Davis Bertans "the Latvian Laser"

[nate33]

This looks like a good time to repost this article:

https://fansided.com/2019/01/08/nylon-calculus-best-advanced-stat/

[doclinkin]

Ok, for starters, let's not argue. It's pointless.

...

There are plenty of subjects about which you know something & I don't, doc. Similarly, there are subjects about which I know something & you don't. Neither kind of subject presents a fruitful area for discussion.

Whereas I didn't think I was arguing at all. Simply providing context to the conversation for people who don't know how the numbers are produced. That said, I have been following these models as they were discussed on:

http://apbr.org/metrics/viewforum.php?f=2

Where people like Kevin Pelton and all (including our own WizardsKev Kevin Broom) were debating David Berri's formulas etc. that you had an affinity for. Before they all went on to get big money jobs with various organizations. (Okay except our Kev). So I have seen the basis of some of their calculations as they were debating them.

We've been over this before on another website, linked you to the boards, but likely you forgot. Is all good, time happens to all of us. (Well, we can only hope...)

But you can run regressions yourself at least to adjust for teammates, with the limited data at 82games.com and their line-up info. Or now at Baskeball reference, where they actually do the regressions one by one. It is all based on box score info.

Even some box score info is arbitrary though. (Assists for instance. It suggests something about passing but isn't really definitive, measures nothing really). Still Plus/minus is a pretty intuitive concept, not all that arcane, and based on that box score data and not the advanced video situational data we don't have access to as easily. The question is: does the squad score more or less with a guy on court? Does this players effect carry regardless of personnel? Just because you yourself haven't programmed the algorithm to calculate it with 5 on 5 squads, doesn't mean you can poohpooh the work of someone who actually has put in the effort. Or dismiss the concepts behind their work simply because you don't have the rigor to recreate it. I don't either, which was why I found it fun simply watching from the sidelines at APBR while the deep nerds do their nerding. (I think it is pretty empty there nowadays, and has been since people started actually making money off this thing. But back in the wild west days there were a lot of ideas being freely exchanged, by many of the brightest minds in the business).

On one point though. No:

... RAPM uses a rating of 0 as its prior–in other words, it’s skeptical of a player who rates as strongly above or below average, unless it has a lot of data to back that up. Using ridge regression reduces the impact of the small sample size problems I described above. Values that stray too far from the prior, especially without lots of data to back them up, will be pulled back into (hopefully) a more reasonable estimate. ... While randomness can still have an effect, the damage is less than it is for APM...

Doc, have you taken the time to grok what the above sentences are actually saying? What they are saying is, "we already know how good players are, & if the actual data makes them look different from that... we just change the data."

What I understood this to mean, here and when this was hotly debated years ago, was that the ridge regression attempts to reduce all stats back to the average. If two teams of players were perfectly equal, there would be no variance in +/-. So: zero plus/ zero minus. That means that only the serious and significant and sustained variances will defeat the ridge regression. And from that you can detect the outliers that significantly stray from the average. It reduces the peaks and valleys so the best and worst players are clearly showing out above (or below) the rest.

[doclinkin]

FIBA World Cup qualifiers vs Romania

Watch on YouTube

[Kanyewest]

FIBA World Cup qualifiers vs Romania

Watch on YouTube

Spoiler:

]21-0 game to start the game for Lativia. Bertans hit around 50% of his 3s including drawing a four point play and getting fouled on another on another 3 point attempt. He did miss 3 free throws but overall pretty solid game although doesn't seem like Romania is the toughest of the competition.

[nate33]

I really wish Bertans would scrap the vegetarian diet and start eating some protein.

[closg00]

Our Davis made the list amongst the John Wall and Westbrooks of the NBA

Watch on YouTube

[FAH1223]

Totally fine with the deal.

Still need a defensive wing and big. But I’m not holding my hopes up.

Read on Twitter

?s=21

**** you FAH1223. You were an idiot and wrong.

[payitforward]

Wrong is something we all are pretty often.

Congratulations on checking in here & calling yourself on it. I wonder how many others will follow suit?

[9 and 20]

How has the title of this not been changed to 'latvian loser' yet???

[WallToWall]

Need a stat… (yearly salary) / (number of 3ptrs made). Gotta make sure he is earning his keep.

[Chocolate City Jordanaire]

What I think coach wuj is doing completely wrong is that he has a set amount of people that he is going to play a certain amount of minutes and it just doesn't work like that

Davis bertans was shooting very quickly and playing very poorly but the way this guy's getting yo-yoed in and out of the lineup doesn't make a bit of sense to me

Spencer Dinwiddie has had some rough shooting himself

You cannot play all the people in please all of the people but you certainly can canvass the opinions and mix it up enough to where guys are engaged without getting totally destroyed

To me there is no reason why you cannot play 15 guys a game occasionally and play at least a nine-man rotation

[nate33]

What I think coach wuj is doing completely wrong is that he has a set amount of people that he is going to play a certain amount of minutes and it just doesn't work like that

Davis bertans was shooting very quickly and playing very poorly but the way this guy's getting yo-yoed in and out of the lineup doesn't make a bit of sense to me

Spencer Dinwiddie has had some rough shooting himself

You cannot play all the people in please all of the people but you certainly can canvass the opinions and mix it up enough to where guys are engaged without getting totally destroyed

To me there is no reason why you cannot play 15 guys a game occasionally and play at least a nine-man rotation

I think you are saying two conflicting things here.

If you have too many players to fit into a rotation as we do, you have two choices: bench some of them permanently and ride the others, or mix and match and play the hot hand, which means some players would play some nights and not play on other nights, and they'd never know what to expect. WUJ has gone with the first option. You appear to be advocating for the second option while simultaneously lamenting that players are getting "yo-yoed".

Benching Bertans is the solution to avoid yo-yoing Avdija, Kispert and Hachimura. I agree that it kinda sucks for Bertans, but I care more about the development of Avdija, Kispert and Hachimura.

He is doing the same thing right now by benching Gafford. This is a much tougher call because we want to develop both Gafford and Bryant, but Harrell has been too good to bench. I figure WUJ is just trying to give Bryant an extended stretch to find a rhythm before making a final decision on who to play. It sucks for Gafford, but I can understand what he is doing. Bryant is a young free agent and we really need to know how effective he can be post injury in order to make a decision on him this summer (or at the Trade Deadline). Gafford is locked up long term so a few weeks of short term frustration won't really matter in the long run.

[9 and 20]

I don't understand what Wes is doing. They are trying to make the playoffs, aren't they? They can't wait to mortgage the future in order to have a shot at that ten seed, but they still are forward thinking enough to spend some time trying to see what they have in Bryant? It doesn't make sense.

Gluing Bert to the bench is much more about performance. He's been garbage for the better part of two years now. Half of that time, he's been out of shape.

[Chocolate City Jordanaire]

Latvian Lats man
Latvian Lover
Latvian Looser because it's fun to misspell loser

LATVIAN lap band... oops I'm the fat bastard... my bad

[Chocolate City Jordanaire]

What I think coach wuj is doing completely wrong is that he has a set amount of people that he is going to play a certain amount of minutes and it just doesn't work like that

Davis bertans was shooting very quickly and playing very poorly but the way this guy's getting yo-yoed in and out of the lineup doesn't make a bit of sense to me

Spencer Dinwiddie has had some rough shooting himself

You cannot play all the people in please all of the people but you certainly can canvass the opinions and mix it up enough to where guys are engaged without getting totally destroyed

To me there is no reason why you cannot play 15 guys a game occasionally and play at least a nine-man rotation

I think you are saying two conflicting things here.

If you have too many players to fit into a rotation as we do, you have two choices: bench some of them permanently and ride the others, or mix and match and play the hot hand, which means some players would play some nights and not play on other nights, and they'd never know what to expect. WUJ has gone with the first option. You appear to be advocating for the second option while simultaneously lamenting that players are getting "yo-yoed".

Benching Bertans is the solution to avoid yo-yoing Avdija, Kispert and Hachimura. I agree that it kinda sucks for Bertans, but I care more about the development of Avdija, Kispert and Hachimura.

He is doing the same thing right now by benching Gafford. This is a much tougher call because we want to develop both Gafford and Bryant, but Harrell has been too good to bench. I figure WUJ is just trying to give Bryant an extended stretch to find a rhythm before making a final decision on who to play. It sucks for Gafford, but I can understand what he is doing. Bryant is a young free agent and we really need to know how effective he can be post injury in order to make a decision on him this summer (or at the Trade Deadline). Gafford is locked up long term so a few weeks of short term frustration won't really matter in the long run.

REALLY???

Only two?

I thought I was **** sleep deprived and not making much sense AT ALL

[Chocolate City Jordanaire]

nate, thanks for taking the time.

I'm MYSTICfied ... no,,, mystified about this coach this team ... NADA clue but I THINK the losses are MUCH MORE on WUJ than Bertans or any player INCLUDING Beal and Dinwiddie.