Dynasty, in Practice: Valuing in the Face of Uncertainty

An example of how to value assets in uncertain situations, and more thoughts on Week 2's action.

The Big Takeaway

Events this week provide a fantastic opportunity to give you a peek behind the curtain on how I try to value players in dynasty leagues. I’ve written before about how value is a function of production minus opportunity cost. To quote:

Accounting for these three main opportunity costs, we come up with the basic rules of player value in dynasty leagues. 1) The more points you score relative to your peers, the more valuable you are. 2) The fewer games required to score those points, the more valuable you are. 3) The less time you will spend tying up a roster spot, the more valuable you are. A player's "actual value" is a function of these three rules.

Valuing a player, therefore, is simply a matter of making assumptions for all three points. The better your assumptions, the better the resulting value. So what are the best possible assumptions we can make regarding Adrian Peterson?

To start with, the easiest assumption is about Peterson’s value per game. The base for our assumption can be Peterson’s historical performance. Pro-Football-Reference.com, in addition to being the best resource on the web for historical player stats, tracks historical fantasy production at the bottom of every player page. It even includes a calculation for the player’s VBD in each of his seasons. PFR’s VBD values aren’t perfect, but here at Dynasty, in Practice we are big fans of never letting the perfect be the enemy of the good. According to PFR, Adrian Peterson has averaged 107 VBD per season for his career. Remember to my discussion of player value, though? Raw VBD doesn’t account for the opportunity cost of games played; a player who scores 100 points in 10 games is more valuable than one who scores 100 points in 16 games. Not to despair, we can adjust Adrian Peterson’s VBD values for games played. It’s not as simple as dividing VBD by games played- we have to pro-rate production over a full season, find VBD, and then divide that total by games played. Luckily, I’ve already run the numbers. Over his entire career, Adrian Peterson has averaged 8.04 points of VBD per game played in standard FBGs scoring, which would translate to 128.7 VBD per 16 games. That value would rank in the top 5 in every season of Peterson’s career except for one. Adrian Peterson’s worst season was still 6.25 VBD per game, which would be 100 VBD over a full season. That total would have ranked in the top 12 in every year of Adrian Peterson’s career. Basically, I’m just putting numbers on what we already knew- Adrian Peterson has been an elite fantasy asset his entire career, and even his very worst season to date resulted in top-12 overall production.

As the saying goes, though, we don’t get to count last year’s points. This is where the assumptions come in. First of all, we have to make assumptions about how many points of VBD per game Adrian Peterson will score going forward. I like to approach this question by looking at how running backs similar to Adrian Peterson have fared. One method might be to say that Adrian Peterson is a Hall of Fame talent, so let’s compare him to how Hall of Fame backs have done at age 29 and older. One problem with this method is that it is subject to selection bias. Many backs are only in the Hall of Fame because they were so great after age 30, so we should naturally expect Hall of Fame RBs to have been exceptional late in their careers. A better method is to compare Adrian Peterson to all running backs who were similar through age 28, which we can do using the Historical Data Dominator. Here’s a list I compiled of the top backs in history in terms of fantasy points from age 22 and 28. How did those backs fare going forward? The top-5 non-Peterson backs in history were Tomlinson, Smith, Faulk, Sanders, and Payton. If we calculate VBD on a per-game basis, those five backs totaled 1361 more points of VBD for their careers, or 272 points each. All of them produced at least 100 more points of VBD. The worst of the bunch was Tomlinson, who had 45 more value-adding games with an average of 3.1 VBD per game over that span. The best was Payton, who had 73 more value-adding games with an average of 7.7 VBD per game over that span. All five backs produced positive value at both age 29 and age 30, and all but two produced positive value at age 31, (the exceptions are Barry Sanders, who retired before his age-31 season, and Marshall Faulk). In total, they produced 15 seasons of positive value, for an average of three each. The five backs totaled 12 more seasons of 3+ VBD per game, which is the equivalent of a 48-VBD season. In recent years, 48 VBD would have ranked an RB as high as 7th or as low as 15th, and is essentially a high-end RB2 season. The five backs totaled 9 more seasons of 6+ VBD per game, which is the equivalent of a 96-VBD season and is usually the equivalent of a mid-RB1 finish. There were even two seasons of 9+ VBD, which is a dominant fantasy season; Adrian Peterson, despite all his success, has only topped 9 VBD per game once in his career, during his magical 2012 season. The average VBD per game across all 15 usable fantasy seasons was 6.1.

So based on that particular list of comparisons, Adrian Peterson seems like a safe bet for 2-3 more years of usable fantasy production, including 1-2 years of quality RB1 production. If we want to expand the list of comparisons to include the top 10 non-Peterson RBs (which adds Thurman Thomas, Eric Dickerson, Curtis Martin, Shaun Alexander, and Edgerrin James), the averages go down. The average number of usable seasons remaining falls to 2.3, the average number of 3+ VBD seasons falls to 1.7, the average number of 6+ VBD seasons falls to 1.0, and none of the new backs ever managed to reach 9+ VBD per game again. In total, the sample of the top 10 non-Peterson backs averaged 183.6 more VBD starting at age 29, at a rate of 5.4 VBD per game. Each back still managed to produce at least one more usable season on a per-game basis, but three of the five, (Alexander, James, and Dickerson), only had one.

Whichever list of historical comparisons we use, it’s clear that players who produced like Adrian Peterson through age 28 have traditionally remained quite productive and valuable for a few years afterwards. If we had no other concerns, my assumption would be that Adrian Peterson would put up about 200-250 more VBD over the next two to four years. And if “ifs” and “buts” were candy and nuts, we’d all have a merry Christmas. With the Minnesota Vikings announcing that they will place Adrian Peterson on the exempt/commissioner permission list, it’s quite clear that Adrian Peterson is going to be missing substantial time. To value him, we must again make assumptions about how much time he is missing. One possible resolution to the Peterson case is that he pleads guilty and incurs a 6-game suspension under the league’s new domestic violence policy. Another possibility is that he sits out the season, goes to court, and wins his case, incurring no future penalty. A third possibility is that he sits out the season, goes to court, and loses his case. In that case, we would likely see Peterson earn a 6-game suspension next season. Or, of course, the NFL could decide that the factors of the case warrant a suspension longer than the base 6-game suspension called for under the new domestic violence policy. Or the league might decide that Peterson’s case doesn’t fall under the domestic violence policy at all and resolve it under the personal conduct policy, instead. Obviously there are a lot of ways Peterson’s case can go, and anyone who tells you they know how it all will play out is lying. I’m not going to lie to you. I will instead lay my assumptions out and let you judge them for yourself.

I am assuming that, given the high-priced and highly-regarded lawyer he has retained, Adrian Peterson has little interest in accepting a plea bargain. I also assume that, given the DA’s willingness to bring this case to a grand jury twice in search of an indictment, the state has little interest in offering Peterson a plea bargain. Given the outcry that resulted when the Vikings announced they were activating Adrian Peterson, I suspect they will be leery about trying again this season. Therefore, I think the most likely outcome is that Adrian Peterson sits out the remainder of the 2014 season on the exempt/commissioner permission list. Given the quality of his legal representation, the prevailing attitudes towards corporal punishment in eastern Texas, the fact that the first grand jury to hear the case declined to indict Peterson, and the concerns surrounding how the details of the case leaked and biased potential jurors, I think the most likely outcome of Adrian Peterson’s legal case is acquittal. Should Peterson be acquitted, my assumption is that the league will not punish him further. Should Peterson be found guilty, my assumption is that the league will give him a 6-10 game suspension under the new domestic violence policy. Given my assumptions about the likelihood of Peterson being found guilty and the consequences of Peterson being found guilty, I would bake an expectation of about 2 games missed to suspension into his 2015 value. I also assume that the Minnesota Vikings will part ways with Adrian Peterson after this season, given both prevailing attitudes towards corporal punishment in Minnesota and the high amount of cap savings that could be earned by Peterson's release.

So, these are my final assumptions: Adrian Peterson will miss 16 more games as a result of his court case, he will be cut from Minnesota, he will be signed by a contending franchise (likely one in the south, where the fans have more permissive views towards corporal punishment), he will produce 1-3 more seasons of positive fantasy value for his new franchise, and when all is said and done I would expect his remaining career VBD to be somewhere in the 100-150 range. And what, exactly, is 100-150 expected VBD worth? Well, based on a whole other series of assumptions, I find that that’s about the value of a mid-to-late 1st round draft pick. And so, in my most recent rankings update, Adrian Peterson finds himself settling in right in front of Carlos Hyde and Bishop Sankey, who I had rated as mid-to-late 1st round picks in the most recent rookie draft.

Obviously the quality of that ranking is only as good as the quality of the assumptions underlying it. I have done my best to base these assumptions on the best and most detailed information I have available. I am constantly re-evaluating my assumptions and trying to improve them. Given the information available today, this is the best estimate I am capable of providing. As more information comes in, my assumptions will change, and so too will Peterson’s ranking. For the moment, though, this is why he ranks where he does.

Producing a set of rankings requires making a whole host of assumptions about every player, many of which are entirely unsubstantiated, most of which will prove to be wrong. Hopefully by giving you a window into my ranking process, you can see how dynasty values are determined and begin using these principles to form your own opinions on player value. In addition, hopefully by laying my assumptions bare, you can see where you agree and disagree and mentally adjust my rankings upward or downward to match.

Second Thoughts

I mentioned last week that Ray Rice was worth rostering until we found out for sure whether his indefinite suspension was going to hold up. One week later, we’re hearing positive news on that front, as the NFLPA has decided to file a grievance on Rice’s behalf. While a successful appeal would still leave Ray Rice without a team, and the negative publicity would still serve as a powerful deterrent for anyone thinking of signing him to play this season, the grievance is still a positive development. If reports that the NFL had seen the video from inside the elevator are accurate, Rice has an excellent chance of winning his appeal on the merits, which would leave him free to sign with any team at any time. With a year off to build positive press, perhaps by next season Ray Rice would again be an attractive free agent for an RB-needy team. At just 28 years old, Rice would probably have enough left in the tank for another season or two of fantasy relevance. If Rice is not rostered in any of your leagues, he’s worth adding strictly based on fantasy considerations. Of course, if you’d rather not roster an NFL player who has punched his wife and dragged around her unconscious body, that’s a perfectly valid reason not to add him. Ultimately, we play fantasy football because it is fun, and we must manage our own teams with that in mind.

Those who have been reading for a while know that I am a big fan of buying injured players. I believe that most owners don’t have enough faith in modern medicine, and I believe that it’s human nature to overvalue immediate returns over long-term gains. This means that, as an asset class, I believe that injured players tend to be underrated by the market. Obviously last week’s Injuryocalypse means that this week has been a busy week for offers in my leagues. Let’s run through a few of the highlights.

Robert Griffin III and Jamaal Charles were probably the two biggest injuries of the week. They also perfectly illustrate my belief that people overrate injuries and underrate modern medicine. The immediate reports after the games suggested Griffin had a dislocated ankle and could be done for the season, while Charles had a high ankle sprain and should be out several weeks. Instead, as the week has gone on, we are hearing that Griffin’s injury might be a subtalar dislocation and he could be back in as little as four weeks. Meanwhile, Jamaal Charles has already returned to practice and looks likely to play this week. Who knows if these injuries will linger or how they will affect these players going forward, but anyone who had bought Charles or Griffin immediately after the injury probably could have gotten a nicer discount than anyone looking to buy Charles or Griffin today. Assuming that initial risk very early in the process when we do not have much in the way of details is often a very good way to earn some extra profit on your transactions, and even if the worst-case scenario comes to fruition, there is still always a good chance for you to make up your investment in the future. Dynasty is a long game, so don’t make moves with a short view.

Other players dealing with bumps and bruises include A.J. Green, Ryan Mathews, Mark Ingram, Knowshon Moreno, Desean Jackson, Vernon Davis, Jordan Cameron, Jordan Reed, Tyler Eifert, Doug Martin, Eric Decker, and Tavon Austin, among others. None of these players’ injuries are serious enough to rise to the level of career-threatening. You can be sure I will be inquiring as to availability with all of their owners in all of my leagues. Many will undoubtedly be unwilling to offer a discount, but all it takes is one panicking owner to turn a nice profit. I’ve built championship squads on the backs of ACL tears (Adrian Peterson and Jamaal Charles), Achilles ruptures (Demaryius Thomas), neck injuries (Peyton Manning- whose injuries certainly were “career threatening”, but whose discount was substantial enough to match the risk), and various other maladies. I have rosters stocked with the likes of Michael Crabtree and Percy Harvin and Rob Gronkowski and Jordan Reed, who I anticipate joining the list of title-winning players who were purchased at a discount.

How should we value all of these injured players? Let's just say that the topic of today's Big Takeaway was no coincidence. Valuing injured players is a process of making assumptions. If our assumptions are better than the market's, we will win out in the long term. Many value fantasy players based solely on gut feel. Dynasty leagues are more of an art than a science, and there is a huge place for gut feel; however, that place is at the end of a process built on logic and reason and not the beginning.


More articles from Adam Harstad

See all

More articles on: Dynasty

See all

More articles on: Player Articles

See all