clock menu more-arrow no yes

Filed under:

Fixing the NBA Draft: Auction-Style

New, comments

Here's one man's humble opinion on how to fix the draft, eliminate tanking, promote smart team building, and make Draft Night even more fun.

Derick E. Hingle-US PRESSWIRE

Fixing the Draft seems to be a constant topic amongst the NBA community. On the heels of the Philadelphia 76ers overt 2014 tank job, the NBA has made noise this offseason about altering the Draft Lottery system to discourage tanking. The NBA's proposal is to essentially keep the same lottery system, but to flatten the Lottery odds marginally so the worst team's odds at a top pick were about half what they are under the current system.

One problem: this proposal to eliminate tanking doesn't eliminate tanking.

Currently there is virtually no incentive to tank from, say, the 13th worst record to the 11th worst record. Yet if you make every pick more likely to garner a top three to six pick, tanking from the 13th best odds to the 11th best odds may triple your chances at "winning" the lottery.

In essence, the incentive for tanking is merely shifted from the worst teams to the not-quite-as-bad teams. Hardly a solution.

Zach Lowe first reported on the infamous Draft Wheel, which is a step in the right direction. Each team is placed randomly on a wheel of thirty draft slots. Each year teams move one spot along the wheel, and the wheel is divided as evenly as possible ensuring that every five years every team would have a top six selection and a bottom six selection.

This proposal essentially fixes the draft order for the next thirty years, and proponents of the system likely welcome the certainty that system brings to team building.  If you know you have a number one pick coming up in three years, you can build that into your strategy, rather than relying on the grace of ping pong balls.*

* And that, at it's very core, is what all Lottery systems are. Grace. These teams don't deserve a reward. They've chosen to be bad (sometimes through questionable logic) because there is the potential of a reward.

The problem with building that much certainty into a system is that it levels the playing field among NBA front offices. Uncertainty favors those that manage it best, and that ability tends towards the smarter people in the room. Personally, I like watching people manage uncertainty; it's exciting, dynamic, and you can do everything right and still lose. Any economist will tell you that markets - whether we're talking about natural gas prices, or the stock market, or the NBA labor market - are some of the least certain territories we have. Put enough independent actors bartering over the same finite resources, and someone is bound to throw an unpredictable kink into the system.

So let's change the NBA draft. Let's make it an auction.

That's right, an auction. First of all, auctions make everything better (gambling and cheese have similar effects). I first heard this idea on the Boxscore Geeks Podcast, and it is truly brilliant. Keep the proposed Draft Wheel, but instead of letting the wheel determine draft order for thirty years, it determines the allocation of "draft credits" for thirty years. Every year the Draft Wheel allocates thirty "draft credits" to one team, but just one "draft credit" to another team. Because the order is fixed, teams know in advance, for thirty years, just how many "draft credits" they will have.

Each team then bids on players with their "draft credits," and the team who bids the most wins that player. Teams already view draft picks as currency. Why not make it official?

The Benefits of an Auction Draft

Switching to an Auction Style draft achieves the following goals:

1). The system favors the smartest people in the room - In fact, any team lacking not only an analytics department, but also specialists in game theory and probably behavioral economics, would be playing checkers while the rest of the league played chess. Frankly, I'd probably hire at least one professional poker player - they are specialists where game theory and projecting ranges of actions are concerned.

I'm a big fan of interdisciplinary thinking. Grantland did some segments on the Kings crowdsourcing their draft process, which led to the hilarity of watching NBA stat nerds talking through analytic presentations to Mitch Richmond and Chris Mullin, one of whom reacted like his kid had just gotten a walk in Little League.

2). A More Perfect Currency - Currently, draft picks are the secondary currency of NBA trades, used as makeweights to balance trades of all types. The problem with using draft picks as currency is that they aren't really currency. Draft picks only have two denominations: 1st round picks and 2nd round picks. Essentially, this currency operates as if dollars only came in $5 bills and $100 bills. To create all of the bills between, teams attach complicated protections on picks to create a pick with a value between a 1st and 2nd round pick. An unprotected 1st round pick > top 5 protected pick > lottery protected pick > 2nd round pick.

And these protections encourage tanking. They also tie up a team's ability to trade another pick while they wait for the pick protection to come due. The Memphis Grizzlies find themselves in this exact scenario. They owe a pick to the Cleveland Cavaliers that has cost them the ability to trade a pick (except in a few utterly Byzantine scenarios which severely limit its worth) until that pick comes due.

Two years ago, if Memphis had just paid Cleveland seven draft credits to take Wayne Ellington and Marreese Speights off our hands, Cleveland would have been able to use them immediately, and Memphis would be free of the debt.

To further the currency metaphor, draft picks have an expiration date as a currency, a "must use by" date: the NBA Draft. Think if our currency worked like that. Your dollar is good until a random day in the Summer at which point you must spend it or you lose it.

3). The system favors long-range planning - Allowing "draft credits" to roll over from year to year allows teams to choose when they would like to cash in their warchest of assets. How many teams would have preferred to save their draft credits from the 2013 draft, rather than spending them on players they didn't even want.

The sad truth is that basketball talent is not evenly distributed every year. We could even tie players' rookie salaries to the number of "draft credits" that were used to acquire them. Why should Anthony Bennett be paid like he's Anthony Davis? Or a better question: why should Anthony Davis be paid as if he's Anthony Bennett?

4). More dynamic team building - Currently the team with the number one pick faces a similar decision every year. Only a handful of players are ever seen as "worth" the number one pick (even though, more often than not, the most productive player in any given draft is not one of the top two picks). At any given pick, teams are projecting which players will fall to them.

Switching to an auction style draft changes this dynamic completely. For instance, suppose Cleveland had been allotted thirty "draft credits." Their decision is now not "Andrew Wiggins or Jabari Parker." It is "should we spend thirty draft credits on Wiggins or Parker" or "should we spend half of our credits on Noah Vonleh and the other half on Nick Stauskas"?

5). Opportunity Cost is Magnified - Opportunity cost - the left turn you eschew when you veer right - is the hidden cost of team building in the NBA. Trading a future draft pick once means you can't trade it again. Every dollar on your cap sheet is one less you can spend in the future.

But currently, NBA draft picks have expiration dates, meaning they must be spent once a year, regardless of whether you want to spend them or not. Switching to a "draft credit" system removes that artificial expiration date on the currency, and conversely, makes spending it an actual, you know, decision.

How It Would Work

There is still plenty of room to tweak this system, but my rough idea would be as follows:

1). The team with the most "draft credits" starts the NBA Draft Auction bidding on the player of their choice. After that player is won, the team with the second most credits at the start of the draft starts the bidding on the player of their choice. This process continues until thirty players are selected. This ensures that, at the very least, thirty players are selected.

The "2nd round" is handled the same way, except teams are allowed to pass if they would rather not bid on a player. Therefore a maximum of sixty players are selected, whereas if every team opts to pass, only thirty are selected.

2). Trading "draft credits" is closed the day of the draft. This essentially freezes the currency the day it is spent, and prevents teams from acquiring more currency during the auction.

3). The NBA draft is televised the same way. Now, instead of waiting to see which players are left for your team's pick, your team has, in theory, the ability to come out of the draft with any player. Or more exactly, your team could be involved with every selection. How much more exciting is that?

4). The maximum "draft credits" a team could stockpile would be capped at sixty, or double a one year maximum allotment. This would be the only "expiration date" in place. "Draft credits" are there to be used, not stockpiled to infinity. Making the maximum too high would incentivize saving them, at least for the first couple years.

5). If two teams have identical amounts of "draft credits," the first team to bid their max would win the bidding (i.e., there is no tiebreaker for two max bids. Whoever bids first wins).

6). In a perfect, non-CBA constricted world the number of "draft credits" would also determine a draftees salary. In a given year, the highest rated player could be bought for sixty "draft credits," whereas the following year over thirty players could be bought for a minimum of one credit each. If the market dictates how much each player is worth in credits, that should carry over to salary, no?

So perhaps it makes sense to give the highest valued player in a given draft the standard salary of a number one pick (the player who goes for the highest number of credits, not the player who gets bought first). In theory, the number one pick should go for thirty credits, the number two pick twenty-nine, etc. Perhaps every credit a player is bought over their slot would be given a 3.3% salary bonus, since one credit basically represents 3.3% of the number one pick.   For instance, if the number two pick were bought for thirty credits (i.e., one over the twenty-nine credits the number two pick is "worth," his starting salary would be 3.3% higher than a normal number one pick). In this system, the maximum salary a number one pick could make, year one, would be 200% the normal salary of the current number one pick, or about $11mm. Not franchise crippling money, but also not chump change.

So there you have it. A new, more perfect NBA Draft process. You're welcome, Commissioner Silver.