Fantasy Basketball Value Investing

As I read The Big Short by Michael Lewis and now watched the movie as well, I became enamored with the story of Michael Burry. In 2000 he started an investment company called Scion Capital, which focused on value investing, buying undervalued stocks and junk bonds. He became famous for shorting the subprime mortgage market by purchasing credit default swaps and becoming rich from it. What are the value investment opportunities present in fantasy?

First, let’s begin by detailing what I’m not referring to. I’m not referring to players such as Lowry, Millsap, and Kawhi who most fantasy experts agreed were excellent players and subsequently exceeded their already high ADP by a little. I’m talking about the trash heap. I’m referring to players who even the experts tend to sleep on.

The value bonds of the fantasy world tend to come in five different forms:

  1. Injury History
  2. Per 36 Darlings
  3. New situation
  4. Flaws that you can tolerate
  5. Rookies

Injury History

Unless someone has a chronic condition, “injury history” isn’t a real thing. It is an issue that experienced fantasy GMs have invented to justify not drafting a player who burned them before. Its foundation is in a reverse Gambler’s Fallacy, the Gambler’s Fallacy being that what happened in the past has a bearing on future events, even when they aren’t connected. From Wikipedia: The gambler’s fallacy, also known as the Monte Carlo fallacy or the fallacy of the maturity of chances, is the mistaken belief that, if something happens more frequently than normal during some period, it will happen less frequently in the future, or that, if something happens less frequently than normal during some period, it will happen more frequently in the future (presumably as a means of balancing nature). It applies only in situations where what is being observed is truly random.

The most basic example is betting on a coin flip. Suppose you have flipped a normal two-sided coin ten times and it has landed on heads ten times in a row. If you poll a random crowd how it will land next, most people will say tails, to balance out what happened in the past. Some people will say heads because they assume it will simply keep landing on heads. Very few people will understand that the proposition is still 50/50 because past events have no impact on future events, assuming situations that are truly random.

The reverse can also be a fallacy in which a gambler may instead decide, after a consistent tendency towards tails, that tails are more likely out of some mystical preconception that fate has thus far allowed for consistent results of tails. Believing the odds to favor tails, the gambler sees no reason to change to heads. Again, the fallacy is the belief that the “universe” somehow carries a memory of past results which tend to favor or disfavor future outcomes.

As applied to fantasy sports, many GMs believe that because a coin has landed on heads ten times in a row that it will continue to land on heads. They believe that because players such as Brook Lopez and Ricky Rubio have been injured in the past they will always be injured. However, if the injury isn’t chronic then the chance of that player being in the future is precisely the same as if that player were never injured at all. Believing that because BroLo has suffered injuries in the past that he is “injury prone” is a reverse Gambler’s Fallacy because his past injuries have no bearing on whether he gets injured in the future, assuming his former injuries have healed, which they have.

This fallacy was the source of great value for me this season as I bet on BroLo in most leagues that I was in. John Cregan at ESPN did some great analysis correlating a player’s z score based on standard deviations to his auction cost. Based on his research, I did some of my own analysis. I don’t want to give my methodology entirely away but it has to do with correlating one’s score in fb-ninja player rater to auction value. In a typical auction league, you are given $200, and for the most part only your top 10 players matter as the others are streamers. Therefore, the average player who receives a score of 0 in fb-ninja in your league settings should be worth $20 because a score of 0 in fb-ninja represents the median of rosterable players in your league. Lopez received a score of .22. Based on my analysis, this represents $31 of auction value. He is worth more if you are pursuing some sort of punt steals/punt assists strategy in H2H formats. His floor in an 8 cat roto league is about $30 and his value goes up from there.

If BroLo is worth $30 at the absolute minimum, what do you think his price should have been in auction leagues? I was able to obtain him for around $25 in most leagues. While this is not an absolute steal, and other players were even better value, it represents an irrational bias against players with injury history.

I am in favor of reducing a player’s value in accordance with risk. However, BroLo is one of the safest bets that you can make. He is a veteran, in his prime, as a franchise player on a bad team. This is the fantasy holy grail. He is one of the safest bets for 3rd round+ value that I can think of. Yet the market pegged him as an average player. The only plausible explanation is an extremely risk averse strategy regarding injuries. However, BroLo does haven’t any chronic injuries. He isn’t like Yao Ming, wielding a massive body with an impossibly small frame. He had a foot injury, he took time off to heal, but it didn’t heal properly. Then in the offseason he had surgery which has a 95% success rate. Even if you include risk of re-injury into his analysis, he is at the absolute least worth $27.

I bring BroLo up only because he was a bet that I made in many leagues, which paid off. He is the paradigmatic example of the market being overly risk averse regarding injuries. The truth is that any player can become injured at any time. Players such as Curry who once had a reputation for ankle injuries, can get over their injuries. Players who have no injury history such as Bledsoe can suffer tragic season ending injuries. Injury risk is inherent to every player. My advice is to make no adjustments to a player’s value based on injury history. That way you can reap the rewards of the market being too risk averse.

Per 36 Darlings

This is a controversial topic because when a player plays 18 minutes per game, they are able to exert themselves in a manner that wouldn’t be possible if they played 36 minutes per game. You cannot simply multiply their production by two and expect to have an accurate result. Conversely, often times a player will produce in a coach’s rotation but only as his last option off the bench. When a player produces brilliantly in a limited role, it’s a good indication that he will be excellent when given a larger role. Recent history is replete with players who became early round value after producing in limited minutes.

NBA coaches aren’t dumb. They know when a player is producing even in limited minutes. Often they are slow to make changes, but the change eventually comes, and the per 36 darlings become amazing assets to own. When that happens-watch out!

A good recent example is Rudy Gobert. Having lived in Utah, I still feel a connection to the Jazz. A couple of years ago Jazz fans were abuzz with rumors about one of the lankiest big men of all time showing flashes of brilliance in limited minutes. Back then almost no one outside of Utah had heard of Rudy Gobert. There was only one problem. Both Derrick Favors and Enes Kanter stood in front of him in the depth chart. The Jazz stood by Kanter for a few months until it became obvious that Gobert’s defense was simply too valuable for him not to start and they traded Kanter. Gobert’s value exploded and the StifleTower was born! Other recent examples include Eric Bledsoe when he was in Los Angeles and Reggie Jackson in OKC. These point guards were excellent when they played off the bench but were buried behind Chris Paul and Russell Westbrook respectively. It wasn’t long before they were traded and have become early round staples.

A good example this season is Gorgui Dieng. Early in the season he was buried behind Karl Anthony-Towns and Kevin Garnett in the depth chart. However, his per 36 line was too good to ignore: 13 PPG/9.4 RPG/1.9 APG/1.7 SPG/1.4 BPG on 52.4 FG% and 83.5 FT%. He’s one of the best two-way players in the game, a quick, lengthy, big man who can pile up stocks while being a good shooter. Many GMs were dropping him as he saw limited minutes early in the season. I actively traded for him wherever I could. I even managed to pick him off the waiver wire in a couple of money leagues! Now look at him. Minnesota finally decided to give him more playing time and he will likely produce top 50 value the rest of the season.

Players with great per 36 numbers playing on restricted minutes are often the best value investments that you can make. Almost every young player started off on limited minutes until they proved themselves so there’s a good chance that any player who is producing in limited minutes will eventually get more minutes. Their value can be difficult to predict because sometimes they don’t produce with more minutes and sometimes they never get more minutes. However, value investing is all about spending a little to bet on long shots with huge potential gain. Players with great per 36 numbers are often the best value investments because they’ve already proven that they can produce, the only question is whether they will continue to produce given the opportunity. Therefore, they are safer than rookies because we’ve already gotten a snapshot of their potential, and the payoff is similar. Spending a little money to see whether they get more playing time is often the best investment you can make, particularly in keeper leagues.

New situation

This is related to the per 36 darlings, but subtly different. Granted, you are hoping that the per 36 darlings eventually get more playing time. However, here I am referring more to players who have produced in the past, perhaps faltered a little, and now have opportunity for renaissance in a new situation.

This can be players such as Reggie Jackson or Ish Smith this year who found themselves in leading roles on new teams. Fantasy GMs know that they must adjust a player’s value based on being in a new situation but they frequently underestimate by how much. I knew RJAX would be better in Detroit, this much was obvious. But I underestimated his value. I didn’t think he would return top 50 value and he has smashed that. Hopping aboard the Ish Smith train now will likely produce great dividends, particularly in keeper leagues. He’s the Reggie Jackson of next season.

Another example is veterans who move to new situations. Deron Williams’s reputation was in the toilet last season. Although he was still rosterable with averages around 13 PPG/6.5 APG, his FG% was terrible, he was often injured, and he was blamed for a lot of Brooklyn’s failures. Anyone objective could see that he wasn’t to blame and that Brooklyn simply was not a good team. Rick Carlisle in Dallas also has a reputation for rejuvenating floundering careers. If 13 PPG/6.5 APG was his floor, I was betting on Williams doing better than that in Dallas, and returning top 60 value. His auction cost didn’t reflect that and he could be had for a song. So far this season his scoring has been up while his assists have been down. More importantly, his percentages have gone up in a better offense. He has returned fifth round value.

Veterans in new situations have historically been one of my favorite targets for value investment. I know these players can play as they’ve shown us that in the past. Don’t underestimate veterans with a checkered past when they find themselves in a new situation. Sometimes a change of scenery can be a fountain of youth.

Flaws that you can tolerate

Here, I am not referring to the obvious punt FT% big men who everyone knows have early round value once you’re punting FT%. Anyone can see that these players have value in H2H and they aren’t cheap. To be a true value investment, the player must be cheap.

Rather, I am referring to players such as Andrew Bogut with real flaws in their game. The critiques of Bogut are valid. He only plays around 20 MPG because often the Warriors employ “small-ball” lineups. He has “injury history”. Even when healthy and playing, he doesn’t score, as he averages under 6 PPG. However, he is tremendous in the big man cats. There aren’t too many 2 BPG big men in the league and he is one of them. He’s a consistent 7 RPG/2 BPG type player, which is as good as you fill find in the later rounds of drafts. Not to mention he chips in a solid 2 APG out-of-position. His FG% is one of the highest in the league as he makes about 2.5 out of 4 attempts per game. While is FT% is under 50%, that number is meaningless as he averages less than one attempt per game. In H2H, you can make great use of him in a punt points build, in which case he is top 30 value. In roto, while you should not punt, he is nearly as valuable for his roto friendly FG% and big man stats. At his absolute worst, in 8 cat roto, he’s still rosterable. However, he is rarely drafted in the top 100, simply because he does not score a lot of points. I actively drafted him in every format and when I didn’t draft him I was able to obtain him off the waiver wire, even in competitive money leagues. The only thing that degrades Bogut’s value is lack of scoring and that is merely one category.


Drafting rookies is a controversial subject because generally they do not work out. I won’t go into too much detail about the likelihood of a rookie returning top 50 value because an article was already written about it on this very site. However, tell that to anyone who drafted Kristaps Porzingis or Karl-Anthony Towns. Chances are that if you drafted either or both of these guys, they’ve contributed to you at least making the playoffs in your H2H leagues, or cashing in your roto leagues.

I don’t think most fantasy GMs correctly understand risk/reward. If you play in 12 team leagues, assuming all GMs are equal, then your chance of winning the title in any given year is 1/12. If you play in 14 team leagues then it is 1/14, if you play in 16 team leagues it is 1/16, and so forth. Regardless of your league size you have less than 10% chance of winning your league, unless you are significantly better than your opponents. Your goal should always be to win leagues, not just be above average, as the payout in most leagues is weighted heavily toward first place. You want to do whatever you can to increase the odds of actually winning your league, even if that means frequently being last in your league. Think about that for a moment. Most fantasy GMs are risk averse, but you should be risk seeking. Most GMs are trying to reduce variance, but you should be increasing variance.

When I played poker tournaments, I would often post in poker forums. One of the most frequent topics of conversation was how to reduce variance in tournaments. I would post that I want to increase variance and people would think I was nuts!

But think about it. It refers back to a previous article where I referenced Velocity. If you can double up in the early to mid-stages of a poker tournament then you can wield your chip lead like a club as you can eliminate any given player but they cannot eliminate you. Having a chip lead makes it easier to get more chips. Often times a risky approach will lead to an early exit.

However, mathematically speaking, you aren’t going to cash in most tournaments anyway. Being one of the first people eliminated in a tournament is actually preferable to being on the bubble because you have wasted less time.

The same is true in fantasy leagues. You don’t want to be the guy who finishes 8th in a 16 team format, barely makes the playoffs, and gets crushed by the #1 seed. Both in real life and fantasy that is basketball hell. That’s way too much time spent to justify not winning anything. By the All-Star Break you either want to be first or last in your league. You want to either have a realistic chance of winning the championship or you want to be in such a poor position that you no longer invest any time, energy, or effort into your league. Don’t be the guy in last place that never sets his roster because everyone hates that guy. But if you’re in last place at least you can let go the hope of doing anything, stop wasting your time trading and scouring the waiver wire, and prepare for next year.

This season I am in first or second place in 8 out of 12 of my money leagues. In the others am I doing well? Heavens no! In the other 4 I’m near dead last. But at least I don’t have to be emotionally attached to those leagues. Like the Frozen song says, I “Let it go!” Still winning 8 of 12 money leagues is a record almost anyone would strive for.

What does this have to do with rookies? Rookies are the junk bonds of the fantasy world. No one has a clue of their value. There may be signs that point one way or another. However, players drafted 30th sometimes product value, players drafted in the top 5 are often busts. There may be slightly more likelihood of a player succeeding if they were a lottery pick, but the chances of any given rookie becoming a starter is low. There are many players who are drafted than ever become starters. However, the potential payoff is enormous. Rookies generally cost a few dollars, but they can produce first round value.

I’m not advocating buying up all the rookies with no regard. Do your research. There are generally signals that a rookie will have value. Karl-Anthony Towns was regarded as one of the best rookies in a while. You can find many articles detailing his fantasy friendly game, including excellent percentages for roto. He is a shot-blocking big man with range to his shot. The most apt comparison is Serge Ibaka. If Ibaka cost $30+ then what should Towns have cost? He cost nowhere near that in most leagues. Granted there is more risk in drafting Towns than Ibaka as Ibaka is a proven veteran at the NBA level. But here’s the thing-Towns has the potential to be even better than Ibaka! If you are truly risk-neutral you would have seen Towns’s profile and paid for him whatever you were willing to pay for Ibaka. Chances are, you wouldn’t have needed to, and you got him for a discount. If you did that you’re likely happy right now as you got a borderline first round pick for a discount.


Many of these concepts have similar underlying philosophy. Where there is a perception of risk, most fantasy GMs discount their draft picks accordingly. Players with unknown ceiling often cost less than players with a low ceiling but safe floor. My advice is to be the guy that doesn’t care about risk. Draft players who have shown potential in limiting playing time, hoping their situation changes. Do your research regarding rookies. Don’t be afraid to draft a player based on a feeling. Go with your gut! You may not be correct, but at least when you’re wrong, you’re wrong drafting a player whom you actually like. Embrace risk to find hidden value.

Posted in Strategy