Jonathan Bales is the author of the Fantasy Sports for Smart People book series—a collection of guides designed to help you win on DraftKings.

I wasn’t sure if I was going to write a new daily fantasy book for NFL this year. I sort of pushed things back a bit, but I bit the bullet and just started writing it a few days ago. I wrote the last book in 13 days and some people thought it was good enough to not immediately throw away, so this one might top Bible sales if I dedicate three weeks to it.

I thought it would be cool to give you a sneak peek…

DFS is my favorite game to play. I love it. But I like to play a lot of other games, too. I like poker, ping pong, darts, Stratego, Monopoly, foosball, and a bunch of weird games I made up and make my friends play. I even like Rock-Paper-Scissors and play it against a computer, like, every day. It’s a great way to practice completely overthinking things.

As I mentioned in a previous article, one game I play all the time is called Fun Run. It’s a mobile game and it’s sort of like MarioKart in that you race and there are question mark things you can get that have weapons to kill your opponents (lightning, a bear trap thing, a sword, etc). I don’t even really play video games at all but I really like this one because there’s quite a bit of strategy involved.

When I first started playing, I just tried to move as fast as possible and finish the race with the quickest time I could. Seems logical…faster is better in any race, right? Maybe, maybe not. You need to perform well to win a race, but the goal of any race isn’t to have the fastest time you can, but to have a faster time than everyone else. Those are similar ideas, but not exactly the same.

Thus, now my strategy is more about creating the largest possible difference between my time and that of my opponents. That might mean going out of my way to obtain an additional question mark, using a super low-variance strategy late in the race if I’m winning, taking more risks when I’m losing, and so on. In fact, the goal is almost never to minimize my time, but rather to maximize win probability.

Let me start a new paragraph and write that again: the goal in any zero-sum game is to maximize win probability.

This should sound familiar. While scoring a lot of points is a prerequisite for tournament success, your primary goal should not be point maximization. It should be win probability maximization.

As I’ve mentioned, I think there are two layers of value in tournaments. There’s the traditional dollar-per-point type of value—expected production minus cost—and then there’s the usable value a player provides you. I think most players either overlook the second type of value completely or don’t understand how important it really is.

In short, we need to be concerned not only with the chances of being right on a particular player, but also (perhaps more important) the benefits it actually provides us if we are indeed correct. What are the chances we hit on a pick and what’s the payoff if that happens?

When you roster high-owned players, you’re probably at least coming close to maximizing your point projection and the odds of your players “hitting.” There’s a reason they’re highly owned; no one is in 30 percent of lineups if they’re a crappy value. But in rostering the chalk in GPPs, we’re also minimizing the potential payoff. If you hit on a quarterback who is in 33 percent of lineups, you’re still completely even with one-third of the field. If you hit on one in five percent of lineups, you have a leg up on 95 percent of users.

Another way to look at this idea is to consider situations in which you’ve been trailing in a tournament. When you’re in, say, 50th place in a big GPP with two players remaining, it certainly feels a whole lot better if those guys aren’t in lineups ahead of you, right? If there are multiple lineups ahead of you with the same players left to go, you have no shot at winning. At that point, you’d certainly trade in those two players for a pair that is unique and gives you a shot at the win, even if they’re slightly less valuable, right? Of course. Being contrarian leads to less fantasy scoring over the long run, but it also gives you more “outs,” so to speak, in the event that you need to jump other users (which happens almost all the time in a GPP). If pulled off correctly, it can maximize your win probability without maximizing points. In my opinion, it is a forward-looking DFS strategy that acknowledges fallibility—as opposed to a shortsighted “I’m going to take the best values because I want to score as many points as possible (and it couldn’t possibly be the case that I’m wrong about these guys or even that I’m right and they don’t work out).”

By going against the grain, you’re leaving yourself room for measurement errors, which is basically what my entire DFS strategy is about. I know it’s challenging to predict fantasy scoring, even in predictable sports, and it’s even more difficult to beat an entire field of really smart people. So for the most part, I’d rather focus on maximizing the benefit I receive if everyone else is wrong. I don’t need to be better than everyone else when it comes to projecting players; I just need to have a comparable level of predictive ability in that area if I’m putting myself in a better position when it comes to reaping the fruits of my labor. In being contrarian, I’m betting on low-frequency events that have asymmetrical payoffs such that, although I’m going to be wrong more than I’m right, I’m going to be overcompensated for being right when that happens.

There are times when we can have our cake and eat it too, acquiring high-value players who are also low-usage. In fact, I’d never recommend being contrarian just for the sake of it, picking off-the-map players with no rhyme or reason. However, the overarching idea is that we need to balance value with ownership, using game theory to determine which players provide not only the greatest traditional value in terms of dollars-per-point, but also those that offer maximum usable value in terms of increasing the odds of winning.

In short, others’ decisions in a marketplace setting are arguably the most important component in determining what’s optimal for you.

Note: I was watching the World Series of Poker yesterday and I thought of another good example to add here that I think displays the power of “uniqueness” in poker, and indirectly DFS. You’re in a poker tournament and two players ahead of you go all-in. You call, putting all of their chips at risk. The others both flip over Ace-Queen. Which hand would you prefer: King-Queen or King-Jack?

It’s obviously the latter, even though it is theoretically the weaker hand. It keeps you alive, giving you outs and a much higher chance of winning the hand. Actually, because the duo ahead of you have the same hand, you’d basically be competing heads-up in order to triple-up—great odds if you can get them—but it works only if your opponents have the same hand and you have two live cards.

The same is true in daily fantasy; the more common your opponents’ lineups and the more yours is differentiated, the better your odds, all else equal.

Further, you know King-Jack is suitable to King-Queen only after you have knowledge of the opponents’ hands. Prior to the hand, you’d very clearly perform King-Queen. The difference between poker and DFS is we can “know” our opponents’ cards before the hand is dealt; because we’re dealing with a large sample size of lineups in a GPP, the individual lineup fluctuations even out and we can very often forecast player ownership such that, if played properly, we can predict our opponents’ hands and play not necessarily the one that’s most optimal in a vacuum, but the one that’s most capable of “filling in the gaps” to beat our opponents.

Also, why the hell did you call two all-in raises with King-Queen anyway bro?