Archive for the 'Analysis' Category

During Chris Bosh's brickfest last night, all I could think of was, "Wow, this is a John Starks-ian performance." Turns out it was even worse, albeit in a much less critical situation than Game 7 of the Finals.

Using offensive statistical plus/minus (OSPM), I put together a list of the most detrimental offensive games in our box score database (this spans 1987-2011 for the regular season, and 1991-2010 for the playoffs). For every game, I calculated the player's OSPM, the team's offensive rating, and what the team's offensive rating would have been had the player turned in a league-average performance. The most detrimental performances were the ones that sucked the most points from a team's offensive rating. I also added one requirement to qualify for the list: the player's offense must have cost his team a win -- i.e., with an average offensive performance from a player in his minutes, they would have outscored the opponent, but instead lost the game.

Let's use Bosh as an example. Last night, Bosh had an OSPM of -15.18, which means for every 100 possessions he was on the floor, he drained more than 15 points away from Miami's offensive rating relative to a league-average performance. Miami's actual offensive rating was 95.3, but if Bosh had just been average, Miami's rating would have been 108.5 -- meaning he cost them 13.16 points of offensive rating over the course of the entire game. Worse yet, Chicago's offensive rating was 99.6, so if Bosh had been average (or even merely below-average), Miami would have won the game. That's why Bosh qualifies for the list, because his poor offense cost his team a win.

Anyway, here are the most detrimental offensive performances in our database (mouse over column headers for descriptions):

Read the rest of this entry »

With noted shot-creator Carmelo Anthony on the move, I was wondering which players have a hand in "creating" the highest percentage of their team's shot attempts when on the floor.

If we're just looking at a player's own shooting attempts, this is pretty easy. You can look at the percentage of team FGA a player takes when on the court (all leaders minimum 1,366 minutes):

You could also take it a step further and factor in free throws as well, calculating the percentage of each team's True Shooting Attempts (FGA + .44 * FTA) each player takes while on the floor:

But I think you need to take assists into account as well.

Read the rest of this entry »

How in the cosmos did I miss this the first time around?

Two weeks ago, while I was busy with Super Bowl/Hall of Fame work at PFR, SBNation's Tom Ziller posted a story -- and an awesome graphic -- about the biggest "black holes" (players who never pass) in the NBA.

Read the rest of this entry »

Despite their high overall marks, apparently neither the Lakers nor the Heat can beat the league's other so-called "elite" teams. Miami is just 6-9 this season against teams in the top 10 in W-L%, and 0-6 against top-5 teams. The Lakers are barely better, going 6-7 vs. top-10 squads and 2-6 against the top 5. Here's a summary of the other teams in the top 10 by either W-L% or point differential:

Read the rest of this entry »

Alex Sonty, who writes ChicagoNow's Load O' Bull blog, has been paying close attention to Henry Abbott's TrueHoop posts about Derrick Rose -- specifically, this post about past MVP winners and their teams' rankings in wins. Henry found that team wins were highly correlated with MVP voting, to the point that 19 of the last 20 MVPs came from a team with a top-3 record. This of course is bad news for Rose, as the Bulls are 6th in winning % at the moment.

But Alex was wondering how past winners stack up in a schedule-adjusted margin-of-victory based metric like the Simple Rating System, where the Bulls are 5th. So here's the master list -- every MVP winner, with their team's rank in both WPct and SRS:

Read the rest of this entry »

Ken Pomeroy (of the outstanding college hoops stat site Kenpom.com) ran an interesting simulation last month with regard to the randomness inherent in single-game plus-minus scores:

A treatise on plus/minus - the kenpom.com blog

According to Ken's simulation, a player with precisely average "true +/- skill" can show up with wildly variant observed +/- values over the course of a game, or even 20 games.

Just for fun, I re-ran this experiment for ten thousand games, tracking the observed +/- impact of the player through various checkpoints. Here were the results:

Even after 10,000 games, a massive sample that would never be possible to achieve in real life, our perfectly average "player" appears to be a half-point per 40 min worse than average by raw on/off-court plus minus. As Ken says, "respect randomness"!

Just as we did last season, let's take a look at which players would have made the All-Star teams if various advanced stats were the only criteria in the voting. To pick teams, I used the official positional designations from the 2011 ballot; each team must have 4 guards, 4 forwards, and 2 centers, with room for 2 wild cards from any position to fill out the roster. Players in bold are starters; "*" designates the player as a member of the real-life All-Star team.

Read the rest of this entry »

Just a brief rant about the media reaction to yesterday's Celtics win over LA...

After the game, especially on the SportsCenter coverage last night, I saw "Too Much Kobe" being held up as an explanation for the Lakers' struggles. Here's a sample:

"Bryant took 29 of the Lakers' 66 field goal attempts (43.9 percent) while he was on the floor. This was the 10th game this season that Bryant took more than 40 percent of the Lakers' shots while on the court. In those 10 games, the Lakers are 3-7. Los Angeles is much better when Bryant shoots a smaller percent of the team's shots while on the court. The Lakers are 23-5 when Bryant takes less than 35 percent of the team's shots when on the floor."

That's a familiar media theme when Kobe scores a ton of points but his team loses; we saw it a lot in 2006, for instance.

As far as I can tell, "Too Much Kobe" is exclusively an offensive criticism. Trouble is, L.A.'s offense was fine yesterday. Against the 3rd-best defense in the league, against whom an average team would expect to score about 104 pts/100 poss. at home, the Lakers scored 110.1. The offense is not why L.A. lost, and therefore "Too Much Kobe" can't be why they lost.

They lost because they allowed the 11th-best offense in the NBA to score a staggering 125.0 points per 100 possessions against them on the road. This may or may not be Kobe's fault -- aside from personal fouls, he wasn't overly active on D, and despite his scoring feats the Lakers were -9 when he was on the court.

But it can't possibly be because Kobe had zero assists.

I finally got around to calculating the standard errors for our team Simple Ratings today:

Then I set up a little Monte Carlo sim to estimate what is the probability of each team being the NBA's best (aka the team with the greatest "true" SRS skill). After 10,000 simulations using the estimates and standard errors above, here were the results:

Read the rest of this entry »

What was the best run ever for your favorite team? What was the worst stretch of seasons? Let's take a look at the raw numbers in terms of NBA winning percentage over an x-year span (including our regressed 2011 WPcts):

Best & Worst 2 Years

Read the rest of this entry »