How the NBA Draft Works, the 2026 3-2-1 Lottery Reform, and NBA Expansion

The Basics: Two Rounds, 60 Picks

The NBA Draft is two rounds of 30 picks each — 60 selections in all — held in late June. First-round picks sign guaranteed rookie-scale contracts on a fixed salary schedule; second-round picks are not guaranteed and have more flexible deals. Every team starts with one pick per round, but picks are traded constantly, so the real draft order rarely matches the original team list.

The order splits cleanly in two. Picks 1 through 14 belong to the 14 teams that miss the playoffs — that group is the “lottery.” Picks 15 through 30 go to the 16 playoff teams in reverse order of regular-season record, so the eventual champion picks 30th.

The Play-In Tournament Decides Who Lands in the Lottery

Before any of that is settled, the play-in tournament sorts out the bottom of each conference’s bracket. The 7 through 10 seeds play in: the 7-vs-8 winner takes the No. 7 seed, the 9-vs-10 loser is eliminated, and the loser of 7-vs-8 plays the winner of 9-vs-10 for the No. 8 seed. Whichever teams come out of that without a playoff spot drop into the draft lottery alongside the 10 teams that finished outside the play-in entirely. That is how the 14-team lottery field gets set each year.

How the Lottery Drawing Works

The lottery uses 1,000 numbered four-ball combinations split among the 14 lottery teams by inverse record. The three worst teams get 140 combinations each — a 14.0% shot at the No. 1 pick — and the best lottery team, the 14th seed, gets 5, or 0.5%. The drawing assigns the first four picks. After those four are pulled, picks 5 through 14 fall into reverse-standings order among the teams that did not win one of the top four. That is why the worst team can drop no further than fifth, and why a team can jump up in the lottery but never slide down by more than a few spots.

If you want to watch that play out, the Lottery Forecast runs the full 14-team lottery from live standings, the Team Pick Sim drills into one team’s odds at one pick, and How It Works walks through the simulation math.

Traded Picks and Pick Protections

Because picks move so often, a lot of the draft order is really about which team currently controls each slot. Picks are frequently sent with protections: a pick that is “top-4 protected,” for example, conveys to the new team only if it lands at No. 5 or later — if it lands in the top four, the original team keeps it and the obligation rolls to a future year. Pick swaps, where two teams trade the right to the better of their two picks, work the same way. This is the part of draft mechanics that trips most fans up, and it is why the same pick can be described differently depending on where it lands. The Team Pick Sim resolves those rules pick by pick so you can see exactly when a pick conveys.

Why the League Is Changing the Lottery

The 2026 draft class was viewed as unusually strong at the top, which made the incentive to lose feel sharper than usual. The current lottery still hands the worst teams the best expected draft slot, and the league decided that pull had gotten too obvious to leave alone. After floating several concepts to the Board of Governors, the competition committee, and the 30 general managers through the spring — including versions that would have pushed the lottery to 18 or 22 teams — the league narrowed everything down to a single proposal, informally the “3-2-1 lottery,” for an up-or-down vote.

What the 3-2-1 Lottery Actually Does

The proposal expands the lottery from 14 to 16 teams. The pool becomes the 10 teams that miss the play-in entirely, the four No. 9 and No. 10 play-in seeds, and the two losers of the 7-vs-8 play-in games. All 16 are in the drawing.

The “3-2-1” is the ball count. Thirty-seven lottery balls are handed out like this: the three worst records — the “relegation zone” — get 2 balls each; the next seven non-play-in teams get 3 balls each; the No. 9 and No. 10 seeds get 2 balls each; and the two 7-vs-8 losers get 1 ball each. The headline effect is that the worst team’s odds at the No. 1 pick fall from 14.0% to roughly 5.4%, and being the third-worst team is now slightly better than being the worst. As with the current lottery, only the top of the order is drawn; the remaining lottery slots fall into reverse-record order.

The Anti-Repeat and Pick-Protection Changes

Two more pieces come with it. A team cannot win the No. 1 pick in back-to-back years, and a team cannot land a top-five pick in three consecutive drafts — both aimed squarely at the multi-year teardown. And going forward, teams can no longer protect traded picks in slots 12 through 15, which removes a common safety valve and pushes more picks to actually change hands.

The Vote, the Start Date, and the Sunset

The Board of Governors votes on the 3-2-1 proposal on May 28, 2026, and it needs 23 of 30 teams to pass. If it clears, it takes effect with the 2027 NBA Draft and carries a sunset provision: the system expires after the 2029 draft, so owners get a built-in checkpoint to renew it, tweak it, or move to something else before the next collective bargaining agreement.

Will It Actually Stop Tanking?

It moves the incentive rather than erasing it. Flattening the bottom three and capping repeat winners makes bottoming out for a single season far less rewarding, and killing 12-15 protections adds real risk to trading future picks down. The trade-off is that the new structure rewards landing in roughly the 4-through-10 band, and it hands play-in teams a live lottery ticket, so the strategic question shifts from “be the worst” to “stay in the sweet spot.” That is a milder, more spread-out version of tanking, which is probably the point. You can pressure-test it yourself: in the Lottery Forecast, the button labeled “Run · Proposed 3-2-1 Rules” reruns the current standings under these odds and rebuilds the mock draft around the result.

Where Things Stand: Seattle and Las Vegas

On March 26, 2026, the NBA Board of Governors voted to formally begin the process of exploring expansion to two cities: Seattle, Washington and Las Vegas, Nevada. This was not a vote to add franchises. It was a vote to authorize an evaluation process, with investment bank PJT Partners hired as the league’s strategic adviser to examine ownership candidates, arena infrastructure, and the broader economic implications of going from 30 to 32 teams. A final decision on whether to proceed, and with which ownership groups, is expected later in 2026, with the earliest possible debut for new franchises being the 2028–29 season.

Seattle is the emotional favorite because the SuperSonics history still hangs over the league, and Climate Pledge Arena is already there. Las Vegas is the economic favorite because the city already works as a major sports market and the revenue upside looks huge. The league clearly sees both as premium expansion destinations, not fallback options.

Each franchise is expected to sell for between $7 and $10 billion. That is a huge reason expansion is moving so smoothly. Owners would be diluting future revenue, but they would also be collecting a massive upfront check.

How NBA Expansion Drafts Have Worked

The NBA has added teams twice in the modern era, and both times used a version of the same expansion draft structure. Understanding what happened then is the clearest guide to what Seattle and Las Vegas can expect in 2028.

The basic structure is familiar. Existing teams protect up to eight players. Anyone left unprotected can be taken by the expansion teams. Seattle and Las Vegas would likely alternate picks, and each would come out of the process with a starter set of NBA-caliber players plus an early draft pick in its first rookie class.

The real tension is in the eighth protected slot. Good teams almost always have more than eight players they would rather keep, which is why expansion drafts tend to hurt depth-heavy contenders more than bad teams.

Consequences for Teams: Winners and Exposed Franchises

The teams most exposed are not usually the worst ones. They are the deep teams with real ninth, tenth, and eleventh men. Contenders and strong development programs have the hardest protection decisions because they have more real NBA players than they can keep off the board.

The league-level upside is obvious: two new markets, more roster spots, cleaner conference symmetry, and a repaired relationship with Seattle. The main complication is that expansion and anti-tanking reform are now tied together. If Seattle and Las Vegas arrive by 2028–29 and struggle early, the 3-2-1 lottery rules — assuming they pass — will immediately be tested in a 32-team league.

Which Teams Will Become the Expansion Franchises?

As of May 2026, no ownership groups have been officially selected. The league has authorized the process, not awarded the teams.

Seattle still looks like the cleaner case because the arena question is basically solved and the SuperSonics identity is sitting there waiting. Las Vegas looks more open-ended because the ownership field is wider and the arena situation still needs more clarity.

Who gets the teams will matter. A big-spending ownership group can try to accelerate relevance. A patient group may be more comfortable taking early losses and building through the draft. Either way, the new anti-tanking rules will shape those first years immediately.