one thing to note:  the "enough" team has a cumulative 95 usage in the starting lineup.  The "more" team has 111.  I've never really paid attention to cumulative usage.

I wonder what differences there would be if I took the "more" team and dropped Shaq (31.9 usage) for someone around 20-22 usage.

Posted by thisredbox on 7/3/2013 8:30:00 PM (view original):

I just submitted a team with all players between 22.7 and 27.9 usage (besides three junkers). Kinda high on the usage side, but pretty consistent. Everybody at four usage points except Timmy at 6. I'm predicting 81-1 (jk).

Name	Pos	Usage%	2pt%#	3pt%#	eFG%	OReb%	DReb%	Ast%	Stl%	Blk%
12-13 Tim Duncan	PF	27.9	50.2	28.4	50.3	6.8	27.7	11.0	1.1	5.0
10-11 Manu Ginobili	SG	26.2	49.5	34.7	50.9	2.0	11.1	20.4	2.3	0.7
80-81 Magic Johnson	SG	24.3	53.8	18.9	53.4	8.1	15.4	24.2	3.7	1.0
11-12 Stephen Curry	PG	24.1	51.4	45.7	58.3	2.2	10.7	24.1	2.4	0.6
07-08 Shaquille O'Neal	C	23.9	59.1	0.0	59.3	11.0	23.6	6.9	0.8	3.0
07-08 Chauncey Billups	PG	23.7	47.6	39.8	52.6	1.9	7.8	28.2	2.0	0.4
08-09 Kevin Garnett	PF	23.5	53.2	24.7	53.2	5.4	25.0	10.5	1.7	2.4
12-13 Chris Paul	PG	22.8	53.7	32.6	52.6	2.6	9.9	36.3	3.4	0.3
11-12 Manu Ginobili	SG	22.7	61.6	41.4	61.8	2.5	12.9	23.3	1.4	0.9
87-88 Rick Carlisle	SG	18.7	45.7	35.9	47.8	3.0	3.9	18.5	2.3	1.2
89-90 Leon Wood	SG	12.9	42.6	19.3	36.7	0.5	6.0	29.3	1.3	0.8
90-91 Tree Rollins	C	11.8	42.1	0.0	42.4	6.9	15.4	1.8	0.4	5.8

*Update:

We finished 52-30, second in the division. Took seven games to get out of the first round, lost in five in the second round. We actually won game one of the second round, but we were beholden to exhaustion. At the start of game two:  Paul - 86%, Billups - 91%, Ginobili - 88%, Garnett - 90% and Duncan - 90%.

 Obviously one of the trade-offs with usage is less minutes per dollar and I think that was apparent here. We came in at about 104% of total regular season minutes minutes: ~17429 on 16758.

Starting lineup at 124.1 total usage points.

More stats to come if anyone is interested.

yeah red, do so.  I'll see what your advanced numbers were real quick and link you...


You got crushed on the boards... that was your main problem.  Winning the turnover battle is only part of the possession war.

Posted by ashamael on 9/7/2013 10:44:00 PM (view original):
one thing to note:  the "enough" team has a cumulative 95 usage in the starting lineup.  The "more" team has 111.  I've never really paid attention to cumulative usage.

I wonder what differences there would be if I took the "more" team and dropped Shaq (31.9 usage) for someone around 20-22 usage.

Actually, I'm thinking about keeping Shaq... looking at a SL of 16 usage points & 101.3 cumulative usage in the SL.  Taking the "more" starting 5 and replacing cp3.  I'll likely work on this team this evening and run it as my next OL team.

Team 3 built... 13.9% estimated tov% going in.  Big jump by dropping cp3 and getting a lower-usage PG in there.  We'll see how it all works out, though.

Name	Pos	Usage%	2pt%#	3pt%#	eFG%	OReb%	DReb%	Ast%	Stl%	Blk%
96-97 Michael Jordan	SG	33.2	50.6	37.2	51.6	3.9	12.5	13.6	2.1	0.8
09-10 Jason Kidd	PG	14.6	41.7	42.4	55.4	1.8	14.8	31.8	2.3	0.7
01-02 Shaquille O'Neal	C	31.9	58.2	0.0	57.9	10.4	20.1	10.4	0.8	3.3
92-93 Dennis Rodman	PF	10.7	46.9	20.7	44.4	15.3	35.0	5.0	0.9	1.1
96-97 Dennis Rodman	PF	10.0	46.0	26.1	45.6	17.8	30.4	10.5	0.7	0.6
94-95 Dennis Rodman	PF	10.9	57.2	0.0	57.1	20.1	36.5	7.7	0.9	0.8
06-07 Shaquille O'Neal	C	29.9	58.9	0.0	59.1	10.2	19.6	9.6	0.3	3.0
03-04 Charlie Ward	PG	19.1	46.2	42.8	55.6	1.3	11.2	28.4	2.7	0.6
94-95 Trevor Ruffin	PG	35.4	46.6	38.2	52.3	2.8	5.1	18.2	1.9	0.4
09-10 Dominic McGuire	SF	9.9	37.1	0.0	37.5	7.6	20.3	4.4	0.8	0.7
04-05 Bo Outlaw	PF	9.4	35.4	0.0	35.3	8.1	16.5	7.2	1.2	3.0
04-05 Cory Alexander	SG	16.2	26.8	42.0	40.8	4.1	11.9	24.7	2.0	0.3

That's 16 usage points in the SL with 101.3% cumulative usage.  Will be just shy of 100% when Shaq comes out & the other Shaq comes in, but will be higher if Ruffin and/or Ward are in for Jordan and/or Kidd.  Dat 118 dreb% tho...

"enough" usage just finished the regular season.  59-23, 2nd in the division.  Went 5-5 down the stretch while the guy that won the division went 10-0 to storm by and take it.

Haven't put in the data yet, but I definitely feel like there is a real advantage to going for "more" usage.  More analysis in the coming days.

"enough" usage lost in round 1.  Extremely disappointing.  Team 2 just finished the regular season (72-10).  Here's some comparisons:

Team 1 had an estimated tov% of 12.7% going into the season.  In actuality, it was 15.6%.

That's a 22.8% increase.


Team 2 had an estimated tov% of 12.3% going into the season.  In actuality, it was 14.6%.

That's an 18.7% increase (21.9% less of an increase than team 1).


That, in my mind, is pretty huge.

ALSO:

On team 1 (51.3% efg), every single player except Steve Novak had lower efg% in the sim than they did IRL.  Novak maintained.
On team 2 (53.6% efg), almost every player either maintained or had a higher efg% in the sim than they did IRL despite playing in a division with much better defenders.

This pretty conclusively demonstrates a gigantic difference in cumulative usage to me.    Team one, while having 14 usage points in the SL, only had 95% cumulative.  Team two had 19 usage points in the SL & 111% cumulative.  I'm running the third squad (listed above) now that has between 100 & 111.  Results so far are pleasing.

Here's the spreadsheet where I'm keeping track of these teams.  It includes the first team I ran with "too much" usage that has inspired this experiment.  That team, fyi, won the whole enchilada.

spreadsheet is updated with the latest team in there now.


This team (only Shaq & Jordan, no cp3) had an estimated tov% of 13.935 IRL.
In the sim, it went 15.7%, or an increase of 12.7%.

As you can see, Jordan shot much better on this team with a 53.7% efg (a 4.2% increase over his 51.6% irl) than the other leagues.  I attribute much of this to having faced lesser defenders than the other leagues, as some of the other players shot below RL numbers while others shot above.

All in all, the trend definitely seems to show that there is less of an increase in turnovers by going above 100% cumulative usage in your lineup & less of a hit in fg%, though it might still always be there (never know what kind of defenders you'll be facing).

Something of interest to note:  In this last team, I went above & beyond on boards - well more than I did on the MJ, CP3 & SHAQ II team.  However, in the end, I ended up with a smaller percentage of rebounds than on that team.

This, along with the other bits of data I've collected during this experiment, indicate to me that there comes a point where getting 100+ cumulative usage in your lineup at all times trumps picking up a few extra percentages of rebounding.

The "experiment" is now closed.  I'm going to continue running teams along this theme but no longer as an experiment.  I feel that this data (and the success of these squads) has confirmed something long believed by others and in turn has made me a better owner.  I will now update the original post to reflect this experiment's findings.

original post overhauled.

bump

Bump

Bump-Diddety-Bump-Bump

the team penalty once in effect in a game impacts squad variants on the floor even when they do hit the base-line minimum: a theory 

Name	GP	GS	Min	Pts	FG%	3Pt%	FT%	OReb	Reb	Ast	TO	Stl	Blk	PF
66-67 Wilt Chamberlain	82	82	47.0	22.2	65.1	20.0	45.4	5.7	15.0	7.2	4.0	1.7	3.9	1.8
96-97 Karl Malone	82	82	37.4	28.7	53.5	8.7	76.2	2.2	8.8	5.5	3.8	1.8	0.8	2.8
90-91 Dennis Rodman	82	82	33.9	7.7	44.2	19.6	61.2	3.5	10.7	1.6	1.5	1.2	0.7	3.5
83-84 Magic Johnson	82	80	31.9	14.5	53.2	17.5	82.4	1.1	4.9	9.7	3.4	2.1	0.5	2.1
95-96 Eddie Jones	82	82	27.1	15.0	45.2	37.7	76.0	0.4	2.2	3.4	2.0	1.7	0.5	2.1
07-08 Jordan Farmar	82	2	20.1	10.0	41.3	37.3	69.6	0.4	2.0	2.8	2.0	1.1	0.0	1.2
07-08 Sasha Vujacic	74	0	16.8	10.0	43.5	42.0	80.0	0.1	1.4	1.2	1.1	0.6	0.1	1.4
12-13 Josh McRoberts	82	0	9.4	2.8	41.0	13.6	75.0	0.4	1.8	1.0	0.5	0.4	0.3	0.7
11-12 Jordan Hill	82	0	7.1	2.7	44.6	0.0	81.8	0.6	2.0	0.2	0.6	0.1	0.3	0.8
94-95 Tracy Murray	82	0	6.7	2.9	35.6	35.0	84.6	0.2	0.6	0.5	0.8	0.2	0.0	0.9
05-06 Brian Grant	67	0	3.2	0.9	36.5	0.0	90.0	0.1	0.5	0.0	0.2	0.1	0.0	0.6
07-08 Kwame Brown	36	0	5.3	1.0	30.6	0	30.4	0.4	1.2	0.6	0.6	0.3	0.0	0.6
TOTALS	82			116.7	50.1	35.8	68.8	15.0	50.2	33.4	20.0	11.1	7.2	17.9
OPPONENTS	82			117.6	50.8	38.4	75.6	13.1	48.4	31.5	19.5	11.3	8.9	18.7