Determining Asset Value

mikereardon

hyde

Back in February, I wrote an article that examined the correlation between a player’s NFL Draft round and their production over the first three years of their career. That article focused on running backs and I’ve always intended to do a follow-up article with the other positions. I return to you now in the wake of the NFL draft to finish those thoughts.

Since it has been awhile, I’m going to re-explain what I was looking to do and the methodology I used.

The idea came from the fact that these days, when it comes to the NFL draft, there are a thousand “experts” and “draftniks” out there, and at times it feels like there’s too many voices. It’s difficult to know which of them, if any, you should trust. There are the film junkies, the information guys and the metrics crowd, but nobody seems to have it down pat.

Personally, not fancying myself an amateur scout myself, I tend to give a great deal of deference to the professionals. While I don’t follow the NFL in lockstep, I do tend to think a first round wide receiver has a lot better chance of working out than a third round receiver. I think most people would agree with that as it’s not exactly a ground-breaking notion. But I still thought it might be useful to put together a study and try to nail down what trends and correlations there are between a player’s NFL draft position and his fantasy value, in the context of picking a player in your dynasty rookie draft.

To determine how to approach this problem, the first thing I had to decide was how to define “value.” Few words are tossed around more in dynasty circles than this one, but interestingly, I’m not sure we all put the same meaning into it. Personally, I’ve always carried around the idea that there are two very different types of dynasty value: production value and asset value.

This is another not-ground-breaking concept, it’s simply putting into words an idea we all know. Production value is, simply, how much production (i.e. fantasy points) I expect a player to produce. Is Percy Harvin going to be a WR1 or a WR2? How much of a positional advantage will Jordan Cameron give me? Things like that.

Asset value is thinking in terms of each player being a dynasty commodity that you can buy and sell. What could I get for Trent Richardson right now? What would I be willing to give up for him? If I sit on Vincent Jackson for another year, what will I be able to get for him next year?

Of course, the real fun in dynasty leagues is trying to decide how these two values should interface, which is when you get to the really tough questions. What will I be able to get for Marshawn Lynch a year from now? Is his 2015 production value worth taking the hit on his 2015 asset value? Is the rest of my team ready to win a championship right now?

There are players who have high production value, but low asset value (Andre Johnson), high asset value but low production value (Christine Michael), players who have high levels of both (A.J. Green), and of course, players who don’t have very much of either (Eddie Royal).

For the purposes of answering the question of what the relationship is between value and NFL draft position, I decided to focus on asset value. This is because you can make a good dynasty rookie pick without having a lot of production right off the bat. The aforementioned Christine Michael is a good example of that. He has not produced any fantasy points, yet his asset value has increased since he entered the league. To me, so far, that makes him a successful draft pick because as an owner, I could turn him into something today that is more valuable than what I paid to get him. If we only looked at production value, Michael would look like a severe bust.

Next, I decided to limit the scope of my review to the first three years of a player’s career. This is a somewhat arbitrary window, basically chosen because I feel like you know whether or not you made a successful rookie pick after the first three years of their career.

To me, average draft position data is the best way to measure asset value because it reveals how the market (i.e. the pool of drafters) felt about a player at a given time. Unfortunately, it’s very difficult to come by a large sample of dynasty ADP data for the time period I wanted to look at, so I ended up using redraft data. I actually did not mind this because I think redraft data offers a perfectly legitimate snapshot of what a player’s perceived value was at a given time. If someone is drafted as the fifth overall running back in 2008 in a redraft league, obviously he was expected by the general fantasy community to be a stud.

More importantly, I was looking to do a like-for-like comparison between position groups drafted in different rounds. That I was using redraft data to make the comparison was not entirely relevant. The important thing is I used a common measuring stick, imperfect as it may be.

Of course, our children’s children’s children will not have this problem because they will have decades of monthly DLF mock data thanks to Ryan McDowell, who I’m sure still be pumping those drafts out long after I’m gone. But for now, I can only work with what I have.

So, with that backdrop in mind, this is how I put together the data used for this analysis:

  1. I pulled NFL Draft information for all quarterbacks, running backs, wide receivers and tight ends drafted in the years 2004-2011.
  2. I pulled redraft average draft position data (ADP) from MyFantasyLeague for the years 2004-2013.
  3. I merged those two lists to create a list of every player drafted over this time period along with their ADP from years one, two and three of his career.

[am4show  have=’p2;p3;p4;p5;p6;’  guest_error=’sub_message’  user_error=’sub_message’ ]

All of the included players were then grouped together by their position and the NFL draft round they were taken in. Names and identities became irrelevant. There was no “Matt Forte” or “JJ Arrington,” only “Second Round Running Back.”

I then put that data together in a convenient way to compare it side by side. The table below is the result. I’ll soon put it in another format that is easier to look at and understand, but it’s important that I do a quick walk through of how the sausage is made first.

chart1

Position Group – the group of players whose data is reflected in each row, which is broken up by position and NFL draft round. The first row is running backs drafted in the first round, second is wide receivers drafted in the second, etc.

3 Year ADP – the cumulative average ADP for every player in that position group over the first three years of their careers. Since we’re using redraft data, this number in itself is not very meaningful, we’re only interested in how it compares to the same number of another group.

Draft % – the percentage of players in that position/round group who were drafted at all. For the purposes of this study, “drafted” means finishing in the top 220. This is not a terrible impressive feat in itself, but it’s important to include so that each group van be viewed on a level playing field. For example, if you only looked at the average ADP figure alone, it would look like fourth round tight ends have a better outlook than third round tight ends as their average ADP’s are134.81 and 169.85, respectively. This, however, is misleading when you consider that third round tight ends were drafted 41% of the time whereas fourth round tight ends were drafted only 15% of the time.

SMS Score – The SMS score is simply a snazzy-metric-y-sounding figure that takes into account both the average ADP and the draft frequency. The number itself in a vacuum is meaningless; it is only useful as a means of comparing a particular position group to another one. We will focus on this figure when comparing each position group as it provides an even representation of the asset value of each of the position groups.

Since we’re not all that interested in the actual ADP figures, we’re going to focus in on the SMS Rating of each position group and use that as are primary means of comparison.

Below you’ll find a visibly pleasing chart that shows the SMS score of each position group ranked from highest to lowest. I’ve color coded the bars so you can also easily look at SMS scores within a particular position.

barchart

Well now that we’ve come this far and have an aesthetically pleasing pastel chart, what conclusions can we draw from this? I sure hope there are some because frankly, I’ve spent more hours than I want to count putting this all together.

First round running backs – Wow. 

So clearly the first thing that stands out is that the first round running backs SMS score towers over the field. This is expected, unsurprising, and frankly not entirely useful since first round running backs are going the way of the unicorn. Unicorns existed once, right?

In any case, suffice it to say that if a running back is drafted in the first round in the future, you are going to get a pretty good shelf-life with respect to his asset value. The best example I can think of from recent times (not included in this data because he was drafted too recently) is Trent Richardson. Richardson had a modestly successful rookie season and instantly became a top five start-up pick. He followed that up with an extremely terrible sophomore season, putting together some really bad film. Yet in most leagues, you can still get a pretty good exit price for him. As of the most recent DLF mock ADP data, Richardson is going at 51.30, ahead of a lot of attractive running backs who were not terrible in 2013. What you’re seeing here is the long reach of draft pedigree putting a floor underneath his asset value and slowing his decline. If he has another bad season, that foundation will start to crumble, but the fact that it has held him up this long is notable.

There’s a strong correlation between NFL draft round and SMS score.

As you would expect, the higher the player is drafted, the higher his SMS score. It’s still nice to see the data work out that way because it’s a sign the methodology is capturing what I want it to and it is notable to see how stark the difference is in some cases. The SMS score of first round wide receivers is almost double that of second round receivers (7.91 to 4.19), which is relatively similar to the percentage difference in second round wide receivers to third round ones (4.19 to 1.70). Interestingly, the asset value of third and fourth round receivers are very similar, which is at least in part due to Brandon Marshall and Cecil Shorts pulling up the fourth round receiver rating.

It’s not really fair to compare quarterbacks to running backs and receivers simply because their ADP is going to be lower due to the lack of positional demand in fantasy. For this position, it’s best to focus on the differences within the quarterback position by focusing on the green bars and recognizing first round quarterbacks have a much better chance to bring you sustained asset value than later round quarterbacks. This may not be notable in standard dynasty leagues, but you A.J. McCarron and Zach Mettenberger owners in deep, two quarterback or super-flex leagues might want to take notice. The odds are not in your favor.

Mid-late round running backs fair far better than any other position. If you add up the SMS score of running backs drafted in the third, fourth, and fifth rounds, you get 10.31. The total for tight ends (3.85) and wide receivers (3.89) drafted in those rounds are far behind. This would be in part due to running backs simply being more coveted than the other positions, but it’s also reasonable to infer it’s more common to find mid-round running backs who end up being successful, which is in line with today’s thinking about the position.

Overall Trend = Defer to the NFL

We all have favorite sleepers or players who we thought would go high in the draft that drop. Many of us hold on to those pre-draft feelings and target players who dropped in the NFL Draft in our rookie draft. There’s nothing wrong with this – if you know a player well and have a strong feeling about him, you should trust yourself. But you must go into the situation with eyes wide open. If you’re taking a fourth round receiver over a second round receiver, according to this data anyway, you are taking a massive leap of faith. Again, the measuring stick I’ve used here isn’t completely perfect, but in a like-for-like comparison based on the aforementioned methodology, in that particular example, you have twice as much expected asset value taking the second round receiver.

Late round NFL picks should be late round fantasy flyers. If you’re in a non-salary cap contract league with deep rosters, by all means, burn your fourth and fifth round rookie picks on a receiver who was taken in the fifth round. In that format, it will cost you very little. However, when you consider leagues with shallower rosters, say 25 or so, then the opportunity cost of holding these late round players gets too expensive. This cost is increased if you also have salary caps in your league, or a total contract years cap. In those cases, you’re investing a draft pick, a roster spot, and other capped resources to hold a player that is very unlikely to ever bring you a positive return.

One Last Look

Before I finish this off, I’m going to show one more way of looking at this data that might help put some additional context around what we’re examining. Below is a table of how often a player from a given position group registered a redraft ADP in the top 100 over the first three years of his career. Unlike our conversation earlier, for this example, we are going to look at this data through the redraft prism, just to give us an additional way to view the data. A top 100 player, generally speaking, will be someone you expect to start for your team.

chart2

Once again, running backs do fairly well, particularly first round running backs. It’s a little suprising to see second round receivers get drafted in the top 100 only 10% of the time. Of the 33 receivers drafted in the second round over the years I looked at, only seven ever cracked the top 100 in their first three years: Reggie Brown (2), Eddie Royal (1), DeSean Jackson (2), Greg Jennings (1), Torrey Smith (2), Vincent Jackson (1) and Randall Cobb (1).

The picture gets even bleaker when we go down to third round receivers. Of the 43 receivers taken in the third round over this time period only two, Mike Wallace (3) and Eric Decker (1) ever posted a double digit redraft ADP. The percentage is so bad that Marques Colston, all by himself, was enough to elevate the seventh round receivers ahead of those taken in the third round.

For the most part, this data speaks for itself. Again, it is “only” redraft data, but I would still keep it in mind when you’re making your rookie choices. Based on this data, don’t expect Cody Latimer to make an immediate impact. And that said, as much as the hit rate for second round receivers is surprisingly low at 10% – it’s more than double that of the group that Andre Williams and Devonta Freeman are from. If Lache Seastrunk does it he’ll have done something that no other sixth round running back accomplished between the years 2004 and 2011, though Alfred Morris did break through recently.

Closing Thoughts

Overall, I think the results of this investigation support the idea that you should, to at least some extent, defer to the NFL. The earlier a player was drafted, the longer the shelf-life of his asset value is likely to be. That doesn’t mean the NFL is always right of course – as I said before, I’m not huge into college scouting, but even I knew Ted Ginn Jr was fantasy fool’s gold and Darrius Heyward-Bey was hugely overdrafted by the Raiders. If you have a strong feeling about someone and it contradicts how the NFL “market” valued a player, I’m not saying you should ignore that. I’m simply saying you should be fully aware of the trends you are going against when you go with your gut.

Would I be shocked if Allen Robinson has a better fantasy career than Odell Beckham Jr.? No. But I wouldn’t bet on it.

As always, thank you for reading, and if you have any questions, criticisms, or opinions on who the best Ninja Turtle was (Donatello), feel free to hit me up on the Twitter @mjreardon62.

[/am4show]