What analytics/databall concepts have most changed the NBA?

[fpliii]

Kind of a wide open question, but what comes to mind (on a team-level)?

From a coaching perspective, two things that stand out:

• trending away from midrange jumpers and towards threes (corner threes in particular)
• eschewing of offensive rebounds in favor of getting back on transition defense

What other key innovations have shaped the league in terms of style of play?

What about from a Front Office perspective? How do you feel the team-building paradigm has evolved?

[Mike G]

It's kind of a chicken and egg thing. Did 3FG-heavy offenses rise because so many guys got good at it? Or do teams thrive because it's easier to find good 3FG shooters?

Similarly, did the resultant makeup of players leave fewer offensive rebounders on the floor?

In 2013-14, there was little correlation (-.08) between OReb% and Win%.
There was however a significant negative (-.34) between OReb% and 3FG made.

Houston led the league in 3FG and FT, 2nd to Mia in 2FG%, and they were also 2nd in OReb%. Do they need to change anything?

[wilq]

Kind of a wide open question, but what comes to mind (on a team-level)?

You've mentioned probably the best examples but overall I think we have a problem with lack of transparency - from the outside we simply don't know whenever some changes were made because analytics suggested it or was it simply an evolution or some blend of good coaching/scouting.

For example, FTA/FGA and fouls has slowly declined over the last decade so has it happened because of FTA's consistently high value or was it simply some shift in the defensive philosophy?
Same question with the approach to the shot clock, has it changed because of more data about efficiency early / late or was it simply an effect of copying D'Antoni's success?

What about from a Front Office perspective? How do you feel the team-building paradigm has evolved?

IMO the most influential changes in team-building were created by new versions of the CBA not by analytics.
Speaking of which, I'm curious about something, if almost every team feels "we have a good PG" shouldn't it decrease their value and contracts across the league? Or more specifically, how many point guards in the league would you give a max extension?

It's kind of a chicken and egg thing. Did 3FG-heavy offenses rise because so many guys got good at it? Or do teams thrive because it's easier to find good 3FG shooters?

I disagree it's a chicken and egg thing, for me the answers are provided by the distribution of shots, good 3-pt shooters a decade ago shot way less than they do know. The same is true even for average 3-pt shooters. IMO it means long-ball was simply underrated as a weapon.
BTW, I still cringe when I look at Reggie Miller's bref page - he was among the best shooters of all-time and majority of his seasons wouldn't even crack Top20 in attempts per game last year.

[Mike G]

Last season, 20 players shot over .400 from the arc and made the Leaders board.
Also, 20 guys had over 400 attempts.

Reggie Miller only had a couple of seasons with both of those qualifications (not counting short-arc years, 1995-97). His best 3fg% -- .429 -- would have been 6th best this year. While he was always one of the best in his time, he'd be just one of many today.

If anything, he likely had more open looks before the shot was fully exploited. These days, all teams protect against it, so it's ever harder to get the shot off.

[kjhogo]

I think analytics has completely changed how teams have built front courts. There's now the quest for a stretch guy/rim protector balance and that seemed to have come from the push for 3 point shooting and the ability to quantify defense and defensive presence around the rim.

I think it would be easier for most guys to give up on a player like a Bismack Biyombo but if you look at what he can do defensively a guy like him gets a chance that he may not have gotten before the "analytics" movement.

[Crow]

There are variables reviewed here viewtopic.php?f=2&t=8662 to consider as change impacting or candidates to be such.

[fpliii]

Thanks for the responses so far.

For example, FTA/FGA and fouls has slowly declined over the last decade so has it happened because of FTA's consistently high value or was it simply some shift in the defensive philosophy?
Same question with the approach to the shot clock, has it changed because of more data about efficiency early / late or was it simply an effect of copying D'Antoni's success?

I think analytics has completely changed how teams have built front courts. There's now the quest for a stretch guy/rim protector balance and that seemed to have come from the push for 3 point shooting and the ability to quantify defense and defensive presence around the rim.

Great points.

IMO the most influential changes in team-building were created by new versions of the CBA not by analytics.
Speaking of which, I'm curious about something, if almost every team feels "we have a good PG" shouldn't it decrease their value and contracts across the league? Or more specifically, how many point guards in the league would you give a max extension?

There are variables reviewed here viewtopic.php?f=2&t=8662 to consider as change impacting or candidates to be such.

wilq - Would be interested on what others have to say about your PG question.
Crow - Thanks. I really enjoyed this follow-up as well: http://tothemean.com/2014/07/06/predict ... e-nba.html

[schtevie]

Just to put it my two cents (to add to the non-trivial pile of two cent contributions on the same topic over the years) the history of the adoption of the three point shot simply does not support the notion that it as a analytics/databall concept (i.e. the ability to multiply by 1.5) is what brought about the change. If so, the rate of adoption would not have been so mind-bendingly slooooooooow.

One needs a theory of "technological" change to organize one's thoughts on these issues, and what the three point shot history reveals is that improving one's offensive efficiency and winning championships has not been the sole (or primary) priority of NBA franchises.

That it was Phoenix last year that first garnered a CotY award, largely for aggressively adopting this tactic/strategy, more than 30 years after the three point shot came to the Association, is illustrative of this fact.

And a similar (but distinct and rather interesting) story can be told regarding the adoption of "stretch4iness" (as that is largely a three point shooting-related concept).

This is not to say that all "analytics/databall" concepts have been doomed to slow adoption, but given this, the notion of "change" needs to be better specified.

[Crow]

Lineup analysis and strategy seems way less developed and implemented by teams than it could / should be imo.

I am hoping we hear something new from the external analytics community on this front soon...