There's a lot of really strong dynasty analysis out there, especially when compared to five or ten years ago. But most of it is so dang practical-- Player X is undervalued, Player Y's workload is troubling, the market at this position is irrational, take this specific action to win your league. Dynasty, in Theory is meant as a corrective, offering insights and takeaways into the strategic and structural nature of the game that might not lead to an immediate benefit, but which should help us become better players over time. (Additionally, it serves as a vehicle for me to make jokes like "theoretically, this column will help you out".)

Dynasty Is A Game of Pricing Risks

Everything about Dameon Pierce's trajectory suggests he's going to be a star. He received effusive praise for his work during training camp before earning the starting job in advance of Week 1. He quickly dispatched all trusted veteran challengers and has seen his role increase until he became a clear every-down back before midseason. Through his first eight starts, he has 678 rushing yards (at 4.6 yards per carry), and while he's not really an asset as a receiver, he's added 98 receiving yards (albeit at just 4.9 yards per reception) and scored a total of 4 touchdowns.

Everything suggests he's going to be a star... except for one little thing. Dameon Pierce was drafted in the 4th round.

In 2013, everything about Zac Stacy's trajectory suggested he was going to be a star. He received effusive praise for his work during training camp, and while he didn't earn the starting job in advance of Week 1, he earned his way on the field in Week 5, quickly dispatched all trusted veteran challengers, and saw his role increase until he became a clear every-down back before midseason. Through his first eight starts, he had 692 rushing yards (at 4.4 yards per carry), and while he wasn't really an asset as a receiver, he added 106 receiving yards (albeit at just 6.2 yards per reception) and scored a total of 5 touchdowns.

Everything suggested he was going to be a star... except for one little thing. Zac Stacy was drafted in the 5th round.

Despite his strong rookie performance, the Rams drafted Tre Mason in the 3rd round following Stacy's rookie campaign. Mason followed the "positive signs" career path, too, taking a bit longer to win the starting job from Stacy but finally wresting it away from the trusted veteran and having a strong second half of the season of his own in 2014 (686 total yards and 4 touchdowns in his first eight starts).

And then the Rams pulled the rug out from under Mason, too, drafting Todd Gurley with the 10th overall pick in the 2015 NFL draft.

This is the NFL reality for later-drafted running backs. Plenty of them make good on the flashes of promise and go on to be the stars they seem destined to become. Undrafted Arian Foster became a 6-year starter for these Texans before injuries laid him low. Undrafted Priest Holmes put up one of the greatest stretches in fantasy football history. 6th-round pick Terrell Davis went on to win two Offensive Player of the Year awards and a league MVP with the Denver Broncos.

But plenty of them never earn the sort of long-term trust and support required to flourish like that. Before Holmes became a fantasy superstar, he got rug-pulled twice by the Baltimore Ravens, posting 1200 yards in 1998 before losing out to free agent acquisition Errict Rhett in 1999 and 5th-overall pick Jamal Lewis in 2000. The same Broncos that committed to Davis pulled the rug from under 4th-rounder Olandis Gary after 1300 yards and 7 touchdowns in 12 games in 1999 and from under 5th-rounder Mike Anderson after a whopping 1650 yards, 15 touchdowns, and a Rookie of the Year award in 2000.

More recently, it's hard to imagine a young running back smashing expectations more than undrafted James Robinson, who was so good during training camps and preseason that the Jaguars shockingly cut former 4th-overall pick Leonard Fournette, who had averaged 100 yards per game for them over the past three years. Robinson was the lone bright spot for Jacksonville, posting 1400 yards (at 4.5 yards per carry) and 10 touchdowns in 14 games despite poor blocking and a terrible 1-15 record. As a reward, the Jaguars fired their coach and spent a 1st-round pick on Robinson's replacement.

Is Dameon Pierce the next Mike Anderson, or is he the next Terrell Davis? Neither. He's the first Dameon Pierce. Comparisons from history can help shape our expectations, but ultimately every player is unique and will follow their own path. But it is important to understand the history here; later-drafted running backs do get replaced at a much higher rate than their comparably-productive, comparably-impressive earlier-round peers.

Again, I'm not saying that Pierce hasn't been fantastic, and I'm definitely not saying I don't think he's here for the long haul. Pierce has been wildly impressive, and it looks to me every bit like the Texans have found their running back of the future. I'm just saying that I could be wrong. It's been known to happen. In fact, it happens quite a bit!

And that brings us to today's thesis; dynasty is not a game that rewards managers who are most right. It is a game that rewards managers who are least wrong. Or, put differently: dynasty is ultimately a game of pricing risk.

I’m reminded how much success in dynasty hinges not on evaluating talent better than the next guy, but on pricing risk better than the next guy.

— Adam Harstad (@AdamHarstad) December 18, 2017

Let's imagine two ways of determining player value. The first is very simple: "I believe A is true, and if A is true, this player's value is X". This is a good starting point. I believe Dameon Pierce is the running back of the future for the Texans, and if that is true, he's maybe the 8th to 10th most valuable running back in dynasty right now. Sure, it'd be nice if he was more of a receiver, but a running back this good and this young (still just 22 years old) should be considered a long-term positional advantage. He could be the next Nick Chubb.

Now let's look at a more nuanced method. "If A is true, then this player's value is X. If B is true, then this player's value is Y." In this frame, a player's value is (odds of A) * (value if A) + (odds of B) * (value if B). Let's say I think there's a 70% chance the Texans don't bring in a challenger for Pierce this offseason, and if that's the case, he's the #8 running back in dynasty. That means there's a 30% chance the Texans do bring in a challenge for Pierce, and if that's the case, he's the #20 back in dynasty. (By way of comparison, James Robinson was valued as around the #14 back in dynasty at the end of his rookie season, and that fell to around #30 once the team drafted Travis Etienne before rising back to 20th once Etienne suffered a Lisfranc injury and was lost for the season.)

There are two things to keep in mind about this more nuanced method. The first is that it's always wrong. If you believe Pierce will not face meaningful competition and value him as such, then in the event that he doesn't face competition, you are right, and in the event that he does, you are wrong. Similarly, if you think he definitely WILL face meaningful competition and value him as such, then if he does, you are right, and if he doesn't, you are wrong. Either way, there's a chance that, come this June, your current valuation of Pierce will be "correct", and there's a chance it will be "incorrect".

But if you're valuing him as some sort of Schrödinger's Running Back who both will and will not face competition next offseason you are guaranteed to be wrong. If the Texans don't bring in competition, you're too low on Pierce. If they do, you're too high on Pierce. There is no scenario where valuing Pierce in this way produces a value that is accurate nine months from now. Managers who deal in absolutes will be more right than you are.

The second thing to keep in mind is that this more nuanced method is hard. Really, really hard. It requires us to make blind guesses and handwave estimations about things we can't possibly know. I can say the odds the Texans bring in meaningful competition are 30%. You might say they're 20%. Another person might estimate it at 50/50. Regardless of what happens, we can never know which of those three estimates were "right". In fact, in a meaningful sense, none of them can possibly be "right". Some philosophers dispute the very act of assigning probabilistic estimates to one-off future events.

But researchers have demonstrated that making these sorts of estimates is a genuine skill, and some people will prove more accurate at it than others. For a skilled prognosticator, 20% of the time they say there's a 20% chance of something, it will happen. And 80% of the time it will not.

And provided such well-calibrated estimates, the greatest feature of valuing players this way is that, over the long run, it is the least-wrong way to do it. While you're guaranteed to miss on every player's true value, the amount you miss by will be the least of anyone. And consistently being the "least wrong" means our errors have the least cost to our overall roster. If we were looking to sell Dameon Pierce like the harshest of skeptics, the expected outcome is that we get too little and lose value. If we were looking to buy Dameon Pierce like the most optimistic of boosters, the expected outcome is that we give too much and lose value.

By embracing the happy medium between the two poles (with our exact position dictated by how likely we think each outcome is), we maximize our expected return. And if our risk estimates are better than our leaguemates', over time, we'll find our roster becoming better and better while theirs become worse and worse.