NBA 2K15 Roster Update Details (11-18-14)

[Real2KInsider]

Quote: Originally Posted by MadassMarsh

Allen Crabbe has spent 96% of his entire career at SF.

82games data needs to be taken with a grain of salt rather than face value. They frequently get positions wrong so you need to look at the actual lineup data and production.

The best example of this would be Goran Dragic. 82games lists him at PG in lineups with Bledsoe despite the following

Bledsoe: 6'2", 28 AST%, assisted on 24% of FG
Dragic: 6'4", 16 AST%, assisted on 42% of FG

Dragic is clearly playing off the ball more this year (and also plays far more mins w/ Thomas than Bledsoe d0es). He is defended by SG and defends SG. But 82games ignores all that. Most of their data is fine, just using Crabbe as a small example.

Quote:

Yeah I know, but that's what I'm saying it's for accurate lineups THIS year. Team specific, 2014-2015 specific. Yeah Batum is like 98% SF 2%PF. The thing with Lopez is, that's what a secondary position is. He and Kaman play together so to help the auto subs (even though they're already quite flawed) one of them should have PF as their secondary position. It's not like I'm changing his primary position to PF.

My logic on these situations is the following.

Lopez has played 4% of the team's minutes at PF versus 52% of the team's minutes at C. That 4% is very low and given the large disparity it misleads rather than informs. The only reason Lopez has played PF is because Aldridge was hurt. However with a healthy rotation that will never, ever happen. As such, that 4% will diminish down to 1% by season's end provided Aldridge stays healthy.

If a player plays the vast majority of his minutes at one position and only token minutes at another, then a second position would not make sense because it's giving an opportunity for the AI to put a dumb lineup out there on the regular. Lopez/Kaman is Plan C, not the season-long Plan A.

An easy example that you already referenced.

Dwyane Wade 2014
PG: 2%
SG: 63%
SF: 3%

Wade-Allen-James and Chalmers-Cole-Wade were the only instances of Wade not playing SG in his top TWENTY lineups. That was 57 of his 1,775 total minutes. He's more than capable of playing either position but for rotation purposes he did not except in extreme circumstance.

[Mikelopedia]

Quote: Originally Posted by Rashidi

82games data needs to be taken with a grain of salt rather than face value. They frequently get positions wrong so you need to look at the actual lineup data and production. The best example of this would be Goran Dragic. 82games lists him at PG in lineups with Bledsoe despite the following Bledsoe: 6'2", 28 AST%, assisted on 24% of FG Dragic: 6'4", 16 AST%, assisted on 42% of FG Dragic is clearly playing off the ball more this year (and also plays far more mins w/ Thomas than Bledsoe d0es). He is defended by SG and defends SG. But 82games ignores all that. Most of their data is fine, just using Crabbe as a small example. My logic on these situations is the following. Lopez has played 4% of the team's minutes at PF versus 52% of the team's minutes at C. That 4% is very low and given the large disparity it misleads rather than informs. The only reason Lopez has played PF is because Aldridge was hurt. However with a healthy rotation that will never, ever happen. As such, that 4% will diminish down to 1% by season's end provided Aldridge stays healthy. If a player plays the vast majority of his minutes at one position and only token minutes at another, then a second position would not make sense because it's giving an opportunity for the AI to put a dumb lineup out there on the regular. Lopez/Kaman is Plan C, not the season-long Plan A. An easy example that you already referenced. Dwyane Wade 2014 PG: 2% SG: 63% SF: 3% Wade-Allen-James and Chalmers-Cole-Wade were the only instances of Wade not playing SG in his top TWENTY lineups. That was 57 of his 1,775 total minutes. He's more than capable of playing either position but for rotation purposes he did not except in extreme circumstance.

For the first part, you're absolutely right as I do not use 82games's data. I use basketball-reference and the inhuman amount of League Pass that I watch, as I do not 100% trust ANY position % stats out there.

In regards to Lopez, where you're mistaken is he and Kaman were not on the floor because of Aldridge being hurt. I made my position change list before Aldridge missed his recent games with injury. They were seeing the floor together at times as a planned rotation as Portland was not using anyone consistently as their backup 4.

As far as Dwyane Wade is concerned, I believe you're referencing last season's data (I know you are in the case of referencing a Wade Allen James lineup). I'm not concerned with previous years, only this season. Wade's ratio of SG to SF minutes this year is about 3:1 and he has not spent a single minute at PG this year. Cole-Chalmers-Wade is not the same Cole-Chalmers-Wade of last season.

I do agree with you on a player not requiring a secondary position when their minutes are broken down to only show less than 5% spent at a given position, and I tried to stay away from that as those numbers are scewed by late game situation lineups as well. In those situations I made judgement calls based on what I've seen in actual games and WHEN I've seen it. I know better than to look only at numbers.

[Real2KInsider]

Quote: Originally Posted by MadassMarsh

For the first part, you're absolutely right as I do not use 82games's data. I use basketball-reference and the inhuman amount of League Pass that I watch, as I do not 100% trust ANY position % stats out there.

BBRef can be just as inaccurate, I'm just saying the lineups themselves are needed.

Quote:

In regards to Lopez, where you're mistaken is he and Kaman were not on the floor because of Aldridge being hurt. I made my position change list before Aldridge missed his recent games with injury. They were seeing the floor together at times as a planned rotation as Portland was not using anyone consistently as their backup 4.

From what I can tell in the most recent game w/ Aldridge healthy against the Pelicans Lopez played 27:46 and Kaman played 20:14 (total of 48:00 at center). Beyond the Nets game where Lopez/Leonard started (only possible because the Nets effectively have a double-center lineup), I'm not really seeing any games where Lopez could have gotten extended run at PF as Aldridge has otherwise played a ton of minutes and the Blazers are still giving minutes to the likes of Robinson and Freeland.

Quote:

As far as Dwyane Wade is concerned, I believe you're referencing last season's data (I know you are in the case of referencing a Wade Allen James lineup). I'm not concerned with previous years, only this season. Wade's ratio of SG to SF minutes this year is about 3:1 and he has not spent a single minute at PG this year. Cole-Chalmers-Wade is not the same Cole-Chalmers-Wade of last season.

It was used as an example regarding Lopez's situation, not on anything regarding this season.

[Mikelopedia]

Quote: Originally Posted by Rashidi

BBRef can be just as inaccurate, I'm just saying the lineups themselves are needed.

I know, like I said I don't trust any data on positions 100%. I don't know where to look for individual lineup data, so I've just been taking my own notes on every lineup used.

Quote:

From what I can tell in the most recent game w/ Aldridge healthy against the Pelicans Lopez played 27:46 and Kaman played 20:14 (total of 48:00 at center). Beyond the Nets game where Lopez/Leonard started (only possible because the Nets effectively have a double-center lineup), I'm not really seeing any games where Lopez could have gotten extended run at PF as Aldridge has otherwise played a ton of minutes and the Blazers are still giving minutes to the likes of Robinson and Freeland.

They were using it the first 3 games of the season and most recently the game against the Clippers on nov 8th i think it was. I remember Lopez played 32 minutes and Kaman played 18. In all fairness, considering those 4 games were the examples and its been 10 days since they used it, and Freeland is getting more consistent run, I can probably change that.

Quote:

It was used as an example regarding Lopez's situation, not on anything regarding this season.

Ah, my mistake.

[Real2KInsider]

[quote=MadassMarsh;2046827986]
They were using it the first 3 games of the season and most recently the game against the Clippers on nov 8th i think it was. I remember Lopez played 32 minutes and Kaman played 18.

The Clippers run a double center lineup when Spencer Hawes (11:50) comes in next to DeAndre Jordan (40:17). Injuries and niche matchups have really been the only time it's been done.

Quote:

In all fairness, considering those 4 games were the examples and its been 10 days since they used it, and Freeland is getting more consistent run, I can probably change that.

Freeland was hurt at the start of the season which brings clarity to the decision.

[hdfhfgh]

Everything in today's update is included in the notes, I did remove a few other free agents, but other than that everything is listed[img]

[dytjdy]

Everything in today's update is included in the notes, I did remove a few other free agents, but other than that everything is listed

[Mikelopedia]

Quote: Originally Posted by Rashidi

Quote: Originally Posted by MadassMarsh They were using it the first 3 games of the season and most recently the game against the Clippers on nov 8th i think it was. I remember Lopez played 32 minutes and Kaman played 18. The Clippers run a double center lineup when Spencer Hawes (11:50) comes in next to DeAndre Jordan (40:17). Injuries and niche matchups have really been the only time it's been done. Freeland was hurt at the start of the season which brings clarity to the decision.

In regards to the initial games being a result of Freeland being out, he played about 5 minutes in each of the first 2 games. Thomas Robinson was also available so I saw this as something they would possibly return to. But you're right, there's more evidence against than to support so I'll change it in my next post.

Btw, Rashidi, where's a good source to find lineup data beyond just the best 5 man lineups?