Is there work showing that pace increases when subs come on? Seems like that would follow from an increase in shots and steals.Mike G wrote:The conclusion is that it must be easier to get shots, rebounds, blocks, steals against subs.
Assists and shooting% seem to be better with more starters on the floor. And fewer fouls, at least for bigs.
2012-13 starter-sub comparisons
Re: 2012-13 starter-sub comparisons
Re: 2012-13 starter-sub comparisons
That's one possibility. It does not preclude the scenario that there are just more missed shots, hence more rebounds and fewer assists.
I've selected players who are sometime-starters -- not end of the bench guys. So they move up in the pecking order when other starters are off the floor.
Here's a summary of the whole league this season, separated into 4 equal sized groups: players who have started every game (or all but one); starters in 20% or more of their games; who have started <20% or played at least 340 minutes; those with no starts and <340 min.
Shooting frequencies (per36) and efficiencies:These are totals, *36, divided by total minutes.
Full-time starters play 44% of the minutes. They shoot notably more and better 2-pt FGA, and more FTA.
Group 3 may be overloaded with 'scorers off the bench' -- JR Smith, Crawford, Kevin Martin, Landry, Ray Allen, ..
Other per 36 minute rates:I especially like the foul rate, with its steady increase off the bench.
Last column is a very rough suggestion as to each group's constituency in their respective lineups. Obviously the aforementioned 6th men are mis-identified as 8th+
But by isolating bench warmers in the bottom quartile, it's pretty clear that there is not a general increase in pace -- at least as indicated by FGA and Reb, steals and TO -- when starters are out of the game.
I've selected players who are sometime-starters -- not end of the bench guys. So they move up in the pecking order when other starters are off the floor.
Here's a summary of the whole league this season, separated into 4 equal sized groups: players who have started every game (or all but one); starters in 20% or more of their games; who have started <20% or played at least 340 minutes; those with no starts and <340 min.
Shooting frequencies (per36) and efficiencies:
Code: Select all
% Starts %Min mpg 2FGA 2fg% 3FGA 3fg% FTA FT% TSA TS%
.97 to 1.0 .44 32.6 13.4 .487 2.6 .359 4.0 .758 15.1 .539
.20 to .96 .28 23.3 11.1 .466 3.2 .358 2.7 .750 12.3 .522
.00 to .19 .23 18.2 11.5 .466 3.5 .363 3.0 .764 12.8 .529
none .05 9.2 10.9 .437 2.8 .289 2.7 .669 12.1 .468
Full-time starters play 44% of the minutes. They shoot notably more and better 2-pt FGA, and more FTA.
Group 3 may be overloaded with 'scorers off the bench' -- JR Smith, Crawford, Kevin Martin, Landry, Ray Allen, ..
Other per 36 minute rates:
Code: Select all
%Start OReb DReb Ast Stl TO Blk PF Pts depth
.97 to 1.0 1.73 4.99 3.80 1.22 2.29 .84 2.5 16.3 1 to 4
.20 to .96 1.63 4.33 2.71 1.09 1.87 .73 3.0 12.9 5 to 7
.00 to .19 1.70 4.27 2.82 1.17 1.95 .77 3.5 13.6 8 to 11
.00 1.55 4.14 2.72 1.13 2.15 .80 4.0 11.4 12+
Last column is a very rough suggestion as to each group's constituency in their respective lineups. Obviously the aforementioned 6th men are mis-identified as 8th+
But by isolating bench warmers in the bottom quartile, it's pretty clear that there is not a general increase in pace -- at least as indicated by FGA and Reb, steals and TO -- when starters are out of the game.
Re: 2012-13 starter-sub comparisons
Last year, there were 58 players who started between 20 and 50 games. Of those who were with a single team, these 34 had at least 1340 minutes: Oladipo, JR Smith, Singler, G Green, R Jackson, J Jack, H Barnes, Kanter, Ja Crawford, Waiters, D Collison, Lin, Sullinger, Tyreke, Belinelli, Diaw, Harkless, Anteto, Henson, Tobias Harris, Varejao, A Anderson, Mozgov, H Thompson, M Barnes, J Dudley, B Roberts, Caldwell-Pope, Nene, Booker, Stoudemire, Humphries, Koufos, Mayo.
http://bkref.com/tiny/7AEQs
I got individual player splits from such pages -- http://www.basketball-reference.com/pla ... lits/2014/
Summing all their raw stats in their games as starters and as reserves, their aggregate per-36-minute rates are generally less when starting. Ranked by ratio:Presumably, starters shoot and pass better, so a sometimes-starter can be expected to get better shots (TS%) and more assists.
Turnover and foul rates are better (lower) when starting.
x = Pts+Reb+Ast+Stl+Blk-TO (per 36)
http://bkref.com/tiny/7AEQs
I got individual player splits from such pages -- http://www.basketball-reference.com/pla ... lits/2014/
Summing all their raw stats in their games as starters and as reserves, their aggregate per-36-minute rates are generally less when starting. Ranked by ratio:
Code: Select all
sta/sub per36 start sub
1.39 mpg 30.6 22.1
1.05 Stl 1.09 1.04
1.02 Ast 2.95 2.90
1.01 TS% .541 .536
.981 OReb 1.55 1.58
.975 FGA 12.2 12.5
.973 Pts 14.7 15.1
.964 TSA 13.6 14.1
.948 TO 2.00 2.11
.939 Blk .62 .66
.921 DReb 4.15 4.50
.906 PF 2.92 3.22
.900 fta/fga .253 .281
.877 FTA 3.09 3.52
.974 x 23.0 23.7
Turnover and foul rates are better (lower) when starting.
x = Pts+Reb+Ast+Stl+Blk-TO (per 36)