Picking the Conference Finals and Playoff Science

Posted on May 18, 2009 by


After the games on Sunday I believe I have now taken over first place in the TrueHoop Stat Geek Smackdown.  And this means… well, let me make my Conference Finals picks and I will then discuss what this means.

Once again, my approach is quite simple.  All I am considering is a team’s efficiency differential (offensive efficiency minus defensive efficiency), home court advantage, and any relevant injuries.  With this approach I have correctly chosen the winner in every series except for Portland vs. Houston.  And now I am going to apply this approach to the Conference Finals.

Cleveland Cavaliers vs. Orlando Magic

The Magic finished the season with a 7.1 differential while the Cavs led the league with a 9.7 mark.  Given these numbers, the Cavs – who have the home-court advantage – should win in six.  But the Magic’s efficiency differential for the season was inflated by the play of Jameer Nelson.  The Magic’s differential in the first half was about twice what we observed in the second half (when Nelson didn’t play).  If we take the second half numbers as more indicative of the Magic’s true quality – and their play against the Celtics suggest this is true – then the forecast should change to…

Pick: Cleveland over Orlando (4-1)

LA Lakers vs. Denver Nuggets

Dean Oliver is a friend of mine and he works for the Nuggets.  And I would like to see Dean and his team advance to the Finals.  But the numbers suggests otherwise.  So although I will be rooting for Denver, my forecast will be…

Pick: LA Lakers over Denver (4-1)

Okay, those are my picks.  Now let me put this contest in perspective.  The biggest story from this contest is the similarity in everyone’s picks.  There have been 12 series to date and in eight of these everyone picked the same winner.  This consistency reflects the fact that everyone is essentially looking at the same thing. I think all of us agree that teams are best evaluated by looking at points scored and surrendered per possession.  And since we agree on this point, we tend to agree on the identity of the “better” team in each series. 

Of course, despite such agreement, I am currently in the lead.  This must mean I know a little more than everyone else. 

Although I like that story, it really is just a story.  In other words, my current lead is probably just luck.  A key component of this contest is the requirement that we pick the number of games in each series, and although I think the data helps somewhat with that question, I am not sure it helps that much.

It’s important to remember that despite what you hear on television, the playoffs are not really designed to identify the best team.  Currently I am reading The Drunkard’s Walk: How Randomness Rules Our Lives by Leonard Mlodinow.  This wonderful book contains the following passage relevant to any discussion of predicting the winner in a best-of-seven playoff series.

“…if one team is good enough to warrant beating another in 55% of its games, the weaker team will nevertheless win a 7-game series about 4 times out of 10.  And if the superior team could beat its opponent, on average, 2 out of 3 times they meet, the inferior team will still win a 7-game series about once every 5 match-ups.  There is really no way for a sports league to change this.  In the lopsided 2/3-probability case, for example, you’d have to play a series consisting of at minimum the best of 23 games to determine the winner with what is called statistical significance, meaning the weaker team would be crowned champion 5 percent or less of the time.  And in the case of one team’s having only a 55-45 edge, the shortest significant “world series” would be the best of 269 games, a tedious endeavor indeed! So sports playoff series can be fun and exciting, but being crowned “world champion” is not a reliable indication that a team is actually the best one.” (p. 70-71).

For the TrueHoop contest we are each offering our evaluation of who the better team is in each series.  And then the series is played to see if we are “right”.  But the words of Mlodinow remind us that a seven game series is simply not up to this task.  In sum, the playoffs are about fun, not science.

In this current contest I expect all of us to pick Cleveland and LA to reach the Finals.  So I should still rank among the leaders when we get to the last round.  Assuming the Conference Finals go as expected, when we get to the Finals we will be mostly guessing.  There simply is not that much of a difference between Cleveland and LA.  So the winner of this contest is really going to be the person who ranks among the leaders entering the Finals and manages to guess right on the outcome of a series between two close competitors. 

So this contest really is not a “test” of anyone’s ability to evaluate teams.  Again, this is because a) I think we essentially have the same evaluation and b) the playoffs are simply not designed to test that evaluation.   Of course if I get this test right, then we will forget everything I just said and conclude that I really do know something :)

– DJ

The WoW Journal Comments Policy

Our research on the NBA was summarized HERE.

The Technical Notes at wagesofwins.com provides substantially more information on the published research behind Wins Produced and Win Score

Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:

Simple Models of Player Performance

Wins Produced vs. Win Score

What Wins Produced Says and What It Does Not Say

Introducing PAWSmin — and a Defense of Box Score Statistics

Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.