Sunday, May 15, 2011

Heat Check: Previewing the 2011 Eastern Conference Finals by the Numbers in the Matchups

I’m the lone member of the Wages of Wins Network that thinks the Miami Heat will defeat the Chicago Bulls in the Eastern Conference Finals. Here’s a rundown of the WoW Network’s picks:

With all of these statheads picking the Bulls, I must be a homer that’s ignoring the numbers, right? That’s where you’re wrong.

I’ve posted a few different previews of this series (see here and here) but I haven’t posted any Heat Produced stats yet. That changes today.

Let’s take a look at two sets of numbers for this series and see who’s wrong:
  • First, the season series won by the Bulls 3-0 and 
  • Second, the matchups at each position. 

This article will use Win Score and Estimated Wins Produced, statistical models created by Professor David Berri from the Wages of Wins Journal, to measure how much a player's box score statistics contributed to their team's efficiency differential and wins. An average player produces an estimated 0.100 wins per 48 minutes (Est.WP48), a star player produces +0.200 Est.WP48 and a superstar produces +0.300 Est.WP48. More information on these stats can be found at the following links:

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

Season Series Notes:
This spreadsheet lists the wins produced stats estimated from box scores of three games against the Bulls this season.

  • Not sure how much the season series says about the Heat since only four players appeared in all three games while nine Bulls players appeared in all three games. 
  • Over three games, both teams battled to a standstill in the first quarter with an estimated 0.5 wins produced each, but the Heat dominated the second quarter with an estimated 0.9 wins produced compared to just 0.1 for the Bulls. The Heat outscored the Bulls by an average 27.5 points per 100 possessions in the second quarter. 
  • An estimated 69 percent of the wins produced in the second quarter came from LeBron James (1.385 est.WP48!), Dwyane Wade (0.468 est.WP48) and Eddie House (0.456 est.WP48). 
  • The big difference between these teams came after halftime when the Heat were outscored by an average 25.4 points per 100 possessions in the second half. The Heat produced an estimated -0.6 wins in the second half while the Bulls produced an estimated 1.6 wins. 
  • The most productive Heat player in the second half was Chris Bosh with an estimated 0.2 wins produced and 0.185 est.WP48. 
  • The least productive Heat player in the second half was Mike Miller with an estimated -0.3 wins produced and -0.420 est.WP48. 
  • This spreadsheet details Heat players’ production by quarter against the Bulls. 

  • Looking at the games across all four quarters, the most productive player in the season series was LeBron with an estimated 0.7 wins produced and 0.395 est.WP48 from 31.7 points on 22.5 shots and 6.9 free throw attempts with 10.4 rebounds, 2.3 steals, 6.3 assists and 2.3 turnovers (all stats are per 48 minutes). That’s just two games worth of work.
  • Mario Chalmers was the second-most productive Heat player with an estimated 0.4 wins and 0.179 est.WP48 from 15.8 points on 11 shots and 2.4 free throw attempts with 3.8 rebounds, 2.9 steals, 5.7 assists, 2.4 turnovers and 6.2 fouls (all stats per 48 minutes). As I mentioned in the point-counterpoint with Big Shot Ron (a blogger and Bulls fan), Chalmers has played very well against Rose in his career as this spreadsheet illustrates.

  • Chalmers was most productive of the four Heat players that appeared in all three games against the Bulls. Here’s the production of the other three:
    • D Wade: 0.3 est.WP, 0.130 est.WP48 in 115.7 minutes
    • C Bosh: 0.1 est.WP, 0.046 est.WP48 in 114.2 minutes
    • J Jones: -0.3 est.WP, -0.278 est.WP48 in 43.3 minutes
  • The most productive Bulls player against the Heat in the regular season was Carlos Boozer with an estimated 0.5 wins produced and 0.231 est.WP48 from 19.0 points on 19.5 shots and one free throw attempt with 13.8 rebounds and 3.3 assists (all stats per 48 minutes).
  • Derrick Rose was just the fourth-most productive Bulls player against the Heat with an estimated 0.2 wins produced and 0.101 est.WP48 from 35.3 points on 30.0 shots and 7.7 free throw attempts with 4.1 rebounds, 7.7 assists, 1.2 blocks and 3.7 turnovers (all stats per 48 minutes).
  • The Bulls’ centers had their way with the Heat in the season series with an estimated 0.6 wins produced and 0.206 est.WP48. Here’s where that production came from:
    • O Asik: 0.378 est.WP48, 0.4 est.WP
    • J Noah: 0.181 est.WP48, 0.2 est.WP
    • K Thomas: 0.338 est.WP48, 0.2 est.WP
  • matchup in the regular season. The Bulls big men beat the Heat on the glass by an average of 11 rebounds per game.
  • The other position the Bulls won in the regular season was shooting guard. The majority of those minutes went to Keith Bogans (0.116 est.WP48), Ronnie Brewer (0.261 est.WP48) and Luol Deng (0.043 est.WP48).

Matchup Notes:
In the regular season, the Heat only won the matchup battle at two positions — point guard and small forward — but there are two problems with focusing on the season series. The first problem is that three games is a small sample from an 82-game season and the second is that only four Heat players appeared in all three games.

Let’s take a look at how the Heat match up with the Bulls considering all 82 games. This spreadsheet lists the Win Scores produced by both teams and their opponents at each position in the regular season and playoffs and estimates what per-minute production (WP48) should be expected in the matchup*.

This is the first time I’ve used this methodology, but one thing I like about it is that it said the Heat had a 15 percent win margin over the Celtics, which means the most likely scenario for the Eastern Conference Semifinals was the Heat winning in five games. It also predicted the Heat winning the matchups at C, SF and SG. That’s exactly what happened in both cases.

This new method said the Heat should’ve produced an estimated 3.3 wins in five games against the 76ers and won the matchup at every position. The Heat actually outperformed that prediction by producing an estimated 3.6 wins but did not win the matchup at point guard with Bibby turning in the least productive performance of any player in the series.

What does this new method say about the Eastern Conference Finals?
  • The Heat are only expected to win two matchups - SG and SF. Despite what happened in the regular season, I expect Wade to have a big series against the Bulls just like he did against the Celtics. I don’t think anyone had concerns about LeBron in this series and I actually expect him to play better than predicted.
  • The PF matchup will be close, but the Bulls limit the production of opposing PFs better than any other team in the NBA. As a result, the Heat are expected to get 22 percent less production from that position in this series than they’ve gotten so far.
  • Center is expected to be the least productive position for the Heat at just 0.035 est.WP48. The only team that does a better limiting production at that position than the Bulls are the Orlando Magic which happens to employ the Defensive Player of the Year at that position. Joel Anthony has been the most productive Heat player outside of the Three Kings in the playoffs with an estimated 0.6 wins produced and 0.093 est.WP48. This will be a tough matchup for him.
  • All of the Dead Basketball Poets Society will be focused on the PG matchup and the numbers say it will favor the Bulls and the new MVP. The Bulls rank third in Opponent WS48 and the Heat rank fourth, so the defense will be there. Everyone expects Rose’s offense to be there, too, but Chalmers’ offense has shown up more often than not against Rose since college. It will be interesting to see how he performs on the biggest stage for their matchups yet.

When the production across all positions is summed up, the Heat have a seven percent win margin over the Bulls. That means the most likely scenario is that the Heat finish their dinner in six games.

I'm not trying to give you no advice... I don't claim to be no philosopher...

MO' fire, MO' White Hot attire
Less money, MO' titles now that Three Kings are hired
MO' Finals for the Heat Empire, y'all won't serve us 
Y'ALL nervous, know Bosh's jumper's full service, ready to fire 
One team, two teams, three teams, four 
Mickey's sittin on paper, Pat's above the law 
I'm shit listin' haters, I ain't fuckin with y'all 
For the awards voters, "How about Riles, Wade & LeBron?" 
And it just gets worser, every time Pat signs his name in cursive 
Just add another title to these verses 
One million, two million, three million, four 
The salary cap's worthless, they're pissin you off on purpose 
Wade's kids are situated, LeBron's mama is straight 
So they're ready for whatever drama should come their way 
And Joakim's just yappin to me, so his drama is fake
The [Bulls] are noodles, the [Heat] got more ziti to bake...

* The expected WS48 was calculated by multiplying the team’s WS48 at the position by the WS48 the opponent allowed at the position and that product was divided by the league average WS48 at the position. 

The Heat's estimated wins produced stats for the entire playoffs are posted on the Heat Produced Page. Opponent WS48 for each team is available in this spreadsheet.

Unless referenced otherwise, original game data used for this post was taken from and


  1. All of this sounds very good. Just wondering, if you used this in the other series, how well would it have done?

  2. In my defence Mosi, I think that Miami is the better team, and I have a feeling that they will win the series. Also, I like the Heat a lot more than the Bulls. But given everyone else's picks in the smackdown (and the bonus points I get for sticking with the Bulls) I decided to leave my picks unchanged.

    Of course, with that out in the open, I can say that I predicted the winner regardless of which team actually wins. It's a genius strategy!

  3. After seeing tonight's game, what I thought initially is what's the case. If Haslem or Miller can get minutes and be effective, Miami's going to the finals. Otherwise, it's over. You can say it's one game and what not, but it's clear as day. Joakim Noah will dominate in this series, and Miami's only hope is playing Miller or Haslem, hoping for the best or do what Atlanta did against Orlando and get blown out in their losses but win the close games. Eric Spoelstra's not gonna play Mike Miller.

  4. @DDignam - brilliant strategy, Devin.

    @OmegaEntiSupreme - I only ran the numbers by position matchups for the Heat series. Didn't have time to do it for the others. I also just used it for the Western Conference Finals, so we'll see what happens. As for Heat-Bulls, I don't think Miller or Haslem will make much difference. Maybe Miller grabs some of the loose balls that Noah grabbed in Gm1 but he can't shoot at this point and Haslem's not 100%. As for Joakim Noah, please. He shot 4-14. Did he grab a lot of rebounds? Yes. Is he suddenly Kevin Love? No. His WP48 for that game was just 0.166, far from superhuman. This series was always going to come down to rebounding. If the Heat big men keep the Bulls big men off the boards long enough for LeBron & Wade to clean up the glass then the Heat will win. It's that simple.