TWolves?

Home for all your discussion of basketball statistical analysis.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

TWolves?

Post by Crow »

Trade one of Wiggins or LaVine. Play Jones 20 plus minutes per game for next 10 games. Unless he fails, trade one or both of Rubio and Dunn.
shadow
Posts: 252
Joined: Fri May 29, 2015 3:38 am

Re: TWolves?

Post by shadow »

I don't necessarily think Minnesota is in terrible shape. They have had the worst luck in close games in the entire league so far this year. They're on pace for only 29 wins based on their current win percentage, but their power rating implies they should be on pace to win 39-40 games (right in line with Vegas win total of ~40) if they simply went 0.500 in close games, as teams are expected to do in the long run. It's pretty rare for a team to have a difference of more than 4-5 games between their expected and actual wins over a full season. I believe Atlanta led the league two years ago when they won 60 games with a luck factor of around +5.5 wins. The next season they regressed pretty heavily and only won 48 games with basically the same exact roster.

That being said, yes, Wiggins and Lavine need to show a pulse on defense. I'm a little bit shocked they're still only 23rd in AdjDRTG under Thibs considering his past success with improving the Bulls and Celtics defenses. Perhaps the way the league has evolved to more pace and space style of play is limiting the effectiveness of his schemes, now that you have guys like Brook Lopez and Marc Gasol popping out for 3's on pick and rolls, it becomes a problem when your big sag on pick and rolls to cut off driving lanes. Then again, maybe the schemes are fine and it's just a lack of effort or willingness to buy into his system from the players.
pdevos
Posts: 36
Joined: Fri Jun 26, 2015 6:18 pm
Contact:

Re: TWolves?

Post by pdevos »

The Wolves do score points on offense fairly well -- could be a lot better though with personnel changes. The irony of Ricky Rubio is he's a great passer but an awful shooter so it negates much of his passing except in pick-n-roll where the post is moving/rolling downhill and he's able to make tremendous pocket and wrap around the defender passes. But the rest of the defense doesn't "break down" to try to help on Rubio. It's really only if Towns can make them compromise.

Then there's LaVine and Wiggins who can off the dribble or "bounce" make teams compromise, but they lack the awareness and ability [as it currently stands] to make those passes. It's really hard to win games when you have high usage players (> 20%) who dish out 3 apg or less.

The other part is they make SLOW decisions on both offense and defense. When a double team comes, they aren't thinking ahead -- they're reacting. Passes are slow on rotation. Same with on defense. When Brandon Rush played we saw a cerebrally ahead player who knew exactly what to do when the ball hit his hands or when a double was going to Towns.

How can fast these guys learn? I don't know. But I'd hate to say it, it might be another couple years with these guys. And even so -- do you really need 3 ball dominant scorers (Wiggins, Towns, LaVine) and where Towns is the best passer (3.1 apg) out of the bunch? Do guys become good passers?

I'd honestly feel a better ball dominant scorer (e.g. Kyle Lowry, Eric Bledsoe) with say, Jared Dudley, Brandon Rush, Trevor Ariza types with Towns would probably get us a lot more wins than Wiggins and LaVine would.

I'd also love to see a stretch 4 instead of Gorgei. He allows his man to creep over to Towns on drives (or anyone who drives). He strangely enough can hit mid-long range, but seems to prefer to float closer to the lane.

Maybe that's it...all our guys float vs are mentally ahead of the game and playing the other team. I've found this is something you see more common with incredible athletes who by luxury of their athleticism can 'react' (sprint, jump) to make up for lack of awareness for many levels of basketball. But the NBA, where the athleticism gap narrows and mental processing is even better -- really exposes that.
pdevos
Posts: 36
Joined: Fri Jun 26, 2015 6:18 pm
Contact:

Re: TWolves?

Post by pdevos »

Crow wrote:Trade one of Wiggins or LaVine. Play Jones 20 plus minutes per game for next 10 games. Unless he fails, trade one or both of Rubio and Dunn.
I would definitely be all for this. I'd try to trade for a more team oriented player (better playmaker/passer, better defense) and try to 'tank' to get into the top 5 picks. As much as Thibs says he's not playing Rush, Hill, or much of his bench is "the young guys have to learn" -- maybe he could take the approach you said. Jones & Dunn play 20-30 mpg.

I wouldn't be surprised if CLE or GS would love Brandon Rush.

Maybe Bjelica + something could pry one of those two 1st RDers from Boston they picked last year (Yabusele, Zizic) and get more minutes as well -- I actually, as much as Shabazz isn't a defender -- I like it when he plays the small-ball-4 often. As he spaces well for Towns.

All frustration fodder likely...but I'm definitely with you. I think a trade needs to happen for a player with more ball awareness.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: TWolves?

Post by Crow »

Is there any new approach to player projections (in NBA already or in draft) or lineup interactions you have pursued and / or want to discuss? It appears from one of your links that you have a lot of technical skills available for use.

I'll draw attention to your mention of player thinking / planning ahead / reacting / decision-making. While non-public SportVu megadata might provide better insight into player "reactions" and "decision-making" times, abilities or types, have you done anything with boxscore or public player tracking data to organize the data, find significant questions related to this area of study, posit theories and test them? I think there could be new stuff found / done here.

One thing that I have been interested in but haven't pursued much is play by play patterns. Do players use what happened on previous play to get "a lot" or "more" on the next ones? Do good teams / lineups / players use similar play sequences more often or do they show more "creativity"? Are there offense / defenses that the data suggests know the opponent patterns and anticipate them better than average? Are there actions within plays that are mainly there for their future positive impact or misdirection as opposed to their immediate relevancy? David Locke touched on this topic re:Spurs on his Locked On NBA podcast recently and probably triggered this mention. If you disrupt the pattern does the train wreck? How much risk / effort should you put into breaking the pattern? Does it get harder as the game goes along as the opposing team and / or coach can think / react better? How long can you milk a successful pattern before it is better to switch to a new one? What about regular season to playoff comparisons? Who changes most / least? Who "saw" what they expected vs got the biggest playoff surprise? Are there players who disrupt or don't use patterns that their team uses better with others? Can you get better results by putting players together who think / plan ahead / react in compatible ways and breaking bad patterns or losing disruptions above & beyond what you can do with other stat study and lineup construction approaches?
pdevos
Posts: 36
Joined: Fri Jun 26, 2015 6:18 pm
Contact:

Re: TWolves?

Post by pdevos »

Admittedly, I was aiming lower as in more conversational fodder on a first post. Lesson learned.

After this post I went around many other rooms and found that quite often you deliver, quite frequently, heavy analytical questions. So I actually thought -- oh Crow -- is going to nail my more conversational post with analytics....

"Crow" is not a noun in your case, but a verb. Peter Pan would be proud. :)
Crow wrote:I'll draw attention to your mention of player thinking / planning ahead / reacting / decision-making. While non-public SportVu megadata might provide better insight into player "reactions" and "decision-making" times, abilities or types, have you done anything with boxscore or public player tracking data to organize the data, find significant questions related to this area of study, posit theories and test them? I think there could be new stuff found / done here.
As you said, the non-aggregated SportsVU data is only available to a few. At least that I'm aware of. You can tell Stephen Shea has access and a few others by the type of content they roll out. He's been able to identify which defenses leave their man to aid on the drive or stay glued to their man despite that. Most of which could still be seen from watching the game, but it does make it a bit simpler to see those things in a 2-Dimensional plane.

Computationally it seems -- from what I've found -- most of the guys with the data are looking at aggregated area views. Not sure if they could get angles or subtleness of players moving to angle for the ball. This is actually where SportsVU data would hurt the view as height matters with passing angles. From the passer to the defenders to the possible receiver.
Crow wrote:Can you get better results by putting players together who think / plan ahead / react in compatible ways and breaking bad patterns or losing disruptions above & beyond what you can do with other stat study and lineup construction approaches?
I would be interested in seeing which players begin to angle themselves better or dive to the basket when their man begins to leave for a double team on the SportsVU side. I'm surprised this hasn't been done more. Toughest part is do players know they are essentially a reverse pivot from being open? Do his teammates know this? I don't see it much with many teams today -- maybe because it's usually a post or a dive to the basket and that shot isn't as valued as much. Or players just don't have the leg strength, body control. LeBron does it often. Barkley, Jordan, Magic, Bird were all great at it too. They could establish position, an angle to receive a pass almost anywhere.

Wolves, the baby Wolves, get to see quick and good decision making, whenever Brandon Rush is in the game. Angles, body control, etc. A lot of this is knowing what your strengths are as a player. Can I shoot this shot with confidence without hesitation as the defense flies to re-assume man-to-man defense after the double OR do I immediately pass it to the next guy who has 5 more feet of close out space required of the defense (and thus more time) to hit the open three? The SportVU might not tell that as easily as you'd likely use that data for more aggregated views/trends vs say Bjelica, LaVine, or Rubio's decision making. I've seen Thibs say 'it', but perhaps in a term or terms they don't fully get. He's said, "If you're open, shoot it!" That's a good shot. He may need to ask them, stop the video. Bjelica -- are you open? His answer, as in the logistics/semantics of it might tell more than the actual answer. Anyone watching the past 10 games or so would say that Shabazz and even Rubio have gotten 'smarter' in this approach and shoot on the catch vs hesitate enough to where all is lost, even the open shot they had.

Thibs has been doing a lot less yelling out of these: ICE!! ICE!! ICE!! (Baby?) and SQUARE!! SQUARE!! SQUARE!! and PICK LEFT!! PICK LEFT!! But the TV cameras seem to be catching a lot more of his other more "colorful" language. If your coach is yelling out everything you should now be doing -- are you thinking? It must be assumed they wouldn't do that.

If I was a basketball coach, I would stop the video many times when it seemed like no one was open and I'd ask: "Who's open?" (Can't be yourself) How? How many 'steps' is that? Who might be more open here? Who has the best mismatch? Each team has 3-5 guys who can probably guard or could be guarding you, tell me your advantages/disadvantages vs all of them. Dribble, drive, shoot, rebound, defend. Then practice your Larry Bird trash talking (cause he knew all of this stuff) -- "Does your coach hate you or something?" "Why?" "Cause he told you to guard me."

I like analytics for narrowing down what can be measured to help identify differences. Then you answer the easier question. At IBM Watson there was really only two areas they had that were good -- text analysis and image analysis. Helping the radiologist to look at 5 similar photos that would help them know what they're looking at was better than 900+. What's the difference between Jordan Mickey (LSU pick by Bos in '15), Ben Wallace, Richaun Holmes (BGSU pick by Phi in '15), and Robert Williams (FR at Tex A&M)? And all those other shot blocking 6'9 'tweeners' who don't become much in the NBA. Maybe it's just the system they were in?

Arizona Coach Sean Miller said Isaiah Thomas (5'9 guard at Washington at the time) was the best player and most difficult player his team faced while Isaiah was in college. He said that while Isaiah was in college. Everyone missed on him, except maybe the Kings, but their giving him away, even after his scoring might suggest they didn't know. How do you 'hit' on these sort of players? 2nd RD picks vs top 5 picks are considerably more easy to acquire.

As far as the technical side...I do have more to offer in that regard -- I haven't put out much on the stats side. Looking for books to help me know which stats tests I need to run to support my model(s) as to not make myself look like any more of an idiot that I am :)
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: TWolves?

Post by Crow »

Yeah I went crow. In case it might be interesting to you. Do / share and not as you wish.

Welcome.
pdevos
Posts: 36
Joined: Fri Jun 26, 2015 6:18 pm
Contact:

Re: TWolves?

Post by pdevos »

Thanks for the welcome. Glad I'll be held accountable. :)
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: TWolves?

Post by Crow »

Towns playing really lousy D according to DRPM (worst mark in league for a Center). Wiggins still below average on RAPM, BPM and WS/48 and worse on RAPM estimate than last season. Dunn abysmal on box score stats and RAPM (by far worst of 3 main PGs). Lots of other issues.

Need multiple trades.
Dr Positivity
Posts: 331
Joined: Thu Sep 20, 2012 6:44 pm

Re: TWolves?

Post by Dr Positivity »

Towns rating below Okafor and Kanter in DRPM is serious
pdevos
Posts: 36
Joined: Fri Jun 26, 2015 6:18 pm
Contact:

Re: TWolves?

Post by pdevos »

Towns has real issues with understanding how much contact he can put on players. He also is upright in his stance when 'on man' defensively.

Some of the issue is when you have horrible defenders like LaVine and Wiggins on the court it has a collateral affect on the team ratings [metrics] as well as morale/effort [shows up in the metrics]. Especially for Thibodeau's schemes which involve a lot of trapping, switching, etc. Where you essentially involve a world class worst defender, then as a big, you try to recover for that player and ultimately get the bill (pts) on your tab at the hoop. Just how it goes.

LeBron isn't as bad a defender as his numbers say. He takes plays off and turns it on when he feels it's necessary (his Drtg is 6 per100 worse than career avg and last year). I'd be a betting man to say his Playoff D-metrics are more in line with his career averages vs a 6 pts per 100 drop it shows for this season.

In the minutes KG was on the floor last year the Timberwolves were among the best defenses in the NBA, when he was off the floor, they were the worst.
http://fansided.com/2017/01/05/nylon-ca ... n-garnett/

But I'm with you Crow, I'd love to see some trades. If the Bulls were willing to trade Butler (by all indications it seems so) for LaVine and Dunn - probably won't now - but LaVine had value. And I imagine some team will pay even more for Andrew Wiggins.

I'd love to see them parlay either of those guys for Trevor Ariza, Patrick Beverley types + draft assets. Both will likely also command $20M+ per beginning in 2018-19 (RFA) as the upcoming 2017-18 is their last year in their rookie deal options. This summer is the beginning of when they can actually sign the extensions... :?

The trick would be getting a draft pick from a team (getting better with Wiggins or LaVine) that owns another team's draft pick so it's not 20+ or later in the 1st RD. Obviously easier than done and probably why teams do pay those crazy $20M+ or $30M+ in the case of Memphis/Conley to keep a player as can't get close to value in assets in return without the trade killing the value of the draft picks they get back in the swap.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: TWolves?

Post by Crow »

Being positive, you could give Thibs some points for patience. But that stops being as much of a positive at some point. Like Feb. 2018 perhaps.
pdevos
Posts: 36
Joined: Fri Jun 26, 2015 6:18 pm
Contact:

Re: TWolves?

Post by pdevos »

Crow,

you a wolves fan by chance?

Yeah. Admittedly, he [Thibs] wasn't my first choice, or second for that matter, for coach of the Timberwolves.

I honestly have no idea if he's patient patient or conservative. He and Layden (GM) have really held their cards tight on all matters. Most interviews are 'by the script' or to it.

At this point the heads of conservative vs patience haven't had a chance to clash. BUT, as you eluded to, Feb 2018 is kinda the deadline for that.

Conservative probably would NOT pay BOTH Wiggins and LaVine $20M+ each per year (have to do it for the Towns the year after too, he could get 5 year max if he makes All-NBA 2x). Patient, might sign them both to those huge deals? Thinking they do improve more than they have this past year. Though....maybe that's passive...

Not doing the Butler trade -- passive, conservative, or patient? I'm still not sure. I probably am thinking patience and wanting to see what LaVine is. Despite Kris Dunn's seemingly weird mental funk on offense, I think Thibs still really likes him. Not hard to see he can be a 5 win defender without any contributions on O.

This Summer will tell a lot. Wolves are one of the few teams with a lot of cap room and flexibility. Will see if any of the big names move and what's leftover for making up rosters after that. =)
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: TWolves?

Post by Crow »

I am not a Wolves fan. I looked at them a few times to see if I could get aboard but little happened this season to encourage me. Unless they make trades I don't see enough to support. Garbage defense irritates me. 18th or 22nd on defensive efficiency, small improvement? I'd grumble but maybe could ride. 27th and 1.5 points worse than last season? Can't stomach.
Crow
Posts: 10565
Joined: Thu Apr 14, 2011 11:10 pm

Re: TWolves?

Post by Crow »

Two biggest minute lineups with LaVine & Wiggins, negative plus minus.
Four biggest minute lineups with just one? All positive, two very strong.

And the insider posture is?
I assume it is... boy we're excited to see them play together again next season.

Yes, sometimes you might fight to try to make something work, if it might be special. But is it? Is it going to be? Based on what team performance facts? Wiggins - LaVine is Wiggins' worst pairing of top 9 pairs. LaVine's second worst. -3.5 pts per 100 possessions. Only slightly better than -4 in previous season. Not working. If it is going to work, it may require a very different approach. Clocking is ticking down.

You could keep both and minimize their time together. Thibs? He increased their minutes together dramatically. Up about 60% from previous season. Will he reverse course or do same? I'm expecting more of same, at least at start of season.
Post Reply