NBA DFS Strategy Guide: How I Win on DraftKings and FanDuel

Everyone wants to know how to win at DraftKings and FanDuel NBA daily fantasy. There is no magic bullet, but there is a real edge in a handful of concepts that most players misunderstand. I'm Alex Baker, and I built my reputation grinding high-volume NBA DFS, so this NBA DFS strategy guide is the framework I actually use, not a recap of someone else's tips. For today's specific projections, ownership, and plays, head to the Stokastic NBA DataHub. This page is about the why behind those plays.

A quick note before we start. I founded Awesemo.com (now Stokastic.com) to put accurate projections and simulation tools in players' hands, so when I reference "our projections" or "our Sims" below, that's the toolset I'm talking about.

In Summary (TL;DR)

• Projections are everything in NBA, but accuracy is the edge, not predictability. Being one minute or a couple of points off a player's true projection can make or break your slate.
• Ownership wins tournaments. Mix a couple of low-owned plays with the best value of the night. Judge leverage player-by-player: get meaningfully under the field on high-upside pieces and stay with chalk only where it's truly worth it.
• Stacking is weak in NBA. Most stats between teammates are negatively correlated. Don't optimize for game stacks the way you would in MLB or NFL.
• Build around injuries. Take the underpriced players who absorb usage from a ruled-out star, then soak up the remaining salary with the highest-priced studs.
• Late swap is the single highest-value action you can take. If you can't watch the news until the last tip, I'd rather not enter that day.
• I build by win probability, not by one projected score. Stokastic Sims simulate the contest tens of thousands of times so I'm optimizing for how often a lineup actually finishes first.

NBA DFS Projections: Accuracy Is the Edge

People love to say NBA players are easier to project than other sports, and to a large degree that's true. NBA has many fantasy scoring opportunities, but each one is low-significance to the overall score, a huge contrast to a touchdown, a goal, or a home run. As a result you can usually guess within about 10 fantasy points of a player's performance on a given night. In my Boom/Bust tool I give you the range of fantasy scoring you can expect for each player, not just a single midpoint.

The misconception is that predictability makes NBA DFS easier to win. Everyone has that same advantage, so it's still an even playing field. What predictability actually does is push the best NBA strategies toward popular players more than in other sports. There are nights when you should play a guy even if nearly the whole field has him. The classic case is a high-usage point guard on a night his ball-dominant teammate sits, because he soaks up the shots and assists that normally go elsewhere. That does not mean you load up on every chalk play, though. You still need the right sleepers to win a large-field tournament.

Accuracy is where projections actually earn you money. Being a minute or a couple of points off a player's true projection makes a good play look like a great one, you over-commit, and the night gets away from you. Accurate projections are also how you tell a sneaky sleeper apart from a plain bad play. We spend all day refining our NBA projections, so use them to build your lineups or as a second opinion on who you're rostering.

NBA DFS Ownership and Leverage

If the highly owned players have the big nights everyone expects, the score it takes to win climbs drastically. So if you load your lineup with chalk, you'll need far more points to win than if you mix in some contrarian pieces, and you're competing with thousands of near-identical lineups that differ by a player or two.

The best approach I've found is a couple of genuinely low-owned players mixed with the best value plays of the night. That hits the sweet spot: a good chance to cash and a real chance to win outright. Reading the room on who'll be popular or sneaky is hard to do by feel, so I lean on our ownership projections and rankings instead of guessing.

One NBA-specific note on math: multiplying each player's ownership together estimates the expected duplication of that exact lineup, a rough proxy that assumes the plays are roughly uncorrelated (which, since stacking isn't a common NBA strategy, mostly holds). That's a useful number, but it is not a measure of how contrarian your lineup is. Leverage is a player-by-player judgment: are you meaningfully under the field on the high-upside pieces, and are you only paying up for chalk where it's genuinely worth it? A low duplication estimate does not by itself confirm the build is contrarian enough.

Stacking in NBA DFS: Why It Is Weaker Than You Think

Game stacking is a popular GPP strategy in other sports, but it works far worse in NBA. There's really only one positively correlated pairing between teammates: assists and points. Most other stats are negatively correlated. Every shot one player takes is one fewer shot for a teammate. There's nothing harmful about rostering two players from the same game, but optimizing for that over everything else leads to worse lineups.

You'll occasionally hit an overtime game, which lifts everyone's fantasy output and makes a game stack roughly a wash overall. So I don't chase game stacks, but I don't avoid them either. Common sense helps. If a team is a huge favorite, loading up only on that team without any opposing players cuts your ceiling, because you actually want the other team to put up resistance and keep the starters playing and producing.

Roster Construction: Build Around the Injury News

The cleanest way to build NBA lineups is to take players who are underpriced because of injuries to their teammates, then soak up the remaining salary with the highest-priced studs. Picture a starting point guard who gets ruled out after salaries were already set. That's a spot to lean into his teammates, because his usage, shots, and assists get redistributed across the roster while their prices stay frozen at the pre-news level.

A common mistake is filling out an all mid-priced lineup on a night full of injuries. That leaves value on the table. The way you maximize NBA fantasy points is to grab a few of the cheapest players who directly benefit from an injury, then spend up on a superstar or two with the salary you saved. Our Lineup Generator and Sims are built for exactly this: the tool finds these salary-relief spots and balances the rest of the build for you, then ranks every lineup by its simulated win rate so you're comparing builds by tournament equity, not by one projected total.

Showdown and Single-Game Strategy

On single-game slates (DraftKings, FanDuel, and similar), you pick a player to earn multiplied fantasy points, but there are far fewer players to choose from overall. The biggest danger is building a lineup that's too obvious, because there are usually a handful of combinations that are clearly "best." Enter those and you'll split the top prize with a crowd. It's not fun to take home a few hundred dollars for first when the same contest could have paid five figures.

On DraftKings showdown, the captain slot applies a 1.5x multiplier to that player on both fantasy points and salary, so I put my highest-ceiling play there: the 1.5x boost magnifies the single biggest-upside outcome on the slate. And because captain choice has almost no relationship to how often a lineup gets duplicated, you don't sacrifice uniqueness by captaining the obvious high-ceiling guy. Get contrarian with the surrounding picks instead, and you give yourself a better shot to win outright or at least split with fewer people. The captain is the wrong place to chase uniqueness.

On FanDuel-style showdown there are more roster permutations because of the tiered multipliers, but the number of good lineups is smaller, since in NBA the odds of a deep-bench scrub outscoring a starter are minimal. Intentionally leaving a little salary on the table is a clean way to land a more unique combination. Most of the field tries to build the single strongest squad, so a couple of deliberately suboptimal choices can be exactly what separates you.

Match your single-game build to the field size. In a small contest, err toward the strongest possible lineups. In a massive-field showdown, take calculated risks aimed at the less likely outcomes that pay the most. On full main slates, I've found the best GPP lineups look fairly similar regardless of field size, so this field-size adjustment matters most in single-game formats.

Rather than eyeball ownership and captain choices on a single-game slate, I run it through Stokastic Sims, which simulate the contest tens of thousands of times and rank every lineup by how often it actually finishes first. That turns the "which captain, which combo" guesswork into a ranked list of builds.

Taking Advantage of Late Swap in NBA DFS

DraftKings, FanDuel, and Yahoo all offer late swap in their NBA contests, and it completely changes how you should build. One mechanic is load-bearing here: you can only swap a player whose game has not yet tipped. Once a player's game starts, that roster spot is locked for the night. That single constraint is exactly why I deliberately roster several players in the late game windows, because those are the spots I can still edit when news breaks. Players who become good plays after lock tend to be either under-owned or buried in over-optimized, heavily duplicated lineups, because the field just re-optimizes for projected points when news drops. The most important thing on these sites is staying tuned to news right up until the last game tips. Miss the news and you'll have a hard time winning over the long run.

The best version of late swap is full control of your lineups from a computer so you can re-optimize on new information. If you have a lot of money in play regularly, set yourself up to edit on the fly. The mobile apps will at least let you swap out a player who gets ruled out, but the strongest move is editing every lineup by hand to pick the best swaps. That's realistic for your most-invested lineups; once you're running dozens or hundreds, you need software to do it at scale. Personally, if I know I won't be able to manage late swap because I'm tied up, I usually just skip tournaments that day. Late swap is that important.

With late swap available, I build lineups around the best-case scenario for each player without putting conflicting scenarios in the same lineup. For example, if a starting forward is questionable, I won't build lineups that need both that forward and his direct backup to pay off together. I'll spread my exposure so I have plenty of lineups with one or the other. When the news lands, I re-optimize the lineups that are no longer valid. It helps to have several players in the late games so you can run 2-for-2 or, better, 3-for-3 swaps, which gives you the salary flexibility to fully capitalize on the news. When there's key news still pending at lock, I use the bulk groups and exposure controls in our Lineup Generator to mass multi-enter this way instead of touching lineups one at a time.

Cash Games vs. Tournaments: Opposite Builds

Everything above tilts toward tournaments, because that's where most of the value and most of the questions are. But cash games (double-ups and 50/50s) need the opposite build. In cash you only have to beat about half the field, so you want the highest-floor lineup, built straight from projections, not the simulated-tournament pool. Ownership and leverage barely matter there. Reserve the contrarian, low-owned, ceiling-seeking approach (and the Sims-based contest emulation) for GPPs, and never fire the same lineup into both formats.

How to Build an NBA DFS Lineup: A Worked Example

Let me walk a real DraftKings build the way I'd actually do it, using round numbers so the logic is clear. Say the slate has a $50,000 salary cap and 8 roster spots, and an hour before lock a starting point guard like LaMelo Ball gets ruled out. His backup, who was priced at $4,200 on the assumption of 22 minutes off the bench, is suddenly in line for 32-plus minutes and a much bigger usage share, but the site price won't catch up before lock.

Here's the build math. I lock in that backup at $4,200, which is roughly $1,500 to $2,000 below what his new role is worth. That salary relief is the whole point: it lets me pay all the way up for two studs (say Nikola Jokic at $11,500 and Shai Gilgeous-Alexander at $10,800) and still fill the remaining five spots without reaching for dart throws. A few times through the Sims, I compare that build against an all mid-priced version, and in my experience the injury-driven build tends to grade out ahead in the Sims for tournaments because the studs raise the ceiling while the value play protects the floor.

For ownership, suppose the projected ownership comes back as Jokic 38%, SGA 31%, and the cheap point guard 45% because everyone sees the obvious pivot. That's three popular pieces, so the leverage has to come from the rest of the build: I'll find a genuinely low-owned spot, say a 10-12% wing in a strong matchup, to get over the field on real upside. Multiplying those ownerships together estimates how often this exact lineup gets duplicated, but it does not tell me the build is contrarian. A Jokic/SGA/chalk-PG core is high-owned by design, so I lean on the leverage from the under-owned wing rather than treating a duplication estimate as a green light.

Build by win probability, not by gut feel. When I run the slate through Stokastic Sims, I'm comparing thousands of versions of that lineup by how often each one finishes first, not by one projected total. Use code SIMS10 for 10% off your first payment on Stokastic+: unlock the NBA Sims and Lineup Generator. (SIMS10 is a proposed launch code; confirm at checkout.)

Putting It All Together

NBA DFS is all about accounting for the latest information, because a lineup that's a winner 30 minutes before lock can be a dud 10 minutes later on one injury report. To succeed you need a process you can execute quickly in that tight pre-lock window: check projections, weigh ownership and leverage for GPPs (or floor for cash), build around the injury news, and stay ready to late swap. Do that consistently, and NBA becomes one of the most engaging DFS offerings there is.

Here's the repeatable process I run every slate:

1. Pull projections and Boom/Bust ranges to anchor every player to an accurate number and a realistic range.
2. Read ownership. Decide where you're with the field (chalk that's truly worth it) and where you're leveraging against it.
3. Build around injury news. Grab the underpriced beneficiaries, then spend up on studs with the saved salary.
4. Decide format. GPP means ceiling, leverage, and a couple of low-owned plays; cash means highest floor from projections.
5. Run the Sims to compare builds by win probability, not gut feel, and let the Lineup Generator scale your exposure.
6. Watch the news until lock and execute your late swaps. This is the highest-value action of the night.

Frequently Asked Questions

What is the most important NBA DFS strategy? Accurate projections plus disciplined ownership. In NBA, projections are reliable enough that the edge comes from being precise (a minute or two of playing time matters) and from mixing a couple of low-owned plays with the night's best value to separate from the field in tournaments.

Should I stack in NBA DFS? Far less than in MLB or NFL. The only positively correlated teammate pairing is assists-to-points; most other stats are negatively correlated because shared shots and possessions are finite. Roster game pairings when it makes sense, but don't optimize lineups around stacking.

How important is late swap in NBA DFS? It's the single highest-value in-slate action. A ruled-out player or a late injury can flip a winning lineup into a losing one. If you can't watch the news until the last tip, it's often smarter to skip tournaments that day.

How do I judge my lineup's ownership in NBA? Multiplying each player's projected ownership together estimates how often that exact lineup gets duplicated, since NBA plays rarely correlate. That's a useful proxy, but it doesn't tell you the build is contrarian. Judge leverage player-by-player: get meaningfully under the field on high-upside pieces and only stay with chalk where it's truly worth it. That player-level edge is what matters in large-field GPPs.

What is the best showdown captain strategy? On DraftKings showdown, put your highest-ceiling play in the captain slot, because it earns the 1.5x multiplier and magnifies the biggest-upside outcome on the slate. Captain popularity has little to do with how often a lineup gets duplicated, so you don't lose uniqueness there. Chase uniqueness in the rest of the build, not the captain slot.

Do I use the same lineup for cash and tournaments? No. Cash rewards the highest-floor lineup built from projections; tournaments reward ceiling, leverage, and lower-owned correlation-aware builds. Run separate lineups for each format.

---

Ready to put this into practice? Get today's NBA projections, ownership, and Boom/Bust ranges on the Stokastic NBA DataHub, then let the Sims and Lineup Generator build and scale your lineups by win probability. Unlock Stokastic+ with code SIMS10 for 10% off your first payment and stop leaving lineups to gut feel. — Alex Baker