FanPost

Leave Josh Smith Alone! (Or, comparing Detroit's big man pairings)

Raj Mehta-USA TODAY Sports

What don't you guys get? Josh Smith was playing out of position all year long!!! How hard is that to understand? The coach wanted him to shoot all of those threes to spread the floor. We should trade Greg Monroe for a bag of chips dicks and let Smith start at PF next to Andre Drummond because Monroe needs the same space Drummond does to be effective. We are not going to win with two players who both need to be in the paint to be effective. You're not going to pay a player $13.5M a year and make him come off the bench. Also, Monroe is a C, not a PF. Here, I'll even show you data that proves Josh Smith was a better PF than Monroe was for the Pistons.

Looking at stats.nba.com, I looked at all of the 5 man lineup data. I'm going to compare lineups of Drummond/Monroe/???/???/???, Monroe/Smith/???/???/???, and Drummond/Smith/???/???/???.

Stat Drummond/Monroe Monroe/Smith Drummond/Smith
MIN 302 711 511
Best OffRtg 189.9 (2 min) 166.7 (2 min) 202.2 (2 min)
Worst OffRtg 33.3 (3 min) 55.1 (2 min) 58.9 (6 min)
AVG OffRtg 110.1 103.6 111.2
Best DefRtg 18.4 (2 min) 40.0 (2 min) 37.5 (4 min)
Worst DefRtg 172.1 (6 min) 141.2 (3 min) 180.2 (2 min)
AVG DefRtg 105.5 97.7 102.7
Best NetRtg 171.5 (2 min) 71.6 (19 min) 140.2 (4 min)
Worst NetRtg -113.2 (6 min) -39.1 (17 min) -82.3 (2 min)
AVG NetRtg 4.68 5.85 8.52
AST%* 44.24% 57.44% 49.27%
AST/TO 0.94 1.17 1.00
AST Ratio 13.1 15.5 15.2
OREB%* 41.57% 28.72% 38.63%
DREB%* 70.49% 73.20% 75.80%
REB%* 57.35% 49.97% 57.66%
TO Ratio 17.8 12.9 15.9
eFG%* 53.29% 46.89% 50.15%
TS%* 54.66% 51.34% 53.08%
PACE 96.97 96.89 99.51
PIE 39.12% 59.74% 55.81%
PTS OFF TO/36 14.7 14.6 16.4
2nd PTS/36 16.3 13.6 17.6
FBPs/36 14.2 12.0 18.9
PITP/36 38.4 30.0 47.0
OPP PTS OFF TO/36 18.9 12.99 13.7
OPP 2nd PTS/36 9.0 13.2 8.5
OPP FBPs/36 9.8 10.22 7.6
OPP PITP/36 34.6 34.18 25.9
FTA Rate 0.43 0.42 0.34
OPP eFG%* 56.45% 48.35% 50.38%
OPP FTA Rate 0.24 0.38 0.31
OPP TO Ratio 19.07 16.91 15.14
OPP OREB%* 23.25% 26.81% 21.07%
FGM/36 30.9 26.1 33.0
FGA/36 65.1 64.24 67.5
FG% 47.39% 40.62% 48.84%
3FGM/36 5.1 5.4 4.6
3FGA/36 12.0 18.9 15.3
3FG% 42.26% 28.67% 29.95%
FTM/36 15.8 16.7 14.0
FTA/36 25.2 24.7 21.9
FT% 62.89% 67.51% 63.97%
OREB/36 15.6 12.42 13.2
DREB/36 21.3 23.3 26.9
REB/36 36.9 35.72 40.1
AST/36 14.6 15.5 17.6
TOV/36 12.5 9.1 11.7
STL/36 8.3 6.2 6.3
BLK/36 2.8 4.5 5.4
BLKA/36 2.2 3.5 2.6
PF/36 13.9 15.9 15.7
PFD/36 21.1 17.0 16.8
PTS/36 82.65 74.30 84.49
PM/36 8.5 2.5 9.0

That's a lot of numbers!!! Let me explain the values for some of them. Anything that has a "*" next to it is what I call fuzzy logic. For instance, I could pull the TS% from stats.nba.com, but I didn't see a way to total it for all lineups that only had those specific players in the lineup. Therefore, using excel, I simply averaged the columns. Anyone that knows anything about math knows that averaging averages does not give you the true average. That was the case with quite a few of the percentages that I couldn't calculate (easily that is, I'm doing all of this with limited time at work). Also, the Best/Worst of the ratings are subjective. I thought of trying to calculate those per minute played, but it was too much work. Also, those Best/Worst did not include those where either the OffRtg or DefRtg were 0, which was about 4 times.

Anyways, if you count up the green cells, it will clearly show you that the best frontcourt player with Andre Drummond is Josh Smith by a count of 22-14. There are 8 more categories that Drummond/Smith are better than Drummond/Monroe. It's so clear!!! As a matter of fact, Monroe/Smith are better than Monroe/Drummond by 1, so we should obviously trade Drummond.

Honestly though, some of this kind of surprised me. I hadn't realized that Monroe/Smith had spent so much more time on the floor than just Drummond/Monroe. I started this post as a "SHUT UP SMITH FANS!!!" but I guess I can kind of see why they may vouch for him. HOWEVER, I still think that due to Monroe's youth compared to Smith's, his character, and his potential (not based on athleticism) are more than enough for me to want to keep him over Smith.

UPDATE - Maybe I should show my work?

Ok, so as I was catching up on comments on this thread this morning, a few things came to mind. First, I really wish I wouldn't have deleted the excel file that I had with all of the lineup data I saved and manipulated (I had to recreate it this morning). Secondly, I think I messed up on some of my math.

As I was going over each stat in the above chart, when I got to "AVG OffRtg" for each pairing, it hit me, I didn't weigh it down. I had simply did an AVERAGE excel function on the values in that column. So this morning I decided to see how different the data would look weighed down, and it was quite different.

Stat Drummond/Monroe Monroe/Smith Drummond/Smith
AVG OffRtg 109.17 103.15 103.36
AVG DefRtg 110.12 99.32 104.58
AVG NetRtg -0.94 3.84 -1.22
FGM/36 30.9 28.0 30.1
FGA/36 63.8 62.5 64.8
FG% 48.41% 44.84% 46.52%
3PFGM/36 3.75 5.21 3.5
3PFGA/36 9.9 14.3 13.3
3PFG% 37.93% 36.38% 26.16%
FTM/36 14.0 15.2 13.5
FTA/36 23.2 21.9 21.4
FT% 60.37% 69.17% 63.36%
OREB/36 13.3 9.5 10.7
DREB/36 21.4 22.9 24.2
REB/36 35.0 32.0 35.0
AST/36 15.0 15.0 16.0
TOV/36 11.9 9.1 11.4
STL/36 7.8 5.7 6.5
BLK/36 2.4 3.9 5.8
BLKA/36 2.6 3.6 4.0
PF/36 15.0 15.0 15.0
PFD/36 17.0 16.0 16.0
PTS/36 79.5 76.5 77.3
+/- /36 1.2 2.9 -0.9
PTS OFF TO/36 13.6 13.4 14.5
2nd PTS/36 14.4 12.1 13.7
FBPs/36 12.3 12.4 15.3
PITP/36 41.7 35.2 40.9
OPP PTS OFF TO/36 15.9 9.6 12.8
OPP 2nd PTS/36 10.5 10.8 11.2
OPP FBPs/36 10.3 10.4 9.0
OPP PITP/36 34.5 31.9 29.8

So if I were to replace each of the new weighted stats into the first chart in the article, and I counted all of the green cells, Drummond/Monroe would have 22, Monroe/Smith would have 18, and Drummond/Smith would have 14. Phew, now I kind of feel better as that makes more sense. However, you have to take into account the Lack of Control as well.

Explaining the collected data

I want to state this for those that don't take the time to read all of the comments. I collected the 5 man lineup data from stats.nba.com. I placed all 250 different lineup combinations in an excel spreadsheet pulling the Basic Columns, Advanced, Misc, and Four Factors data. I then separated out the names of the players into their own columns and then had three columns I used to determine which big man pairing was used:

=IF(AND(OR(A2="Drummond,Andre",B2="Drummond,Andre",
C2="Drummond,Andre",D2="Drummond,Andre",E2="Drummond,Andre"),
OR(A2="Monroe,Greg",B2="Monroe,Greg",C2="Monroe,Greg",D2="Monroe,Greg",
E2="Monroe,Greg"),AND(NOT(A2="Smith,Josh"),NOT(B2="Smith,Josh"),
NOT(C2="Smith,Josh"),NOT(D2="Smith,Josh"),NOT(E2="Smith,Josh"))),"DM","")

=IF(AND(OR(A2="Drummond,Andre",B2="Drummond,Andre",
C2="Drummond,Andre",D2="Drummond,Andre",E2="Drummond,Andre"),
OR(A2="Smith,Josh",B2="Smith,Josh",C2="Smith,Josh",D2="Smith,Josh",
E2="Smith,Josh"),AND(NOT(A2="Monroe,Greg"),NOT(B2="Monroe,Greg"),
NOT(C2="Monroe,Greg"),NOT(D2="Monroe,Greg"),
NOT(E2="Monroe,Greg"))),"DS","")

=IF(AND(OR(A2="Monroe,Greg",B2="Monroe,Greg",C2="Monroe,Greg",
D2="Monroe,Greg",E2="Monroe,Greg"),OR(A2="Smith,Josh",B2="Smith,Josh",
C2="Smith,Josh",D2="Smith,Josh",E2="Smith,Josh"),
AND(NOT(A2="Drummond,Andre"),NOT(B2="Drummond,Andre"),
NOT(C2="Drummond,Andre"),NOT(D2="Drummond,Andre"),
NOT(E2="Drummond,Andre"))),"MS","")

I then sorted by those three columns so I had them grouped together. I then inserted rows to use as Total/Average rows. For MIN, I found the total, and then for every other column, I HAD (before this update) used an AVERAGE() function and simply dragged it across each column. I admitted that the percentages would be fuzzy logic, but I had forgotten to weigh down the AVG Rtg columns. Here's what the weighed down function looked like for Drummond/Monroe:

=SUM(AF2*$J$2,AF3*$J$3,AF4*$J$4,AF5*$J$5,AF6*$J$6,
AF7*$J$7,AF8*$J$8,AF9*$J$9,AF10*$J$10,AF11*$J$11,
AF12*$J$12,AF13*$J$13,AF14*$J$14,AF15*$J$15,AF16*$J$16,
AF17*$J$17,AF18*$J$18,AF19*$J$19,AF20*$J$20,AF21*$J$21,
AF22*$J$22,AF23*$J$23,AF24*$J$24,AF25*$J$25)/SUM($J$2:$J$25)

The AF column has the OffRtg values and the J column has the minutes for each lineup combination. I did the same for the DefRtg, NetRtg and any /36 column, as well as did the same for Drummond/Monroe and Monroe/Smith.

Lack of Control

As MFMP has stated in the comments, there is no exact control for these numbers. Drummond and Monroe were in 24 lineups that didn't include Smith, Drummond and Smith were in 32 lineups that didn't include Monroe, and Monroe and Smith were in 30 lineups that didn't include Drummond. Then you have the discrepancy of those minutes played on top of which teams they were against. So you have to take all of these numbers with a LARGE grain of salt.

FanPosts are user-created posts from the Detroit Bad Boys community and do not necessarily reflect the opinions of all fans or the staff at DBB. The DBB staff reserves the right at any time to edit the contents of FanPosts as they reasonably see fit.