An Out-There Proposal to Fixing the NBA Lottery and the Perceived Tanking Problem

Feb 14, 2015; New York, NY, USA; NBA commissioner Adam Silver speaks at a press conference at Barclays Center. Mandatory Credit: Brad Penner-USA TODAY Sports

Tanking is a huge problem the NBA is facing right now– even if you don’t consider it so, just the fact that it dominates the news perpetually is bad for the league. There have been at least a few proposals to fix this issue, perceived or real, the notable ones being aggregating records over multiple seasons, tinkering with the lottery odds and The Wheel.

The solutions that address and try to tinker with the weighing of the lottery suffer from the fundamental flaw that they only dilute the effectiveness of tanking, but do not remove the incentive to do so. All sorts of problems arise from using weighted probabilities to determine draft position– there are arbitrary (at least in the macro perspective of long term franchise success) breakpoints where bad Eastern Conference playoff teams don’t get to join, even if they are not much better than teams that didn’t make the playoffs, stunting their growth. Even if these odds were changed to even out the weighing of lottery balls, it wouldn’t remove many of the unhealthy behaviors that are criticized.

There is a good reason to prop up and help teams that are struggling of course, every city and fan base should be given a reasonable opportunity to quickly turn around their franchise. We want a fair and balanced league where everyone has a shot to turn it around. In essence you can think of the lottery as a sort of re-balancing of the league, and the different weights as how often you want that to happen (and with what variance).

The Wheel suffers from the problem of eliminating that chance of revival, or at least postponing it. If you happen to hit on a three-year period of picking 29th, 20th and 17th when you are bad and messed up lottery picks in the previous years, your franchise is destroyed. Three or five year periods in sports lasts forever, and fans would surely check out. On the other side of that, you might have this season’s Warriors picking in the top five and establishing an unbreakable stranglehold on the league.

Both lottery systems and purely (or nearly fully) deterministic methods have merits and problems that go much further, but at their core one suffers from an incentive issue and the other from the fact that it doesn’t work to balance the league. Although some don’t care about that and think college players should come in as free agents, the vast majority of people who follow the NBA probably agree that we want some mechanism to try and help the bad teams get better.

So the question is how to build an structure that doesn’t suffer from the incentive to lose, yet assists and gives hope to teams that need it? And moreover, what else could be in a system like this to encourage and reward actions we consider good– smart management, player development etc. My proposed solution to this is a bit odd, and I should mention that I’ve only thought about this for a day while trying to figure out a new way to approach the subject.

The first step would be to as accurately as possible to predict regular season win totals based on team strength, and deciding multiple credible people/organizations to take up that challenge and then aggregating those results into a consensus of how many wins each team is expected have by the end of the season. By the end of the year teams will have win totals that differ (and the goal is to have it so that the difference is as small as possible), but instead of rewarding losing, teams would gain draft positions and odds by overachieving. You turn the mechanism upside down, and reward teams that tally up more wins than were projected and punish teams that perform poorly.

For example, take the Phoenix Suns, who might now be fundamentally screwed after a fabulous 2013-14 season. The Suns were the feel-good story of the year– a fun, young, up-and-down team that blew away all expectations and won 48 games, only missing the playoffs because of a historically strong field in the Western Conference. Vegas had the Suns’ over-under before the season at 19.5, the second worst in the NBA, wouldn’t it have been great if the Suns had actually gained the first position in the lottery instead, and have been allowed to continue their momentum towards a better future with the help of a high draft pick for one more year (considering their situation right now)?

The exact mechanism for determining draft positions could be a lottery or an absolute mechanism where you try and surpass teams for better draft picks (wouldn’t that make a fun standings to follow) where teams start off at a certain draft position (the team with the best record with the 30th pick, or weighted probabilities between the last 5-7 picks, provide ceilings and floors like 25 wins and 55 wins), but it doesn’t really matter and that system could be tweaked in a way where the system works the best. What’s key is providing a baseline for where each team is, and rewarding performing as well as possible while keeping the element of helping teams that need it.

This system provides a solution to team’s doing morally ambiguous things to lose (the Warriors going for the Harrison Barnes pick comes to mind), and provides incentive for the organization to do things that improve the play on the court as much as possible, from hiring a good coach to building a team with good chemistry. It’s a system that rewards winning and punishing losing while not losing the fact that bad teams are given a chance for a quick turn around (instead of overhauling the current system, this becomes a balancing issue within the new framework).

Strengths

• Reward teams trying to win as many games as possible
• Provide an extra year of momentum for the feel-good team of the year
• No more incentive to lose

Weaknesses

• Needs a way to adjust for injuries that doesn’t feel like pulling stuff out of your behind
• Weird system for players– the better they play, the better the player who tries to take their job will be (at least theoretically)
• System could become unnecessarily convoluted and complex. It feels a bit weird too.

Additionally, the NBA draft has two defining characteristics from an analytical perspective– the quick decline in value of draft picks, starting from the drop from first to the second pick, and that variance is your biggest friend and enemy. The NBA draft is mainly about hitting home runs, and there are fewer ‘safe’ picks than people realize. And these are the sorts of things that should define the details of current and future systems.

Comment below if this doesn’t seem completely stupid (after thinking about it at least for a minute).