My thoughts on http://www.grantland.com/blog/the-triangle/post/_/id/86940/the-nbas-possible-solution-for-tanking-good-bye-to-the-lottery-hello-to-the-wheel...

One problem with this is that it assumes talent is arranged linearly throughout the draft—the difference between slot #s is proportional to the difference in talent. But NBA talent isn’t linear, it’s more like logarithmic:

http://www.basketball-reference.com/blog/?p=2740

Using that chart, the biggest difference between any two teams in a 5-year period would be 5 draft slots (one team’s picks would add up to 75, another’s to 80), but the biggest difference in expected value would be 37.6 wins (one team would have 174.3 expected future wins from its draft slots, while another would have 136.7).

Likewise, even in a 10-year period there would be differences of more than 22 future expected wins between two teams despite the biggest difference in cumulative draft position being 3 slots. Only over the full 30-year period would it completely even out.

In practical terms, this means you’d be able to identify blocks of 5-10 years where you knew your team was at a systematic disadvantage in the draft. Whether that would impact fan behavior is anyone’s guess, but it cuts into the seeming balance of short-term and long-term fairness in this proposed system.

Reply · Report Post