A Plan for NBA Expansion That Finally Solves the Tanking Problem

If it’s all right with you, I’d like to just skip the intro. I’ve read enough columns, blog posts and manifestos about NBA draft lottery reform—and I’m guessing you have, too—to know this is the part where they usually start with an explanation of why reform is needed and some particularly egregious examples of shameless, blatant tanking. I think, given how much it’s been in the news lately (first, commissioner Adam Silver came out and said reform was coming, then ESPN reported three potential reforms that will be discussed at a Board of Governors meeting in May, then the NBA Players Association countered with its own set of tweaks to consider) we can just cut right to the good part.

The bad news is, I don’t think any of the ideas I’ve seen batted around actually eliminate the warped incentive structure where teams benefit from losing, the toughest but most crucial nut to crack in this long-running debate. The good news is I think I have a plan that does.

The other piece of good news is that, although my plan would require a few years to implement, which wouldn’t satisfy Silver’s desire to make change immediately, it would specifically take into account the other massive piece of NBA news that has been floating around lately: that the league is planning to expand to 32 teams.

Let me first explain what I’m currently calling the Goldich Plan (though I’m open to better names), and then I’ll explain the benefits. I’ll get around to addressing some of the flaws in the plans actually under consideration a little later, too.

Ground rules

To start with, let’s make the assumption that the NBA is going to expand from 30 teams to 32, which everyone seems to believe is going to happen sooner than later. Let’s also assume the league’s 32 teams will be split into eight divisions with four teams each, just like the NFL. That seems most likely and is necessary for my plan.

Finally, I’m going to focus on the 16 teams that miss the playoffs. I think the teams that make the playoffs should slot into the final 16 picks of the first round. My preference would be to order those picks NFL style, based on playoff finish and then slotted in by record. But you could also stick with the NBA’s current house style of slotting all playoff teams in by regular-season record, regardless of how they do in the postseason. That doesn’t really impact my plan for the non-playoff-bound teams. For my argument’s sake, let’s also assume they keep a version of the play-in tournament, and we’ll treat play-in losers like all the other teams that miss the playoffs.

O.K, O.K., on to the new and interesting stuff.

The Goldich Plan

Step 1: Calculate the cumulative regular-season records for all eight divisions.

Step 2: Give all non-playoff teams within any division the same lottery odds, regardless of where they finish within the division.

Step 3: Give the best odds in the lottery to each of the non-playoff teams from within the division that has the best cumulative regular-season record. Give the second-best odds to the non-playoff teams from the second-best division, etc., etc.

To illustrate, here are some fake post-expansion standings with new teams in Seattle and Las Vegas. Don’t get distracted by how I broke up the divisions; that’s a debate for another day.

Proposed win/loss records for NBA teams — Sports Illustrated

I’ve also gone ahead and broken down which divisions have the most cumulative wins to set the fake lottery order. Again, remember it’s just the lottery order, not the guaranteed draft order!

Proposed NBA records for an anti-tanking policy — Sports Illustrated

You may have questions, so let me explain what I see as the benefits.

Benefits

1. The No. 1 benefit is that you eliminate the incentive to lose games.

Under my plan, teams actually have a better chance in the lottery the more games they win, because they are contributing to their division’s better overall record. Teams are incentivized to be competitive all season. They don’t go into a season hoping to lose. They don’t start trying to lose six weeks into a season when they realize it’s hopeless. They don’t shut down a star player for six weeks at the end of the season. They should try to win as many games as possible. I think this solves what is, if not the league’s biggest issue at the moment, one of its most widespread negative issues.

This is also why it’s crucial to my plan that every non-playoff team within a division receives the same lottery odds, even if one team finishes two games out of a playoff spot and one finishes 20 games out. You want your division to rack up wins, and there’s also no incentive to finish fourth in your division instead of third.

Teams will still be bad. Teams can still prioritize developing young players. They can still sell at the deadline, trading players for future picks. Those concepts—which Silver was probably referring to when he talked about the difference between a rebuild and a tank—can still be beneficial. But controlling for all other variables, a win is more valuable than a loss, even for a bad team. That’s a major difference.

2. Teams getting a boost with better odds in the draft are still teams that need it.

What is the purpose of the draft? You could argue it’s for competitive balance, league parity, offering glimmers of hope to small-market teams or, if you’re a cynic, that it’s a nefarious scheme to suppress rookie salaries. But I think the No. 1 motive is, in some form, to give help to the teams that need it.

My plan still does that, just in a slightly different way than the league’s current system does. Instead of giving the most help to the teams with the worst records, the Goldich Plan is often going to tilt the balance in favor of the teams stuck in divisions with really good teams.

You could certainly argue that a truly terrible team in a mediocre division needs help more than a pretty good team in a deep and competitive division. But if you’re stuck in a division with the peak Curry Warriors (or if the Thunder and Spurs are paired up in the same four-team division), your situation may feel just as hopeless. Under my plan, those great teams are going to pull a division’s record upward and the non-playoff teams will get some help. In the model above, the best odds in the lottery would go to Utah (40–42, the 19th best record in the league) and Memphis (29–53, tied for 24th). You may prefer worse teams to have better odds, but I think it’s a reasonable enough trade-off to fix the incentive structure.

Are you worried that if the teams with the best odds win the lottery it would concentrate too much talent in a single division? That doesn’t bother me. We can have a great division (like the NFL’s NFC West right now, or historically MLB’s AL East) with a rising tide forcing teams to compete with each other.

One thing that’s kind of lame about the current NBA is how little divisions matter at all. Major outlets, including the league’s own website, default to showing the standings for each conference all year, unlike leagues where divisional rivalries and standings are much more central to the leagues’ structure and ethos. Suddenly under the Cumulative Division Record Plan (see why I’m sticking to Goldich Plan?), the divisions are very important. Playoff-caliber teams still want their rivals to lose as they jockey for seeding, but lottery-bound teams don’t mind if their divisional foes win some games that boost the division’s cumulative record. One interesting wrinkle of the cumulative records is that intradivisional games won’t count toward the lottery odds at all (the division is guaranteed to go 1–1 in those games), so you never have a reason to throw a game. But, I’m banking on those games impacting the playoff standings and further entrenching rivalries. I also imagine teams will play their division-mates more often. Another debate for another day, but you can make an 82-game schedule by playing six games against each of your three division rivals, four games against each of the other eight teams in your conference and two games each against the 16 teams in the other conference. That volume alone would make intradivision games matter a great deal to the standings even if they didn’t impact the lottery odds. And playing more games against your divisional opponents—which I think would be good for the league and could cut down on some travel, another issue the league has discussed—gives even more reason to help teams like Memphis and Utah in my example above. Being stuck in a tough division matters even more when an increase in divisional games makes those teams more directly block your path to the postseason.

3. The lottery still adds some uncertainty.

Remember, it’s not a guarantee that the teams from the best divisions will get the first crack at the best players in the draft, because I have intentionally left the lottery in place. First, because I like the lottery. As do millions of other people, I think. Second, because I do think a little uncertainty over the order prevents direct manipulation that would be harmful to the regular season.

The NFL deals with tanking, most commonly in the final weeks of the season when teams know that specific outcomes can lead to specific draft picks. With the Goldich Plan leading not to specific draft slots, but merely to specific lottery odds, it does at least dampen the incentive to try something similar. It eliminates a situation where a team can, say, tank a game late in the year that would force them to drop from the No. 5 seed to the No. 6 seed when doing so would also prevent their division rival from getting the No. 1 pick. The relationship between wins and draft slots isn’t quite that direct.

Having a cumulative effect where four teams have an equal impact on where each team slots in the lottery also lessens the ability of any one team to be an outlier (or bad actor) that has an outsized effect on the odds—which oftentimes in the current NBA forces other teams to try to keep up.

Why it’s better than other plans

My issues with the proposals currently on the table are that they still incentivize losing if the five or 10 worst teams in the league still get the best odds in the lottery, as two of the league’s reported proposals offer. And that’s true even if you use a two-year record, as the third one prescribes. That proposal also includes first-round playoff losers in the lottery, which could give top picks to teams that are way too good to receive them. (Like, for example, if the Pistons got the No. 1 seed in the East but lost in the first round because Cade Cunningham was injured.)

I’ve read up on many other proposed reforms kicking around the internet (Dan Clayton at Salt City Hoops has a very comprehensive list published before Silver officially put three on the league’s agenda), and of the plans that have generated a lot of attention over the years, I can think of a few that, like mine, tried to eliminate the incentive to lose. One is to simply give the best non-playoff team the top pick (or best odds for the top pick in a lottery). But I do think that has the chance to leave teams at the very bottom truly stuck in a hole that’s hard to dig out of. It would also likely result in teams tanking their way out of the playoff picture entirely (a little more on that later, but rest assured my plan does not incentivize teams to do that as much).

The other popular proposal that forces everyone to play hard through April is to give teams lottery points based on how many games they win after being eliminated from playoff contention. In addition to being a little too convoluted and difficult to follow, I think that creates negative incentives to simply tank earlier in the season. And it has consequences for teams that have genuine major injuries either in the first half or the second half. The same is true for the similar, simpler idea of freezing the lottery odds earlier in the season.

Speaking of convoluted, I did see the one going around social media recently about forcing every team to trade its pick to a different team. I think on a scale from one to 10 where one is sticking with the current system and 10 is the radical plan to make every team trade its pick every season, mine is relatively tame and simple.

My elevator pitch is that it’s a pretty elegant solution that incentivizes every team to win as many games as possible, gives hope to teams that need it, preserves the quirkiness and uncertainty of the lottery, and isn’t too hard to follow or understand. I incorporated the numbers pretty easily into my fake standings above.

Loose ends

There are some loose ends my plan still needs to tie up. I will defer to someone with more of a math background to figure out how we should weight the lottery odds for the 16 teams in it, keeping in mind there will be groups of division-mates tied at various levels. I come bearing a framework, but not all the tiny details.

I also want to acknowledge there is one problem I haven’t been able to solve. I said I have virtually eliminated the incentive to tank, but there is one rather insidious version of it I haven’t, and that’s teams missing the playoffs on purpose. I don’t think I’ve stopped the 2023 Mavericks–style late-season tank, positioning yourself to miss the playoffs and enter the lottery instead of going in as a low seed. I don’t know the way around that one (some have suggested changing rules on pick protections). I think my plan solves most tanking, but probably not all.

I’m also not naive. I’m not expecting this to make every regular-season game feel like it has playoff intensity. Even good teams still have so-called schedule losses and clunkers or occasional no-shows in the middle of the season. Plus, a lot of players won’t be particularly motivated just to improve the lottery position of a team they may not even be on the following year (which is the issue with any proposal that has some sort of tournament where teams play specifically for draft positioning, which I’ve also seen floated unrealistically.) It won’t eradicate load management or back-to-backs, or change anyone’s mind if they feel the regular season is simply too long to matter when the best teams can separate themselves with four rounds of best-of-sevens come playoff time.

But it would stem the most widespread and egregious examples of tanking, which have become such a pressing issue, looming over the whole season.

Benefit of waiting

Finally, I know Silver has said he’d like to implement reform now, but if I may imagine a world where he chooses this plan and announces it’ll be instituted after the 2028 season, when the league likely expands (even with one of the other measures instituted in the short term), I think there are actually a few benefits.

First, a lot of teams have already traded picks in the upcoming drafts. Some of them might not like the idea of a radically different lottery plan being implemented this offseason or next with little to no warning. (This is a minor point, given that picks beyond the expansion date have already been dealt, and will continue to be dealt, and it’s a fact of NBA life that the value of picks can change after you deal them.)

More importantly, it feels, anecdotally at least, like the early sentiment on expansion is more negative than it would have been three to five years ago. A lot of people, from fans on social media to NBA thought leaders with large platforms, are saying it’s not a great time for expansion because of some of the very issues lottery reform is designed to fix.

Pairing a new style of lottery reform with expansion and divisional realignment may be a way to increase enthusiasm for what could be a significant overhaul of the league’s structure, with the divisions, schedule balance and draft reform all coming at once. Maybe the best way to build national excitement for two new teams is to kill two birds with one stone: We’re expanding the league, realigning divisions and finally have a plan to fix the issue of widespread tanking.

I think I may have come up with a way to do it.

More NBA From Sports Illustrated

Listen to SI’s NBA podcast, Open Floor, below or on Apple and Spotify. Watch the show on SI’s YouTube channel.