PDA

View Full Version : The Value of a Blocked Shot in the NBA: From Dwight Howard to Tim Duncan



6deep
03-11-2010, 05:51 PM
I am here at MIT Sloan Sports Analytics Conference, and looking through the number of great papers one of them caught my eye. It was called “The Value of a Blocked Shot in the NBA: From Dwight Howard to Tim Duncan” written by John Huizinga, (A professor of business at the University of Chicago) who has been Yao Ming’s NBPA registered agent since Yao’s inaugural season, representing Yao in his dealings with the Houston Rockets, as well as with major endorsement partners.

Data

Before getting into the data and what it tells us, it was important for us to learn how the data was gained. Mr. Huizinga developed a database called Chances. The database uses data provided by STATS, LLC. and allowed everyone to know the context of the action before the block. The sample of the data used in the presentation was players with over 100 blocks over the last 7 years (this is when the data has become available). In total, this ended up being 170 player-seasons.

Type Of Block

Is blocking a lay-up more valuable than blocking a jump-shot? Mr. Huizinga’s data says yes. In his presentation, he said that it all comes down to expected value. A jumper has an expected point value of 1.04 while a lay-up has an expected point value of 1.54. Looking at it this way, Brendon Haywood, who many people is a very good defender (me included) actually is a less valuable shot blocker than Jermaine O’Neal.

Haywood gets 69% of his blocks on jumpers, meaning he only blocks 31% of the more valuable lay-ups. On the other end of the spectrum, 91% of Jermaine O’Neal’s blocks were on lay-up attempts, while only 9% of his blocks were the less-valuable jump shots.

“Russells”

Many people who have seen Bill Russell play (or have seen highlights) know that Bill Russell was remembered for blocking shots for his teammates, starting a fast break (called by Bill Simmons as “Russells”. Mr. Huizinga showed that this doesn’t really happen in the NBA anymore. There have only been 7 players (in the 7 season where the data was tracked) who accumulated more than 20 “Russells” in a season.

Preblock Situation

One of the most important things to take away from Mr. Huizinga’s presentation is expected value of the preblock situation. Or in otherwords, what happened right before the block took place. Naturally, a block coming off of a live-turnover situation on a lay-up (think a LeBron chasedown) is going to be more valuable than a block coming off of a deadball situation. Again, this comes down to expected point value. The expected point value of a live-turnover situation is higher than a deadball situation because coming off of a live-ball turnover, the defense doesn’t have a chance to get back.

The best shot blocker in the NBA when it comes down to this situation ends up being Andrei Kirilenko, as 16% of his blocks come against this shot type. The worst ends up being Greg Ostertag. This makes sense considering that Ostertag isn’t really known for his footspeed.

Putting It All Together

So whose blocked shots are the most valuable? Mr. Huizinga closed the presentation by going over what he calls “Block Value.” To determine block value, he used the formula Points Saved + Points Created where Points Saved equals the effect of a Block on Opponents Expected Points during this possession and Points Created equals the effect of a Block on Own Team’s Expected Points During the next possession.

Using this formula, we found out who had the best season since the data started being collected (2002-03) in terms of overall block value. It ended up being Theo Ratliff during his 2003 season. Ratliff accumulated a block value of 300 (287 coming from points prevented while 13 came from points created), which when transformed into wins ends up being right around 5.

Interesting Numbers

Just thought it would be interesting to include some numbers towards the end of Mr. Huizinga’s presentation, showing how number of blocks can’t really be used when determining who is the best “shot blocker.”

2003 season

Stromile Swift | 119 blocks with a block value of 74
Rasho Nesterovic | 117 blocks with a block value of 124
So why was Mr. Huizinga’s paper called From “…Dwight Howard to Tim Duncan?” Well as he explained, through a series of charts, Tim Duncan has had the best season in history when it came down to value/block with 1.12, meaning he saved 1.12 points with every block and Dwight Howard ended up with the worst season in terms of value/block with with .53 (both came during the 2008 season).

http://nbaplaybook.com/2010/03/06/the-value-of-a-blocked-shot/

Jays Claw
03-11-2010, 05:54 PM
Cool. :cool:

JermanJaysFan
03-11-2010, 05:58 PM
Cool. :cool:

x2

Paulliwali
03-11-2010, 05:59 PM
mhm good read

wolf3742
03-11-2010, 06:03 PM
isn't this the same as the Peter Keating article?

TheKing23
03-11-2010, 06:08 PM
Really interesting read.

ldc62
03-11-2010, 07:42 PM
isn't this the same as the Peter Keating article?

But with statistics.


A very interesting metric stat.

Raph12
03-11-2010, 07:44 PM
Good read.

HoopsDrive
03-11-2010, 08:06 PM
Cool. :cool:

x3

IversonIsKrazy
03-11-2010, 08:26 PM
very interesting and can agree on how Dwights block are showboating, while Duncan's are a lot more important

giventofly
03-11-2010, 08:39 PM
Kind of seems like basketball statisticians are way behind...

brisk002
03-11-2010, 08:42 PM
so I guess this means Bogut is better than Dwight Howard.... LOL

ManRam
03-11-2010, 09:14 PM
A lot of cool stuff coming out of this MIT conference. This is no exception.