Meet LottoProjector 1.0


With roughly a quarter of the season remaining, the playoff picture has begun to settle. So too have the bottom-feeders begun to emerge.  Perhaps owing to what many projected to be a stellar draft class, the sure fire lottery teams have received an unusual degree of attention.  Indeed, “tanking” has become a household concept, permeating amateur blogs and the Sloan conference alike.

A once largely tepid, or even simmering concern about the incentive to lose has reached a boil.   On the heals of trades that sent Spencer Hawes and Evan Turner packing for very near 10 cents on the dollar, the Sixers have been singled out as the most egregious offender of the unwritten rule that basketball teams must at least try to win.  Of course, Philly’s hardly blazing a trail in uncharted territory.  The life of the powerful incentive to lose coincides with the existence of the draft lottery system.  Its newfound notoriety results not from shifting incentives, but from heightened rationality engendered by the proliferation of analytics, and, more generally, smarter management personnel.

As long as the draft system rewards losing with increased expected lottery value, in most instances, rational teams without a legitimate chance to at least make (and arguably succeed in) the playoffs should prefer to lose.  That’s no revelation.  But with increasing frequency, teams are dispensing with the thinly veiled guise of “trying.”  It says here that this is no cause for alarm — instead of spoon feeding their fans sugar coated absurdities, teams are trusting that fans have become sufficiently sophisticated to handle the truth.   If the league, or particular teams are offended by that candor, well, the answer isn’t likely to be found in charades.  Change the incentives, not the etiquette of responses thereto.

Diatribe concluded, as an unabashed Lakers fan, I’ll admit that I’ve been carefully analyzing their odds of scoring a premium lottery pick.  In that analysis, I found Strength of Schedule to be a woefully inadequate predictor, and developed a new tool to project the lottery standings, “LottoProjector,” exhibited in the table below.  Here’s how it works:

  1. Start with wins to date.
  2. Divide each team’s remaining games amongst one of three categories: (i) vs. Lower Tier teams (“LT”), i.e., those included in the table, (ii) vs. Elite teams, which we’ve defined as Miami, Indiana, OKC, San Antonio, Houston, and the Clippers, or (iii) vs. the Field, which includes all non-LT and non-Elite teams.
  3. Assign a win probability to each category.  
    • For games against Elite teams,, this is simple, as LT teams are unlikely to prevail under any conditions.  We’ve assigned a 5% win probability to this category. (Note: ordinarily, that percentage would be higher — perhaps as high as 15%. Pairing the enhanced incentive for LT teams to lose with the extra motivation for Elite teams to optimize playoff seeding, however, significantly reduces the probability that an LT team will win these matchups.)
    • It’s likewise relatively easy to project a reasonably accurate probability of LT teams winning against the field, which we’ve pegged at 20%, a number that represents Philly’s current winning %.  Given the above stated late season incentives, we feel that this percentage is an reasonable estimate.
    • For games against fellow LT teams, the calculus becomes more complex.  With roughly commensurate incentives at play, factors like home-court and back to back (“B2B”) become more influential.   Thus we’ve assigned a 65% win probability for home games against LT teams (reduced to 52% for B2B games), and a 35% win probability for road games (reduced to 28% for B2B games).


It is by no means perfect — it is an estimate after all.   Off the top of my head, I see several flaws that could be adjusted with further analysis, but are somewhat difficult to entirely cure:

  • Playing the Detroit Pistons obviously is not the same as playing the Milwaukee Bucks.  Some of the LT teams are worse than others.  Trouble is, this varies greatly from night to night, and that variation is likely to become more pronounced as the lottery picture becomes clearer.  That is to say, Cleveland looks great for two games, but once missing the playoffs becomes a statistical reality, performance is bound to change.
  • LT teams are more likely to win at home than on the road against both Elite teams and the Field.  Lottery Projector takes this into consideration.  For instance, an LT team likely has something closer to a 30% chance of winning versus the Field at home, but only a 10% chance on the road.  We spit the difference.  Obviously, this skews the metric a bit for teams that play a disproportionate number of games home or away.
  • In an attempt to secure an optimal playoff opponent, we’re likely to see some “tanking” by playoff teams in their last five games or so.  At this point Lottery Projector does not, and cannot take this effect into account.

Conclusions:  It would take a minor miracle for Milwaukee and Philadelphia to fail to secure the #1 and #2 spots.  The Magic are quite likely to be right behind them at #3, while the Lakers and Celtics appear poised for a photo-finish to determine #4 and #5, as do the Kings and Jazz for #6 and #7.  Conversely, #8-11 remain quite fluid. And finally, the Nuggets appear to be a lock to finish #12 (the bright side is that they own the Knicks’ pick too).

One response

  1. Pingback: Mock Draft 1.0 — LottoProjector Update «

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

%d bloggers like this: