Threes and Transition Defense – A Follow Up

May 13, 2015; Atlanta, GA, USA; Washington Wizards guard Bradley Beal (3) attempts a shot against Atlanta Hawks guard Kent Bazemore (24) in the third quarter of game five of the second round of the NBA Playoffs at Philips Arena. The Hawks won 82-81. Mandatory Credit: Jason Getz-USA TODAY Sports
May 13, 2015; Atlanta, GA, USA; Washington Wizards guard Bradley Beal (3) attempts a shot against Atlanta Hawks guard Kent Bazemore (24) in the third quarter of game five of the second round of the NBA Playoffs at Philips Arena. The Hawks won 82-81. Mandatory Credit: Jason Getz-USA TODAY Sports /
facebooktwitterreddit
May 13, 2015; Atlanta, GA, USA; Washington Wizards guard Bradley Beal (3) attempts a shot against Atlanta Hawks guard Kent Bazemore (24) in the third quarter of game five of the second round of the NBA Playoffs at Philips Arena. The Hawks won 82-81. Mandatory Credit: Jason Getz-USA TODAY Sports
May 13, 2015; Atlanta, GA, USA; Washington Wizards guard Bradley Beal (3) attempts a shot against Atlanta Hawks guard Kent Bazemore (24) in the third quarter of game five of the second round of the NBA Playoffs at Philips Arena. The Hawks won 82-81. Mandatory Credit: Jason Getz-USA TODAY Sports /

Yesterday at Fancy Stats, I looked into that old saw that too many threes lead to long rebounds, which in turn lead to fast breaks for the opposition. While the first part is true, long shots do lead to long rebounds, the other part? Not so much:

RebTrans
RebTrans /

"[N]ot only do missed threes, especially the above-the-break threes…lead to the proportionally fewest transition attempts, those attempts which do occur are not especially damaging. Unlike those quick ripostes which tend to follow missed layups. While some factors serve to balance this gap out such as the higher rate of offensive rebounding on closer shots, not to mention the generally far higher shooting percentages, shooting threes certainly does not appear to spur great numbers of run outs going the other way."

Which is basically my circumlocutious way of declaring that particular piece of conventional wisdom to be bunk.  There were, however, a few interesting follow-up questions which I thought were worth exploring briefly. The first inquiry, and I’m paraphrasing, was can any effect of defensive discombobulation be seen later in the shot clock depending on the initial shot location? For example are threes perhaps more likely to lead to cross-matches and all the potential ensuing drama?

The short answer is no such effect was discernable. I’d drop a graph here, but it would look more like a multi-colored Twizzler or the innards of a bomb[1. Red wire? No Green! Blue!] than a chart from which analysis could be drawn. Basically, the eFG% on the ensuing possession seems to be consistent across the shot clock regardless of where the previous shot was taken, and the damaging effect of missed shots at the rim was more or less the result of the greater proportion of early opportunities arising from those missed layups:

In case the legend is tough to read, the green line represents shots following misses in the restricted area.

The other main question was regarding the identity of the shooter. Given that the examples I used in the original piece were of wings either shooting from deep or missing at the rim, how does changing the shooter affect transition defense? Well, consistent with my theory that the main culprit, guards, particularly point guards[1. Positions defined by basketball-reference.com parsing of play-by-play data, which I don’t think is perfectly accurate, but will do in a pinch.] getting missing shots at the rim and in the paint in general appears to have a negative effect on transition defense:

RebTransPos
RebTransPos /

The one countervailing factor perhaps being big men shooting above the break threes allow for decent floor balance, but allow a higher percentage on those fast breaks which do occur. Entering into pure speculation, perhaps when they are beaten back after shooting from deep, they are beaten back badly?

As a final little aside, I’m always interested in the value and effect of a blocked shot, so I decided to look at if shot blocking had any discernable impact or correlation with transition play:

BlockTrans
BlockTrans /

Sure seems like blocking shots in the paint can help generate some always-welcome fast break points.