where they don’t factor as significantly into the game-plan. His reasonably consistent campaign bodes well for the outcome of his projected larger role in Green Bay’s offense.
  • Vincent Jackson did have six ‘duds’ but only one outlier performance and still managed to finish as the 6th best Wide Receiver. Andre Johnson had just the one outlier as well – a full 21+ points greater than his average – and is being discounted significantly in most draft rooms.

So there you have it – none of this is a reason to take a player off your draft board, and if you anchor the position with a player who has low variation from week-to-week then perhaps a player like White with clear week-to-week homerun potential becomes more appealing. Still, you can’t bank on the big weeks and that’s why I’m looking for high-weekly-floor receivers in the early rounds.

Leave a Reply

Your email address will not be published. Required fields are marked *