AdviceFootball

Prospect Comparables Are Out of Control – And We’ve Done Our Part to Make it That Way

We’re in peak draft season so the NFL prospect to NFL player comparables are going to be reaching peak saturation in no time. In fact we have and will continue to add to the noise that results from the practice of comparing NFL prospects to NFL players. For that reason (we don’t intend to stop) I thought it might be worthwhile to discuss some areas where I see the comparables process to be problematic, and also where I think it has utility.

Let’s start with the cold hard truth: most NFL prospects will not become multi-year stars in the NFL. The fact that we start our mental similarity searches with the players freshest in our minds due to survivor bias (the multi-year stars are freshest in our minds), is not ideal. The lower that a player is picked in the draft, the higher the likelihood that we will never hear from them again in terms of the player having fantasy value.

Prospect to star comparables reach their most ridiculous when a prospect is compared to an NFL star, but then is noted to be lacking a known shortcoming of the star. It’s possible to hear undrafted quarterbacks compared to “Jay Cutler, but more consistent” – we’ve even engaged in similar analysis, although I will say that I regret that we have. I think the point of that comparison is to illustrate the upside that you see in a prospect, but then to make sure that you don’t turn off readers with your comparison. But stop to think about that comparison for a second. First, you’re starting with a multi-year NFL starter. Then you’re saying that this prospect lacks the thing that is most commonly cited at the starter’s shortcoming. You’re coming close to describing the perfect QB.

In part, the problem with prospect to star comparisons flows from an overall lack of awareness of prospect hit rates. When I post a list of running back comparables almost invariably the response will be “that’s not a great list” because people are not used to thinking about the true odds that a player will hit. Most likely they are driven by the most optimistic outcome.

To further my point from the above paragraph, consider that running back prospects are said to be “the best running back prospect since Adrian Peterson,” even though Peterson himself was the seventh overall selection in his draft while two running backs since Peterson have gone in the top four picks. Why aren’t prospects “the best running back prospect since Darren McFadden and/or Trent Richardson?” Maybe you disagree about the specifics of how those players should be viewed as prospects and would argue that still Peterson was the best prospect of the last 10 years. However, the fact that McFadden and Richardson would never be in the discussion illustrates my point that largely we’re so biased by outcomes when these comparisons are made, that it undermines the entire process. If we can’t accurately remember, or are afraid to recall, how McFadden and Richardson were viewed as prospects, it infects the entire process.

So what is a reasonable use of comps if I’ve spent five paragraphs dumping on them? I think they’re still useful to generate a realistic range of expectations. Primarily I think this should be done by comparing prospects to prospects, rather than prospects to pros. Most of the time that’s going to result in increased sobriety about a prospect’s chances, because you’re going to see names like Braylon Edwards pop up as a comp (whereas the only other time that a player would be comped to Edwards would be if the evaluation was almost entirely negative in advance of the comparable being offered, and totally disregarded that Edwards was one of the most highly regarded prospects at his position). But sometimes it will result in increased excitement for a player because a fourth round pick will end up sharing traits with a number of past second round picks for instance.

Comparables are the natural response to the question, “How successful have other players that share these attributes been?” It’s just that between the asking and answering of the question it’s pretty easy to confuse what you’re even trying to do.

Subscribe for a constant stream of league-beating articles available only with a Premium Pass.

By RotoViz Staff | @rotoviz | Archive

Comments   Add comment

  1. Lots of great points here, and obviously really important for our readers. It's why, for instance, in my Prescott piece, I liked to discuss the comparisons to guys like Kellen Clemens and other players that clearly aren't of note now, but cost higher draft capital than what Prescott will probably go for right now.

    I think in that way, we can find tons of arbitrage opportunities just by saying "This guy will probably go in the fifth round, but looks a lot like other players who went in the second round" regardless of the actual outcome of those players.

  2. I also like that this finally provides a platform for RV to stop giving away the secrets to fantasy success.

    But seriously this was a very helpful article. I'm guilty of looking at comps lists and basically disregarding the bad ones. As you mentioned, I'm guilty of focusing too much on the upside guys. Your point about probability also hit me because now I feel like the dude who was chasing gut shots for my entire stack because I didn't realize how stupid the odds were.

Discuss this article on the RotoViz Forums