Managing A Fantasy Baseball Team Is Sorcery

This is FanGraphs which means we’re all predisposed to translating reality into numbers. In our world, every event on a baseball field has a run value, and we can assign credit for these events to specific players. We like things to be tidy. A 3.0 WAR player is better than a 2.5 WAR. It’s right there in the numbers.

Except… well… per Dave Cameron, WAR comes with a +/- of about 1.0 over a full season. These are estimates. When comparing a three-win player with a 2.5-win player, it more accurate to say that there’s something like a 60 to 80 percent chance that the three-win player was better. The math is even fuzzier when comparing players of different positions. Especially pitchers and non-pitchers. And let’s not get started on who will be better. Ooh boy.

Our tendency to believe in the power of numbers can be hazardous for many reasons – most notably because humans are famously terrible at thinking probabilistically. Returning to our WAR example, there are people who will insist that the 3.0 WAR player was better than the 2.5 WAR player. No exceptions. Larger numbers are always better, right?

You’ll notice, I’ve set this up as a sort of “don’t trust the math” post. It’s not that. Our attempts to quantify every aspect of baseball are a good thing on the whole. They’ve taught us so much about the sport. I’ve been participating in FanGraphs since 2008. What we “knew” then – much of it wrong – pales in comparison to what we “know” now. I expect the same to be true in 2029.

We’ve grown beyond “this player is lucky/unlucky because BABIP” to breaking down batted ball traits and susceptibility to shifts. I swear we used to think half of baseball was luck. In one sense, it’s almost all luck. Each individual play is like a throw on the craps table. Anything could happen with some events more likely than others. However, over the long term, there is no luck in craps. You will lose money. These days, we rarely think a full season player was (un)lucky. We usually can explain what happened. We do even better with larger samples.

As fantasy baseball managers, we have the tools to make mostly rational decisions based on existing information. However, each decision requires knowledge – and in some cases foreknowledge – of how it will affect your roster. For a given format, there is no one-size-fits-all path to success.

Here’s an example. I often preach loading up on star talent in FGpts ottoneu leagues. Put simply, the formula is this: you MUST have elite performers to win a FGpts league. So roster the guys you know are the best bet to be elite rather than trying to find a bunch of underpriced breakout candidates. You also MUST have a complete roster of core performers or better. This is where it can get hairy with the stars and scrubs approach. If all of your $1 bets fail, you’re stuck in good-not-great territory.

There are alternatives. I’ve won by loading up on elite SP, spending $70 on my bullpen, and limping by with only Mike Trout on offense. I’ve won with a loaded lineup of elite bats and $70 spent on my pitching staff. I’ve seen others win by entering the season with 10 premium prospects and converting them all into elite talent during the season. In each case, the elite are involved, but they’re all very different paths.

Backing up a step, each decision is made in the context of your roster. You make your early picks based on knowledge of your strengths, weaknesses, and preferences. Since I scrounge the waiver wire as a job (I write about it daily), I usually skew towards a stars and scrubs approach. I trust myself to create value after the draft. In redraft leagues, that manifests as a willingness to gamble on Adalberto Mondesi and other seemingly volatile players.

Once the foundation of the roster is laid, it’s time to fit players to what you’ve already built. There’s little sense reaching for Joey Gallo when you’ve already selected Giancarlo Stanton and Aaron Judge. You probably desperately need Mallex Smith. Or a starting pitcher.

Our concepts for value do a half-assed job of incorporating context. Most people work from a simple rankings sheet or auction dollars guide. These become obsolete for your team early in the draft. Methods like Standings Points Gained do a decent job of incorporating context based on a mean or median projection. However, projections can only take you so far.

This is where the sorcery comes into play. Not all projections are created equal, and it’s up to the end user (you) to determine which players are likely to exceed or fall short in the categories you need. Imagine you’re considering two players with a 30 home run projection. You really need 35 or 40 home runs, but this is all that’s available. One has a batted ball profile like Christian Yelich. The other, like Rhys Hoskins. Pretend stolen bases and lineup role are equal.

In this case, Hoskins’ pulled fly ball approach is far more likely to provide the extra home runs above the projection. The Yelich-like hitter will probably post a higher batting average, but there are more ways for him to fall short of the 30 home run projection. You have to make a choice on the tradeoffs between the two profiles based on your needs.

These considerations extend to in-season management with your bench. Do you load up on pitching depth, stash prospects, or run platoon hitters? Once again, the correct answer is context dependent. Obviously, the construction of your roster is a big part of the equation. However, you are part of the context too.

We all have our blind spots. If you’re able to identify them, you can try to design your team and strategy to mitigate them. For example, I tend underestimate the importance of good starting pitchers. My standard adjustment is to over-invest in elite, non-closing relievers and pour extra effort into identifying April breakouts. Of course, this issue comes with an even easier fix – draft enough good starters in the first place. While that’s easy to say (and do) in a snake redraft, it’s easier said than done in the auction or dynasty settings that comprise most of my leagues.

Fantasy management is a balancing act requiring deft maneuvering, imaginative tactics, and brutal self-honesty. The best managers are akin to sorcerers, weaving spellcraft in the form of fake shares of real baseball players.