The cost of tanking

For the low low cost of 40 wins.

Looking back over our tanking pieces I realize we haven’t painted a holistic position of our view. That has caused some confusion and for that I apologize. I do want to spell out what we are saying a little more clearly. Before I do though, I want to add one last piece to the puzzle. So that means you may get even more tanking posts!

When we look at what tanking is, it is in essence a tradeoff. As a team you consiously make a decision to trade wins for a better draft pick. In fact, since the advent of the current version of the draft lottery, that is exactly how it works. The worse a team you are, the better your odds are for a good pick. I used the math for this to give you the exact expected draft pick based on your team’s lottery rank:

The value of losing

Expected Pick return for lottery rank
Lottery Rank Value of expected pick
1 2.6
2 3.0
3 3.4
4 4.0
5 4.7
6 5.5
7 6.7
8 7.6
9 8.7
10 9.8
11 10.8
12 10.8
13 12.8
14 13.8

As we can see the worst record in the league doesn’t guarantee you the top pick. Only the worst two teams in the league can feel confident in a top three pick. We now have a very clear expected pick for each lottery rank. Let’s get to the next part.

The cost of a pick

Cost in wins for each lottery slot averaged from 1991-2011
Lottery Rank Relative Spot Cost Playoff win cost
14 -1.9 -1.9
13 -2.8 -4.6
12 -2.7 -7.3
11 -1.8 -9.1
10 -1.5 -10.6
9 -2.1 -12.7
8 -1.6 -14.4
7 -2.4 -16.7
6 -1.7 -18.4
5 -1.4 -19.8
4 -2.7 -22.5
3 -2.8 -24.1
2 -2.4 -26.5
1 -3.0 -29.7

Just to clarify it is -2.0 wins  to go from making the playoffs to the 14 lottery rank, -2.9 to go from the 14th rank to the 13th and so on. This is the average value between ranks for the last 20 drafts.

The cost to get into the playoffs is often thin, only about two wins from making it. After that each spot further down the ladder costs between 1.5 and 3.0 wins to cement. We can also see the total cost. For instance, if a playoff worthy team decided (far in advance obviously) that they would like to aim for #3 pick in the draft: based on the last 20 years of data it would cost roughly 30 wins.

The cost of tanking

Finally let’s put it all together. Here is what our pick relative to the wins it costs looks like.

Playoff pick vs. wins cost
Lottery Rank Pick value Spot Win Cost Total Win Cost
14 13.8 -1.9 -1.9
13 12.8 -2.8 -4.7
12 11.8 -2.7 -7.3
11 10.8 -1.8 -9.1
10 9.8 -1.5 -10.6
9 8.7 -2.1 -12.7
8 7.6 -1.6 -14.4
7 6.5 -2.4 -16.7
6 5.5 -1.7 -18.4
5 4.7 -1.4 -19.8
4 4.0 -2.7 -22.5
3 3.4 -2.8 -24.1
2 3.0 -2.4 -26.5
1 2.6 -3.0 -29.7

Looking at it this way makes our problem a little easier. The first thing I want to point out is that even in the best case scenario a team only gets the best odds for a #1 pick. In fact, the worst team has only won the weighted draft lottery three times.

The second is a cost benefit analysis. The rules of the NBA have mostly guaranteed contracts. Additionally, trade rules prohibit giving players away (unless the team has enough cap and that is rare) so usually teams have to make unfavorable trades to offload unwanted players. The question becomes, is the cost of losing your assets, coupled with the expected return on a rookie performance enough to overcome the drop in wins required to tank?

What makes this analysis hard is that there are indeed players worth tanking for. The Spurs took a 39 win dive to acquire Tim Duncan. He made up for that and more in his rookie contract. Of course, when the Spurs did poorly they weren’t guaranteed Tim Duncan. Their poor record gave them a shot at Tim Duncan but realistically promised them Chauncey Billups (the third pick in the 1997 draft). And while Chauncey Billups would go on to be great, he was certainly not worth 30 wins in his first four seasons.

There is one final tanking aspect I will concede. A full out tank job is clearly not worth the risk. However, in most cases to gain a spot or two only costs a few wins (for instance from 7th pick to 6th) In this case I am curious the value of tanking. What I will say here is that when it comes to most draft picks the position doesn’t matter. Teams are bad at picking out players so the 7th vs.the 6th pick isn’t a huge advantage to the team. The question to me is any potential trade value. In fact that may be the only way to benefit from tanking.

An average player (WP48 of 0.100 and around 2000 minutes of playing time) is worth around 4 wins a year. Most rookies play below average and expecting to get a good performance, even from a good draft pick, isn’t a great strategy. Playing your hot prospect will often not get you back the wins you might have had, had you opted to make a playoff push if it were viable. However, if other teams overvalue a draft pick, especially those in the top five, then perhaps tanking, if your season is already lost, makes sense.

Summing up

Examining all the teams in the current weighted draft lottery format we see that getting a good lottery pick at the expense of a playoff boon is quite costly. In fact, trading wins for lottery picks in most cases is a poor strategy. With that said, I still see there might be cases to slightly make your record worse in order to get better trade fodder. A bad team could make itself slightly worse to get a better draft pick and then make an intelligent trade decision. Of course, that means we are hoping for a team that is bad without tanking and also makes intelligent trade decisions. I suspect such a team is about as rare as finding a Tim Duncan in the draft by tanking.


Comments are closed.