This is part four of my series on the daily fantasy lineup optimizer I rebuilt this year (the series starts with the origin story). Everything up to now has been about finding the single best lineup. Here’s the uncomfortable truth that took me a while to accept: the single best lineup is the easy part, and it’s not what wins tournaments. The hard, interesting problem is building a set of lineups that are each strong but smartly different. That’s where the real work — and the real edge — lives.
Why one perfect lineup isn’t enough
If you enter a big tournament, you’re usually entering more than one lineup. The naive move is to submit your single best lineup twenty times — but projections aren’t certain, and if that one lineup misses, all twenty miss together. You’ve put everything on one guess. The smarter move is a portfolio: twenty lineups that each look strong on paper but cover different outcomes, so when the night unfolds in a way you didn’t perfectly predict, some of your entries still hit. The optimizer’s job stops being “find the best lineup” and becomes “build the best set of lineups.” Those are very different problems.
Stacking: betting on a good night together
The first tool is stacking — deliberately putting several hitters from the same real team in one lineup. It sounds like it breaks diversification, but it’s the opposite. In baseball, when a team has a big offensive night, its hitters score together: one drives in another, who was driven in by a third. Their fantasy points are correlated. Stacking four or five hitters from one team is a bet that that team erupts, and if it does, your lineup is stuffed with the players who cashed in. A common pattern is a five-and-three: five hitters from one team, three from another, so you’re leveraged to two offenses instead of spreading thinly across eight. The optimizer lets you require these stacks instead of hoping they show up by accident.
Exposure: how much of any one player you want
The second tool is exposure control. If your favorite pitcher lands in all twenty lineups, your whole night rides on him. Exposure caps let you say “no single player appears in more than, say, half of my lineups,” which forces the set to spread its bets. The tool also has a cooldown option — a player used in the last few lineups has to sit out the next one — which is a simple way to keep your lineups turning over instead of clustering around the same core. These are the same controls the expensive commercial tools sell; there’s nothing proprietary about them, and I think it’s more useful to explain them than to hide them.
Reading the exposure grid
After it builds your lineups, the tool shows an exposure grid — players down the side, lineup numbers across the top, a mark in every cell where a player appears, and a shaded percentage column on the right. This is straight out of my old Excel sheets; I used to live in those pivot tables. At a glance you can see whether you’re too concentrated on one player or one team, which stacks are doing the heavy lifting, and where your set is thin. It turns “trust me, these are diverse” into something you can actually look at and judge.
Why correlation is the whole point
Say a team puts up ten runs on a good night. The hitters who did the damage score fantasy points more or less together — the single, the guy who scored on it, the one who drove him in. If you had five of those hitters stacked in one lineup, that lineup just went nuclear. If you’d instead spread your five hitters across five different teams to feel “diversified,” you’d have caught one slice of that explosion instead of the whole thing. In a top-heavy tournament where you need a huge score to win, the concentrated bet is the correct one. Stacking isn’t reckless; it’s matching your roster to how baseball actually scores.
Cash games and tournaments want opposite things
How aggressive you get depends on the contest. In a tournament with thousands of entries you only care about a huge score, so you stack hard and lean into the upside. In a cash game, where you just need to beat half the field, you want the opposite: safer, steadier players and smaller exposure swings. Same tool, opposite settings. That’s the reason the stacking and exposure controls are knobs you turn rather than decisions baked in — the “right” amount of diversity depends on what you’re trying to win, and only you know that.
How I’d actually set it
If you’re staring at the controls wondering where to start, a sane tournament default: require a four- or five-hitter stack from one team, a smaller secondary stack from another, and cap any single player around forty to fifty percent of your lineups so no one play sinks the set. Generate, read the grid, and adjust — if a player is in nearly every lineup and you didn’t mean that, tighten his exposure and run it again. The deeper version of this problem, which the most sophisticated tools chase, is optimizing all your lineups together to maximize the chance at least one wins, rather than building them one at a time. That’s genuinely hard and a good subject for a future post; for now, stacks plus exposure caps get you most of the way there.
The honest part
None of this guarantees you win — projections can be wrong, and variance is the whole reason these contests exist. What a good portfolio does is make sure you’re not accidentally making one big undiversified bet when you think you’re making twenty. That’s the difference a real optimizer makes versus picking lineups by feel. If you want to see your own set laid out in the grid, the MLB lineup optimizer is free — set a stack, set an exposure cap, generate, and watch the grid fill in. Next in the series: what it actually took to move these models off Excel and onto the web.
Things that I use, like, and am affiliated with:
Mint Mobile offers great cell phone service for $15 flat, get $15 off using the link. Get discounted phones with service activation and no contract.
I never spend money before I check Mr Rebates or Rakuten to get cashbacks, rebates, discounts, coupons or cheaper gift cards.