MLB DFS Multi-Entry Strategy: Build A Stack Portfolio

Most people I talk to think the hard part of daily fantasy baseball is finding the right team to stack. It isn't. The hard part is accepting that you will not find it, and building anyway. An MLB DFS multi-entry strategy is a portfolio problem before it is a research problem, and the figure that governs the whole build is not a projection at all. It is a gap between two numbers. I'll come back to it.

In Summary

• On a big MLB slate, even a great offense is only about 10% to 15% to finish as the top stack.
• That math is the argument for multiple entries: you are buying correlated shots, not one answer.
• The number that matters is the gap between a team's odds of being the top stack and its projected ownership.

Watch The Video

Chris Rando and I walked through this entire build on camera, from the first pass to the exported lineups.

Watch on YouTube

About the example slate. The build below comes from a 2019 slate Chris and I broke down on video back in the Awesemo days. Every salary, ownership figure, and roster in it is from that night, and the players have long since moved on. The names are dated. The process is not, which is why I still teach it this way.

Why One Perfect Stack Is A Losing Plan

Stacking means rostering a group of hitters from the same team against the same opposing pitcher, four on FanDuel or five on DraftKings depending on the site. The reason it works is sequencing: a baseball offense scores in bunches, and the players who benefit are the ones batting around each other.

Chris put the payoff plainly. Jose Martinez comes up with the bases loaded, Matt Carpenter and Paul Goldschmidt among the runners, and hits a grand slam. You do not just bank Martinez's 30 points. You collect roughly another nine from the three runs those baserunners scored. That is correlation doing the work, and it is why MLB DFS correlation beats rostering eight good hitters from eight different games.

Here is the problem. On a full slate, the best offense on the board might only be 10% to 15% to actually finish as the top-scoring stack. Even at the high end of that range, there is an 85% chance it is not the team. A single lineup asks you to be right about something that lands your way roughly one time in seven.

So you stop trying to be right once. You enter many lineups, own many teams, and let the slate decide which one goes off. Every decision below follows from that single concession.

Step One: Build The Chalk Before You Fade It

Before I set a single rule, I run one pass with nothing constrained: just projections, four-hitter stacks, no exposure limits, no randomness. I am not building lineups yet. I am building a map of where the field will be.

On the night we recorded, that pass returned exactly what you would expect. Colorado came up first, at home. The Angels came up next, drawing a Baltimore staff nobody wanted to face. Those are the chalk stacks, and knowing them is worth more than avoiding them, because the whole point of the next step is to price them.

You want a piece of the chalk. Chris takes 10 to 15 of his lineups, depending on how large his entry pool is that night, and simply plays it straight, because a popular team is popular for real reasons and sometimes the obvious thing happens. The rest of the pool works down from there.

The Leverage Read: A Worked Example

This is the gap I promised. Our Top Stacks Tool gives every team two figures side by side: its odds of finishing as the top stack, and its projected ownership. Neither one is interesting alone. The distance between them is the whole game.

Team	Odds of top stack	Projected ownership	The read
Angels	15%	15.9% ownership	Priced correctly
Rockies	8%	8% ownership	Priced correctly
Red Sox	9%	6% ownership	Underpriced

Look at the Angels. Fifteen percent to be the best stack on the slate, rostered by 15.9% of the field. They have a great matchup and the field knows it, so you are paying full retail for a real edge. The Rockies at 8% and 8% are the same story. Boston is the row that pays. A 9% chance to be the top stack against 6% ownership means the field is under-rostering a team that is more likely to go off than the Rockies everyone is piling into.

The supporting evidence was sitting right there. Boston carried an implied team total of 6 runs at home, which Chris read as second or third highest on the slate. The prices at the top of that order were high. Mitch Moreland, the projected cleanup hitter, was barely rostered, and Xander Bogaerts sat at 1% owned because his FanDuel price ran well past his big-game ceiling. He is not a home run hitter, and the number said so. Which is precisely why nobody wanted him.

Price is not the whole story when a team is about to bat around, though. If Boston scored 10 to 12 runs, Bogaerts and Moreland were each coming to the plate five or six times, and plate appearances are production no matter who is holding the bat. The field had priced that out of both of them.

That gap is the reason to build. The tool surfaces it in one column, but you still have to decide what to do with it. If you want the ground floor underneath all of this, how to win at MLB DFS covers it.

Build this pool without the spreadsheet. MLB Data + Sims gives you the Top Stacks Tool, Ownership Projections, and Contest Sims in one place, so the leverage read above takes seconds instead of a night. Code MLBENTRY10 takes 10% off your first payment: start with MLB Data + Sims.

Stack Shape: Why Four Is The Floor

Once you know which teams you want, you choose how to hold them. Chris runs 4-4 (two teams, four hitters each), 4-3-1 (a four-stack, a three-stack, and a single high-conviction one-off), occasionally a 4-2, and rarely a 2-2. The constant is the four.

The reason is brutal and simple. If you take three hitters from the team that goes off, everyone holding four passes you. You did the hard part, identified the right offense, and still lost the tournament to the people who committed one roster spot further.

The counterargument is salary. Four Boston hitters and three Angels hitters cost less than four and four, and that leftover money buys you a minimum-priced bat somewhere else. Call it the real trade in 4-3-1: you give up a sliver of correlation to buy roster flexibility. If you want the full argument for the stack itself before you tune the shape, MLB DFS stacking covers the ground underneath this section.

Three shapes cover almost every build:

• 4-4 — two four-stacks. The default on a big slate, and the most correlated thing you can roster.
• 4-3-1 — a four-stack, a three-stack, and one high-conviction one-off you lock into every lineup.
• 4-2 — a four-stack plus a two-man mini-stack, when the salary is tight and you want a third game exposed.

One more shape worth knowing: the wraparound. Instead of hitters one through four, take 8-9-1-2 or 7-8-9-1. The bottom of an order is cheaper and lower-owned, and if the team bats around, those spots score exactly as much as the top. Tommy La Stella hit ninth for the Angels the night before we recorded, and if you did not have him, you did not win. That is the kind of bottom-of-the-order game that never shows up in a projection. Speed guys who live down there, Billy Hamilton and Terrance Gore, carry stolen-base upside nobody is paying for.

Differentiate The Pool: Exposure Caps And Randomness

Now the portfolio. Left alone, any lineup builder will hand you the same players over and over, because it keeps solving the same optimization. I ran 50 lineups and Shohei Ohtani appeared in 48 of them, which is 96% exposure to one player. Nobody wants that. Cap him at 25% and the builder is forced to find a different answer in the other 75%.

The cleanest way to set those caps is to let the projection do it. Take each player's projected points, use that number as his exposure percentage, and move on. Mike Trout projected for 19 points, so he lands in about 20% of my lineups. It is crude, it takes one column in a spreadsheet, and it produces a genuinely wide spread of teams.

Randomness does the other half. I set 25% randomness on that build, which nudges each projection up or down before every lineup solves. Two things happen. Your lineups stop being clones of each other, and the builder starts surfacing players you had written off. Chris tells a story about an earlier slate where his build kept spitting out Cedric Mullins, a minimum-priced Baltimore outfielder he says he would never have looked at, until it stopped making sense to leave him out. Mullins went for about 30 points. A 1% owned player can win you a tournament, and you will not find him by hand.

Randomness also tells you when you are about to be less unique than you think. With Boston locked as the first stack, my build paired it with LA 81% of the time. Two chalk-adjacent teams, married in four out of five lineups. So I removed LA from that crunch entirely and Colorado came back at 60%, with Toronto, Baltimore, and the White Sox filling in behind. Fifteen different partners for Boston, which is what a portfolio actually looks like.

Remember the 85%. That number is not just the argument for entering more lineups. It is the argument for capping every player inside them. If you are wrong about a team six times out of seven, you cannot afford to be all the way wrong at once.

The last pass is subtraction. If Jackie Bradley Jr. and Sandy Leon show up in every single lineup, pull them out of the pool and rerun. What comes back is not a body swap. Removing a player changes the salary the builder is working with, and the whole shape of the lineup moves.

Slate Size Changes The Whole Build

Everything above assumes a big board. On a 13-game or 14-game slate, 4-4 stacks are strong and you want Boston paired with all fifteen of those different partners, because there is too much opportunity to concentrate.

Shrink the slate to four or five games and it inverts. Chris pairs his one favorite offense against every other team, rotates the hitters inside it, cuts the low-projected players entirely, limits himself to two or three pitchers, and builds 25 to 30 high-probability lineups rather than spraying. Fewer games means fewer places to hide, so uniqueness has to come from construction instead of coverage, which is a different application of the same leverage principle.

There is also a structural trick FanDuel gives you for free. Its roster positions are rigid. Find two teams where neither is starting a shortstop, and the crowd running a straight 4-4 across that pair will not end up with that combination. Almost nobody builds it, so almost nobody owns it. Pairing those two offenses is a lever the field leaves on the floor purely because of how the roster form is shaped.

Enter Smarter, Not Just More

Chris used to hand-build 40 to 50 lineups a night. He does not hand-build at all anymore, and he now fires 100 to 150. The hours he was spending on data entry moved to research, which is the only part of this that compounds.

Volume alone is not the win, though. Chris's read, and mine, is that lower buy-in tournaments are a little easier to win than the expensive ones. So take the entry fee you were going to spend on one $25 Grand Slam seat and split it across five cheaper entries. Five correlated shots spread your outcomes across more of the slate instead of resting the night on one lineup, and it is frankly more fun to sweat five teams than one.

Hand-building costs you something you cannot see, too. When a lineup changes late, you feel it, and you start making decisions with your gut instead of your process. Exporting a pool and uploading it once removes that.

Building This With The Stokastic Sims

The workflow in the video was an optimizer workflow: upload projections, set stacks, crunch, cap exposure, export. Our Sims do that job now, and they do one thing an optimizer structurally cannot. An optimizer ranks lineups by projected score. The Sims run the contest thousands of times and rank your pool by how often each lineup actually wins, which is the question you are trying to answer. That distinction is worth understanding before you build anything, and MLB DFS sims vs optimizers is where I would start.

Practically, you set the same rules: stack shape, exposure caps, randomness, locked one-offs, banned players. You read the Top Stacks Tool and Ownership Projections for the leverage gap. Our Late Swap tool covers you when a starter gets scratched after lock, and DFS Player Compare, which is free, lets you weigh two bats against each other before you commit. Our Live Before Lock show walks the same board every afternoon if you want to see it done in real time, and the MLB DataHub carries the projections and ownership underneath all of it. The free Sims will show you the shape of it before you pay anything.

FAQ: MLB DFS Multi-Entry Strategy

How many MLB DFS lineups should I enter? Enough to cover several stacks, since no single team is better than roughly 10% to 15% to be the top stack on a big slate. Chris moved from 40 to 50 hand-built lineups to 100 to 150 once he stopped building them by hand.

How many players should I stack from one team? Four is the floor on FanDuel and five is available on DraftKings. Three hitters from the right team loses to everyone who took four.

What is a good exposure cap for one player? Around 25% for a popular bat you would otherwise be stuck with in nearly every lineup. Setting each player's cap equal to his projected points is a fast, workable default.

What does randomness actually do to my lineups? It shifts each projection slightly before every lineup solves, which differentiates the pool and surfaces cheap players your projections buried. Cedric Mullins gets into a build exactly this way.

Should I still play the chalk stack? Yes, in part of the pool. Ten to 15 lineups on the popular team keeps you alive when the obvious thing happens; the rest of the build is where you take the leverage.

Bottom Line

The night we recorded, the best-priced team on the board was not the Rockies at Coors and it was not the Angels against Baltimore. It was a Red Sox side the field had priced at 6% while the numbers said 9%, sitting on a six-run implied total with a shortstop nobody wanted. I never knew whether Boston would hit. I knew Boston was cheap relative to its chances, and I knew I could own it fifteen different ways.

That is the shift, and it survives every roster change since. You stop asking which team is going to score 12 runs, because nobody knows, and you start asking which teams the field is mispricing and how many differentiated ways you can hold them. The stack is the unit. The portfolio is the strategy.

Ready to build one? MLB Data + Sims runs the contest thousands of times over your pool and ranks every lineup by win probability, with the Top Stacks Tool and Ownership Projections feeding the leverage read. Use code MLBENTRY10 for 10% off your first payment: build your MLB pool with the Sims.

Prefer to tail a proven card instead of building your own? The value-first habit above still applies; you just outsource the legwork. DFSnDonuts Picks on Tails is a DFS-native expert (#8 on the MLB Hot Streak leaderboard as of July 16) whose plays come with the receipt every Tails expert carries: a full graded track record you can check before you tail. Start with the free expert picks feed, where free picks post every day.