A Team Greater Than The Sum Of Its Parts

[Crow]

What if we looked at the 3PTA/FGA of the most efficient teams offensively rather than playoff teams?

Good point. That brings it back up a bit to 10 of 16 playoff teams in the top 16 in the regular season on 3PTA/FGA. But that is not real high.

[Nima]

What if we looked at the 3PTA/FGA of the most efficient teams offensively rather than playoff teams?

Good point. That brings it back up a bit to 10 of 16 playoff teams in the top 16 in the regular season on 3PTA/FGA. But that is not real high.

Thanks for looking into that.

Adding to that point, teams that are better defensively generally get into transition more, and generally, teams probably take a lower rate of threes in transition than they do in half-court. I would suspect if we looked solely at the most efficient half-court teams, the number of teams would increase.

[Crow]

Maybe. I was making a quick and blunt assessment. I am all for greater precision if you have the time to do it.

The bottomline will remain that some good playoff teams are out of step or not really leaders on this fundamental shot distribution issue.

[Crow]

Your data supports a high usage initiator more than a high assist concentration in that guy. That would be good news for Westbrook and the Thunder if they had other guys well capable of and making passes for assists.

This much of the analysis can be used to support Kobe being Kobe, given his assist capable teammates. It looks like 2 pt shots are fairly concentrated for the Lakers but maybe too high for Kobe and too low for Gasol and Bynum. At the end of the shot clock it is hard to get it inside to big men but if you don't give it to them fairly early, they and Kobe basically have made the choice for him to end up taking the biggest number of those shots.

[Nima]

Maybe. I was making a quick and blunt assessment. I am all for greater precision if you have the time to do it.

The bottomline will remain that some good playoff teams are out of step or not really leaders on this fundamental shot distribution issue.

I agree. It would be odd if every team had the same formula.

[Nima]

Your data supports a high usage initiator more than a high assist concentration in that guy. That would be good news for Westbrook and the Thunder if they had other guys well capable of and making passes for assists.

This much of the analysis can be used to support Kobe being Kobe, given his assist capable teammates. It looks like 2 pt shots are fairly concentrated for the Lakers but maybe too high for Kobe and too low for Gasol and Bynum. At the end of the shot clock it is hard to get it inside to big men but if you don't give it to them fairly early, they and Kobe basically have made the choice for him to end up taking the biggest number of those shots.

I agree with this also. It's definitely more about the player initiating the offense and drawing attention rather than getting the assist necessarily. The shot doesn't need to come on the first past, and often shouldn't. I agree with your assessment of Kobe as well, as he can be such an initiator as he is a great threat to score and attracts defensive attention. He need only pass up some of those early shot clock mid-range twos, particurly when he's double and triple teamed; but late in shot clock when he can't get to the rim, that's when I feel it is most appropriate for him to take those tough mid-range twos off the dribble.

I think one of the big issues is that players think too much. Most perimeter NBA players are capable of catching and attacking the rim when space has been created for them by someone initiating the offense. The problem is when they think about what to do rather than just reacting.

John Wooden taught his players not to think, and to just react, and obviously John Wooden knew what he was talking about.

[Crow]

Celtics get no support for their strategy by expanding the view to OR%. Dead last in the league on that. Probably hurt a bit by the higher reliance on mid-range shots vs 3s. Broaden all the way out to offensive rating and they were 27th on that in the regular season. Used their own formula and really sucked on offense in the regular season and were below average on that in the playoffs too (10th of 16).

Was having / making the 2nd leading assist-maker in the league a good idea? Not really by that part of your data. Especially an initiator with a relatively low level of shots and almost no ability to hit from 3. But yeah stay the course with him...


Houston trades or lets go of their top 4 offensive rebounders and spends big to bring in an offensive rebound leader. That seems to go against your data analysis. But maybe they end up sharing the load more equitably overall than before (despite Asik's dominance)? Will it be at a higher or lower overall team OR% level than last season? I'll guess lower.

Chicago 9th on at the rim frequency and high on FTA but only 23rd on at the rim FG%, 27th of FT% and 17th on 3pt attempts. Plenty to refine there. With stat analysts (and fancy approaches) or not.

Terry Stotts is an offensive genius... and then the Mavs fall to 23rd on offensive efficiency in this regular season and are below the efficiency average in the playoffs? Not helped by being bottom 20% at OR% and getting to the line.

[Nima]

I think Rondo might be an exception to the rule. First, he's not a significant threat to score outside of his ability to get to the rim (which is something he is rather good at, despite the fact that defenders slack off of him because of his weak, though improving, jump shot), yet he's one of the best at finding the player with the open shot. Maybe because Rondo is so good at finding the guy with the best shot and making the pass effectively so that they can take the shot, that the extra pass is not needed as often. That said, Boston was not one of the most efficient teams, so I also think that supports the findings. However, Rondo does bring other things to the table, with his rebounding ability, nose for the ball, and though I think he probably underachieves defensively, the potential to be a great defender.

Regarding offensive rebounding, we can't just blindly use the findings, meaning we have to think about them logically. For example, if two lineups have the same 4 players (let's say athletic players who are solid offensive rebounders for their position), but different 5th players, with one being a far better offensive rebounder (like Asik, who you mentioned) than the other, although the lineup without Asik will have a lower standard deviation and more even distribution of OR%, the lineup with Asik will very likely be the better offensive rebounding team. However, that's not the case here. As you noted, Houston let go of their top 4 offensive rebounders (assuming they played in the same lineups at times, rather than always in different lineups, in which case they are simply upgrading the offensive rebounding of a single spot in the lineup).

Something to keep in mind is that the analysis is done independent of the sums, so it really looks at distribution alone when all things are equal. So, if adding Asik means losing an equal amount of offensive rebounding at other spots in the lineup, I don't think it would be beneficial. So, I would guess lower as well, unless Asik is such a dominant offensive rebounder that it more than makes up for the loss at other positions, or if players step up their offensive rebounding to account for the lost players.

[Mike G]

Though Miami was just two points better than average on the season, how were they during the finals? From observation alone, it seemed like they turned things around once their backs were against the wall against Boston. They were particularly disciplined during the finals and I believe their shot selection improved.

Battier hit 15-26 from the arc, Miller 7-11. Combined 22-37 in total 232 minutes.
They played 19% of the team's minutes, had 38% of their 3FGA and 52% of their 3FG.
On all FG, they were 28-47; thus 6-10 on 2FGA.

During the season, Miami averaged 15.4 3FGA and 5.5 makes.
During the Finals, it was 16.3 and 7.0
Their attempts didn't change much, but their success rate did. They got about 4.25 PPG more than they normally would from those attempts.

If they were evenly matched going into the Finals, a 4.25 ppg boost makes them likely to win 65% of their games.
In fact, they won by 4.0 ppg average, taking 4 of 5 games.
In game 2, Battier was 5-7 from the arc, providing about 7 points more than his avg from 7 shots; Mia by 4.
In game 3, he's 2-2, and they win by 6. A 38% shooter is more likely to go 0-2 than 2-2.
Even the 15-point blowout in game 5 is due almost entirely to 3-pt success: Miller and Battier combine for 10-15: about 12 points more than normal from 15 attempts.

[wilq]

I interpreted the findings of my research analyzing the distribution of various statistical categories to draw some conclusions about how to best build a team that will be greater than the sum of its parts.

Excuse me but I have to... how your concept will help to create a team GREATER than the sum of its parts?
You have to assume underrating of 3-pt shooters around the league, right?

... three-point attempts should be taken more liberally, while mid-range shots should be taken more conservatively.

If you've been around these parts for very long, you'd know that LOTS of people have come to this conclusion. It would only be a slight exaggeration to say Everyone has held this theory.

True but does it indicate there's optimal distrubition of shots and teams can't lower long two attempts any more? Because it's surprisingly constant...

[Mike G]

Last year (2010-11), the Cavs gave up 41% shooting from the arc, an all-time worst. And this is with several years of knowing that the 3 is one of the most effective weapons out there.

That 41% is from 509/1489 opponent shooting. There are only a handful of 3-point aces who have hit 41% on their first 1489 attempts. But the whole league can now do it against an NBA team that is a bit lax at defending it.

This year's Celtics allowed just 31% from the arc, the lowest opponent % in 8 years (covering the peak of league 3-pt effectiveness). Obviously, defensive concentration can seriously cut into the percentages.

The long 2-ptr is the default available shot, when there's no fast break. It's probably the best complement to the 3, because it can be a derivative play, when the shooter is chased from the arc. It may be that successful teams will tend to be those that hit 45%, rather than 40%, from 15-20'.

[huevonkiller]

The Heat seemed to underrate the three-point line until the playoffs started. The Heat's best line-ups had two wing players who could stretch the floor alongside their Top 3 players.

[Nima]

I interpreted the findings of my research analyzing the distribution of various statistical categories to draw some conclusions about how to best build a team that will be greater than the sum of its parts.

Excuse me but I have to... how your concept will help to create a team GREATER than the sum of its parts?
You have to assume underrating of 3-pt shooters around the league, right?

I think the idea as that when teams have multiple guys who initiate offense, whether it is creating for themselves or others, often through isolation, they aren't maximizing what they get out of each. I believe, for example, Amar'e Stoudemire likes the ball in his hands, and likes to create for himself, as does Carmelo, but they both can't have the ball at the same time. So when Carmelo is initiating the offense, anyone else's ability to initiate is going to waste and it is best to have guys who can finish, whether from 3, or at the rim, and athletes who can rebound and play D, as opposed to more players who want the ball in their hands.

three-point attempts should be taken more liberally, while mid-range shots should be taken more conservatively.

If you've been around these parts for very long, you'd know that LOTS of people have come to this conclusion. It would only be a slight exaggeration to say Everyone has held this theory.

True but does it indicate there's optimal distrubition of shots and teams can't lower long two attempts any more? Because it's surprisingly constant...

This is a good question. I don't know. From observation alone, I think a lot of long two-point attempts can be passed up in favor of passing (or attacking/driving and passing), with the goal of getting the ball closer to the rim, or getting a three-point attempt.

But, I don't have a definitive answer to that question from the analysis itself.

Again, the findings of the research were with regard to the relationship between offensive efficiency and the standard deviations of the percentage of various statistics each of the five player in a lineup accounted for. The most recent post is simply my interpretation of what those findings mean, based on my own thoughts of the game and some logic.

[Nima]

The Heat seemed to underrate the three-point line until the playoffs started. The Heat's best line-ups had two wing players who could stretch the floor alongside their Top 3 players.

And one of their top 3 players, Bosh, also stretched the floor, whether it was with the corner 3, or as the main player who took mid-range jump shots off the catch.

[Mike G]

... From observation alone, I think a lot of long two-point attempts can be passed up in favor of passing (or attacking/driving and passing), with the goal of getting the ball closer to the rim, or getting a three-point attempt. ..

Well, since it's in the best interest of all players and teams to get the better shot, why don't they just do that, if it's an option?

Occasionally, a player will take an "ill advised" shot. This is generally when the shot does not go in.
Over the course of a season, or even a single game, there's a big difference between 45% effectiveness and 50%. On a given play, after the fact, you can say "That was a good unselfish pass to a player in a better position".

But on a more immediate scale, a 45% shot is WAY better than a turnover -- shot clock violation, offensive foul, bad pass. So it's pretty hard to say that when a player has an open look from 20' he should never take the shot.

When I wrote earlier that a player in 3-point position is rather "out of the offense", I was referring to his physical distance from the "action" near the rim. He can't really initiate much offense from out there. He can't really get a pass and do anything other than shoot, or bail out of his shot.

When a team has great initiators and/or a great post player, they can use a couple of 3-pt specialists. If they don't have these rare assets, removing some players from the "action zone" probably doesn't help their offense.

It's almost impossible, for a normal person, to get into the offensive action from 10-17' out, and then to fly back to the arc, position and fire an accurate 3. The world's best can do it when they're playing near their best.