Lakers Plan Staff Move To Beef Up Analytics Operation

Looks like it's time to brush up my resume'

Anyone know if they have a 401k?

 

yes, and maybe. but a good example of a failure would be the three point stuff. the idea is: ignore context, you shoot better on threes than non-dunk twos, so take those only. that's a heuristic. and it works, provided that the threes you're taking are of the same quality as the ones that produced the data in the first place, which, imo, is where the failure is located.

the heart attack stuff is related to my point from before: replace doctor with coach and losing a life with losing a game. so, even when you have something that works, getting it into practice is understandably difficult, imo. the problem is that I don't actually think there's something that works that well in the nba right now in the first place.

we'll see, perhaps some crack analytics team will find it, and their club will break the bulls' win record. so far, we've got morey creating a mediocre offensive team and philly losing like crazy. I just haven't seen the evidence of the competitive advantage, in the draft, front office, or in coaching. the latter is actually the case I'd buy the most, but even then, I'd argue is less about analysis and more about intelligent data collection, which can be done by non-scientists fairly easily.

 

If i was a crappy coach (like Byron) i would surround myself with best assistants and analytics possible.
Real talk.

 

He won't. There's nothing wrong with more information and applying that information in a positive way. The positive I take out of this is we're moving in this direction and Byron may be the odd man out and we could move on...

 

What are the "odds"? I work for a business that provides Quant Analysis to a wide range of client types. We employ Psychologists, Medical Doctors, Mathematicians, Attorneys and "C Suite" individuals from a variety of industries. We do a lot of work developing processes for the healthcare industry as well as manufacturing and the military. Application of analytics is not typically the hurdle, rather identifying and eliminating rouge variables is. In Healthcare this is especially true.

One of my colleagues has been working in his spare time to apply some of our methods to the NFL. He's reviewed and reworked what passes for analytic stats for the last few years trying to find a real application with little success. To your point about finding "it", the number of variables that need to be controlled in a team sport played in constantly changing environments is a significant undertaking.

Where most of these efforts fall short is they are overly player centric. Success needs to be defined as well. Is it fostering individual statistical performance or interdependent success of a team/coaching staff/owner/fan base that is the desired outcome?

 

I get the thought that all the information may not end up translating well on the court, but to just dismiss it outright as Byron has done is just short-sighted.

 

Morey in Houston is the analytic's guru, yet his coach Mchale is a do it by feel coach. Mchale was in the running for coach of the year. I feel he did a great job using his old school coaching, passing on the stat heavy new wave style. They should increase their analytic's team and Byron should pay attention, as a guide, but let him do it his way or fire him, which they ain't gonna do.

 

Predictive analytics needs data to analyze and a determined outcome to predict. If each team is doing this as many are saying I can assure you that there is no "common" method, data set or determined outcome. To have a statistical advantage you would search for data that you believe to be determinant while at the same time crafting the decision product to be relevant in decision making.

Trodgers mentioned the ED RMA process for patients presenting with a potential cardiac condition. The desired outcome wasn't so much to have a "diagnosis" however a "rapid medical assessment" (RMA) that moved a potential cardiac condition into an area where the proper treatment/diagnosis can be done rather than go through a the typical prolonged Triage process. So data typically collected at Triage was moved out of this process because it lacked value in determining the probability of a cardiac event. This is example of "value stream mapping" a process between a data set and decision. Everything not of value in making the decision is discarded. I can see that this a a way of using analytical methods to evaluate player traits, behavior and environmental influence.

 

The Lakers are embracing analytics?

 

Why would anyone willingly watch the Raiders?

NFL stats tell even less than NBA ones about the true value of a player. Typically 16 to 19 guys make up a typical "possession" on each side of the ball. The so called "analytical" stats for the NFL are really laughable. Made for "Fantasy" leagues more than reality. It's not much different with the NBA advanced stats.

 

bonk's player-centric complaint is a good one. I think it's tied to the correlation-causation issue that plagues a majority of the media who love the advanced stats. coaches like popovich and Carlisle do a good job finding roles that highlight players' strengths. this is why you see reclamation projects there (and, conversely, guys who have less success outside of the system). it's also why you see them keep some guys' minutes low despite great production in their court time, much to fans' chagrin. these guys are good and contribute to winning *when they are playing in an appropriate role*. to me, the beginning of the causal chain, as it matters to our desired outcome (winning) is appropriate employment of the player. but we could go round and round on that, I suppose.

going back to the three point discussion: I think teams that can generate a lot of good three point shots are good teams, but that doesn't mean that just going out and shooting the three makes you a good offensive team (again, see Houston). the combination of players and system that leads to the right guys getting the right looks is what's driving the relation. or, in gs's case, having two guys for whom any look is a good look playing heavy minutes.

if Byron scott told his team to just stop taking midrange shots, for example, the lakers offense could possibly get worse. seriously.

again, not trying to support scott as a tactician.

 

Kings just fired Dean Oliver. Time to hire him now, Buss family!