Sam Mitchell's plan and development (Warning- +/- stats involved)

Any And All Things T-Wolves Related
User avatar
longstrangetrip [enjin:6600564]
Posts: 9432
Joined: Tue Jul 09, 2013 12:00 am

Re: Sam Mitchell's plan and development (Warning- +/- stats involved)

Post by longstrangetrip [enjin:6600564] »

khans2k5 wrote:
longstrangetrip wrote:
khans2k5 wrote:
Q12543 wrote:
alexftbl8181 wrote:
Q12543 wrote:
alexftbl8181 wrote:
Q12543 wrote:
alexftbl8181 wrote:You so sure as hell ask a starting player on your team to be able to do more then set picks, space the floor, and move the ball. Pretty sure you could find any Jeff Adrian off the scrap heap for 50 bucks to do that


....or Robbie Hummel. He was always a guy that knew how to play and where to be on the floor.

Can you now explain to me how KG and Prince are hurting the team on defense or are they being carried by the others in that department as well?


hurting the team? no. As good as people think they're looking? Hell no. Good enough on defense to offset their offense? Oh Hellll nooooo


Well, according to the numbers, they are a net positive. That's a fact through 4 games, no matter how much you don't like it. The question is whether it's sustainable. You obviously don't think so, but through the first four games it's an undeniable fact.


according to some numbers, not others


Through 4 games, when KG and Prince are on the floor, we outscore our opponents. I'd say that's a fairly relevant outcome, but perhaps you can explain why it doesn't matter.

Can you point to a different statistic that says we don't outscore our opponent when the two of them are on the floor?


Using their plus/minus stats to prove their worth is the biggest bullshit stretch on this board. They score a combined 4 points a game. Their usage is at about 15% combined. They're getting carried hard core when they are out on the court by Rubio and Towns. Everything except plus/minus says they are playing terrible this year. Why does 1 stat that is team based trump every other stat there is? That's just a horrible use of statistics to prove a point. Rubio is almost at a full win share (0.9) in 4 games where we won 2 of those games (the statistical equivalent of him almost winning 1 of our two wins by himself). Towns is at .5. KG and Prince...a rounded up 0.1 each (which is in line with Dieng and Payne). Payne surpassed KG's WS/48 in his 17 minutes this year over KG's 63. 1 stat doesn't trump all. They deserve 15% credit for their plus/minus numbers because that's all they are giving the rest of the team. The rest is credited to the guys carrying their corpses on a nightly basis. Ask yourself this. If they are so damn good/valuable because of their plus/minus, why doesn't Sam close out games with his starting 5? Statistically they aren't going to lose him the game if they have a lead and their plus/minus is so big they theoretically could come back from a big deficit pretty easily for come from behind victories as well. Is he just a fucking moron or does he see what every other stat says and goes with a different option because he knows they'd struggle to score and get stops to close out games? Plus/minus is a terrible stat to use if everything else is bad.


Khans asks "why does 1 stat that is team based trump every other stat there is"? I love stats because they provide insights into individual aspects of a players' game. But ultimately there is only one stat that matters, and that is team +/- at the end of every game. If you are plus, you win...if you are minus, you lose. It's that simple. So, that is why this team-based stat trumps every other stat there is. And why how individual players contribute to this +/- is so meaningful.

I fully admit that +/- can be misused in a very limited sample size. Damjan Rudez could play garbage minutes with a teammate who is making everything against a team that is missing everything, and look like a great player with his +/-. Four games is also still a small sample size, and KG and Prince may drop from their lofty positions as the season progresses. But I would argue that the way Sam is using the two vets makes it very easy to analyze their +/-. They are playing 15-20 minutes at the start of halves almost exclusively against the opponents' starters. And the fact that they are dramatically winning the +/- battle against the other teams' starters is very meaningful. I hear a lot of anger directed toward Sam for starting halves with KG and Tay, but I have yet to hear a legitimate numerically-based argument as to why his starting lineup isn't effective.

This one fact is indisputable...when KG and Tay are on the court, we outscore our opponent more than with any other player. How can that not be a good thing?


Robbie was almost always positive in plus/minus yet we still lost 66 games. That's why plus/minus actually isn't a good stat. There are countless examples of guys with negative plus/minuses in wins and positive plus/minuses in losses. The game is about scoring more than your opponent. They score 4 points combined. Are you telling me they give up 4 points or less to their direct opponents because that is literally the only way they earn their plus/minus numbers otherwise it is almost 100% because of who they play with that they have those numbers? There's way too much noise in plus/minus. Basketball-reference tells a much different story with their offensive and defensive ratings and 82games will tell us at the end of the month as well how accurate these plus/minus stats are for our starting 5 man unit versus other possibilities we've rolled out.


First, a quick correction of one of your statements. As much as I liked Robbie Hummel's overall game, he actually ranked poorly in +/- last season...at -7.2 per 36 minutes, he ranked 12th on the team (Rubio was first among players with sufficient minutes at -.7). And as more evidence that over the course of a season +/- is a good evaluation of a player's value, you only need to look at who led the league in +/- per 36 last season...Steph Curry at 12.7.

I don't think the parties on either side of this discussion can agree because of one basic difference in perspective. The Tay/KG detractors continue to cite points scored by individuals as the key measurement of offensive performance, while the Tay/KG supporters define offense much more broadly and choose to focus on points scored by the team. And as long as the points scored by the team are more than the points given up (the definition of +/-, by the way), it doesn't really matter who actually puts the ball in the basket...we still win!

Tay and KG are never going to win any individual scoring contests...that's not their role. But as long as the 5 they are part of continue to outscore the opposing teams' starters so significantly, they are welcome to start on my team.
mjs34
Posts: 2408
Joined: Sun Jul 07, 2013 12:00 am

Re: Sam Mitchell's plan and development (Warning- +/- stats involved)

Post by mjs34 »

khans2k5 wrote:

Robbie was almost always positive in plus/minus yet we still lost 66 games. That's why plus/minus actually isn't a good stat. There are countless examples of guys with negative plus/minuses in wins and positive plus/minuses in losses. The game is about scoring more than your opponent. They score 4 points combined. Are you telling me they give up 4 points or less to their direct opponents because that is literally the only way they earn their plus/minus numbers otherwise it is almost 100% because of who they play with that they have those numbers? There's way too much noise in plus/minus. Basketball-reference tells a much different story with their offensive and defensive ratings and 82games will tell us at the end of the month as well how accurate these plus/minus stats are for our starting 5 man unit versus other possibilities we've rolled out.


Khans, you're putting the the noise into this. KG and Tay are for the most part only playing with the other three starters against the opponents starters. That gives a definite +/- in those situations, so you can calculate the difference with other guys replacing KG and Tay. Robbie's +\- wasn't necessarily higher than the other starters. If he was only playing with that group and was higher, than you would say he was a plus player. Even further, is that being a plus player doesn't necessarily extend to wins. The Wolves starting lineup in Love's last season was significantly better than the opponents in +\-, but still won only half their games.

KG and Tay don't have to hold the opponents to 4 pts because their opponents are likely taking signifcantly more shots.

I personally think that starting out with tougher defense not only sets a tone, but also holds down the confidence of your opponent on shooting. Something that can have lasting effects throughout the game.
User avatar
SameOldNudityDrew
Posts: 3127
Joined: Sat Jul 20, 2013 12:00 am

Re: Sam Mitchell's plan and development (Warning- +/- stats involved)

Post by SameOldNudityDrew »

I'm with Tim, Q, and LST on this one--that seems to happen a lot.

I do want to see an end to the LaVine at backup PG business, although I understand Sam is in a tough position. The team just plays better with Prince so far, and doing so focuses development on our two most important investments for the future (Towns and Wiggins). And I also want to see more Bazz, and Martin has been a consistent scorer. I'm not as high on Zach as some guys here, but he's also an investment in the future, and an exciting one on the floor. Man, though, it's tough to watch the offense when he's at the point. I'd like to see Tyus get some minutes at the point against some backups who wouldn't overwhelm him physically. Maybe playing Bazz at the 4 for short stretches could help open up some minutes for Zach at the 2 instead of the 1.
User avatar
SameOldNudityDrew
Posts: 3127
Joined: Sat Jul 20, 2013 12:00 am

Re: Sam Mitchell's plan and development (Warning- +/- stats involved)

Post by SameOldNudityDrew »

sjm34 wrote:
khans2k5 wrote:

Robbie was almost always positive in plus/minus yet we still lost 66 games. That's why plus/minus actually isn't a good stat. There are countless examples of guys with negative plus/minuses in wins and positive plus/minuses in losses. The game is about scoring more than your opponent. They score 4 points combined. Are you telling me they give up 4 points or less to their direct opponents because that is literally the only way they earn their plus/minus numbers otherwise it is almost 100% because of who they play with that they have those numbers? There's way too much noise in plus/minus. Basketball-reference tells a much different story with their offensive and defensive ratings and 82games will tell us at the end of the month as well how accurate these plus/minus stats are for our starting 5 man unit versus other possibilities we've rolled out.


Khans, you're putting the the noise into this. KG and Tay are for the most part only playing with the other three starters against the opponents starters. That gives a definite +/- in those situations, so you can calculate the difference with other guys replacing KG and Tay. Robbie's +\- wasn't necessarily higher than the other starters. If he was only playing with that group and was higher, than you would say he was a plus player. Even further, is that being a plus player doesn't necessarily extend to wins. The Wolves starting lineup in Love's last season was significantly better than the opponents in +\-, but still won only half their games.

KG and Tay don't have to hold the opponents to 4 pts because their opponents are likely taking signifcantly more shots.

I personally think that starting out with tougher defense not only sets a tone, but also holds down the confidence of your opponent on shooting. Something that can have lasting effects throughout the game.


Boom. Exactly.
User avatar
khans2k5 [enjin:6608728]
Posts: 6414
Joined: Tue Jul 09, 2013 12:00 am

Re: Sam Mitchell's plan and development (Warning- +/- stats involved)

Post by khans2k5 [enjin:6608728] »

The setting the tone defensively argument is crap also. Here's our starting unit's defensive rating:

Ricky 97.6
Wiggins 92.7
Prince 85.5
KG 79.4
Towns 91.9

The second unit that replaces them:

Lavine 104.3
Martin 105.2
Bazz 101.6
NB 99.1
Dieng 104.3

And that's against worse competition. What tone is being set defensively if the second unit comes in and is a bunch of matadors? Setting a tone means everyone starts and keeps playing defense the whole game. At the end of the day the Nuggets game is skewing all of the team defensive data with their horrid 78 point showing. The other three games we gave up 112, 106 and 96 and we are -16 in those games. I'm glad to see the board is giving credit to two corpses who are just on the court watching their +/- go up because of everyone around them. By the way, NBAdotcom has our starters as the third best lineup we have for +/-. Ricky/Martin/Prince/NB/Dieng is at the top at +8 and in second is Ricky/Lavine/Wiggins/NB/Towns at +7. Our starters are third at +5.8. Since we are ok with very small sample sizes I decided to include that.
User avatar
longstrangetrip [enjin:6600564]
Posts: 9432
Joined: Tue Jul 09, 2013 12:00 am

Re: Sam Mitchell's plan and development (Warning- +/- stats involved)

Post by longstrangetrip [enjin:6600564] »

One of the things I love about this board is that we have so many lively discussions about tough issues between smart guys...this is one of them. Contrast this with the knuckleheads over at rubechat, where there is complete agreement about these opinions:

1) Sam Mitchell is a complete idiot
2) Tay Prince should never see the court again

At least here we have a good discussion going about these two matters. Keep it up, guys!

I thought sjm really nailed it when he talked about the psychological impact of making a defensive statement at the start of games. Most of us have played basketball at one level or another, and we know how much better our games are when we start out hot. But then khans had a good comeback when he displayed the horrible defensive ratings of our second team. I think what khans is missing, though, is that our starters are making a defensive statement against their starters which hopefully continues to have an impact on their starters the entire night. Our reserves are largely getting lit up by their reserves.
User avatar
Q12543 [enjin:6621299]
Posts: 13844
Joined: Thu Jul 11, 2013 12:00 am

Re: Sam Mitchell's plan and development (Warning- +/- stats involved)

Post by Q12543 [enjin:6621299] »

khans2k5 wrote:The setting the tone defensively argument is crap also. Here's our starting unit's defensive rating:

Ricky 97.6
Wiggins 92.7
Prince 85.5
KG 79.4
Towns 91.9

The second unit that replaces them:

Lavine 104.3
Martin 105.2
Bazz 101.6
NB 99.1
Dieng 104.3

And that's against worse competition. What tone is being set defensively if the second unit comes in and is a bunch of matadors? Setting a tone means everyone starts and keeps playing defense the whole game. At the end of the day the Nuggets game is skewing all of the team defensive data with their horrid 78 point showing. The other three games we gave up 112, 106 and 96 and we are -16 in those games. I'm glad to see the board is giving credit to two corpses who are just on the court watching their +/- go up because of everyone around them. By the way, NBAdotcom has our starters as the third best lineup we have for +/-. Ricky/Martin/Prince/NB/Dieng is at the top at +8 and in second is Ricky/Lavine/Wiggins/NB/Towns at +7. Our starters are third at +5.8. Since we are ok with very small sample sizes I decided to include that.


....Except the second unit isn't being a bunch of matadors - at least not yet. Those are actually solid defensive ratings. Go look at LaVine, Martin, Bazz, and Dieng's defensive ratings from last season. There is a dramatic difference, so I would absolutely say there is some credence behind the defensive tone narrative.
User avatar
MikkeMan
Posts: 817
Joined: Fri Jul 19, 2013 12:00 am

Re: Sam Mitchell's plan and development (Warning- +/- stats involved)

Post by MikkeMan »

Garnett has still at least two elite skills. Team defense and defensive rebounding. His Drtg numbers in last two years have been 100.5 and 102.5 with most minutes played in pretty bad Brooklyn team. Those are pretty good numbers compared what Wolves rotational players have had in last couple of years. (Better than Rubio's numbers for example)

Also if you check Garnett's defensive rebounding rate from advanced stats, it is still in same (or acually even better) level than during his prime years. Last year it was 31.1 and in previous year it was 32.1. Both numbers were better than numbers during his all star years.
User avatar
khans2k5 [enjin:6608728]
Posts: 6414
Joined: Tue Jul 09, 2013 12:00 am

Re: Sam Mitchell's plan and development (Warning- +/- stats involved)

Post by khans2k5 [enjin:6608728] »

Q12543 wrote:
khans2k5 wrote:The setting the tone defensively argument is crap also. Here's our starting unit's defensive rating:

Ricky 97.6
Wiggins 92.7
Prince 85.5
KG 79.4
Towns 91.9

The second unit that replaces them:

Lavine 104.3
Martin 105.2
Bazz 101.6
NB 99.1
Dieng 104.3

And that's against worse competition. What tone is being set defensively if the second unit comes in and is a bunch of matadors? Setting a tone means everyone starts and keeps playing defense the whole game. At the end of the day the Nuggets game is skewing all of the team defensive data with their horrid 78 point showing. The other three games we gave up 112, 106 and 96 and we are -16 in those games. I'm glad to see the board is giving credit to two corpses who are just on the court watching their +/- go up because of everyone around them. By the way, NBAdotcom has our starters as the third best lineup we have for +/-. Ricky/Martin/Prince/NB/Dieng is at the top at +8 and in second is Ricky/Lavine/Wiggins/NB/Towns at +7. Our starters are third at +5.8. Since we are ok with very small sample sizes I decided to include that.


....Except the second unit isn't being a bunch of matadors - at least not yet. Those are actually solid defensive ratings. Go look at LaVine, Martin, Bazz, and Dieng's defensive ratings from last season. There is a dramatic difference, so I would absolutely say there is some credence behind the defensive tone narrative.


The only solid rating among them is NB. Even Bazz's rating on a team level is lottery level defense and Lavine/Martin/Dieng would be bottom 9 in the league. How is that solid? It's not, it's just better than last year. You can't be bottom 10 worthy in something and be solid.
User avatar
Q12543 [enjin:6621299]
Posts: 13844
Joined: Thu Jul 11, 2013 12:00 am

Re: Sam Mitchell's plan and development (Warning- +/- stats involved)

Post by Q12543 [enjin:6621299] »

khans2k5 wrote:
Q12543 wrote:
khans2k5 wrote:The setting the tone defensively argument is crap also. Here's our starting unit's defensive rating:

Ricky 97.6
Wiggins 92.7
Prince 85.5
KG 79.4
Towns 91.9

The second unit that replaces them:

Lavine 104.3
Martin 105.2
Bazz 101.6
NB 99.1
Dieng 104.3

And that's against worse competition. What tone is being set defensively if the second unit comes in and is a bunch of matadors? Setting a tone means everyone starts and keeps playing defense the whole game. At the end of the day the Nuggets game is skewing all of the team defensive data with their horrid 78 point showing. The other three games we gave up 112, 106 and 96 and we are -16 in those games. I'm glad to see the board is giving credit to two corpses who are just on the court watching their +/- go up because of everyone around them. By the way, NBAdotcom has our starters as the third best lineup we have for +/-. Ricky/Martin/Prince/NB/Dieng is at the top at +8 and in second is Ricky/Lavine/Wiggins/NB/Towns at +7. Our starters are third at +5.8. Since we are ok with very small sample sizes I decided to include that.


....Except the second unit isn't being a bunch of matadors - at least not yet. Those are actually solid defensive ratings. Go look at LaVine, Martin, Bazz, and Dieng's defensive ratings from last season. There is a dramatic difference, so I would absolutely say there is some credence behind the defensive tone narrative.


The only solid rating among them is NB. Even Bazz's rating on a team level is lottery level defense and Lavine/Martin/Dieng would be bottom 9 in the league. How is that solid? It's not, it's just better than last year. You can't be bottom 10 worthy in something and be solid.


I think collectively it's solid, especially compared to last season, although I'm not sure they can hang on at this rate given the fact we haven't played OKC, Golden State, Clippers, etc. yet.

The point is that it is a much better rate than last season, which suggests to me that there is some merit behind the fact our starting unit out of the gate is setting the tone for the rest of squad (and the fact that Mitchell might actually be getting through to these guys).
Post Reply