i have no doubt that harden is going to blossom and kill it in NY under a normal zebra regime now and it will be by playing essentially his same analytics-shaped game he played here. that being said, i had been posting for several years now that we needed an old school bball guy working w/ morey on every decision (this is something i saw DD first propose in this forum). Long Term Capital Management is just one example of quants not being able to oversee analysis entirely on their own. in the end speculation is still an art requiring the human mind's experience, foresight, and judgement
some people’s eye tests told them that there wasn’t that big of a gap between Harden and Kyle Lowry just a few years ago
Numbers don't lie. Morey did concentrate too much on numbers over chemistry though. No room for growth, only known production. Same with MDA. I'm looking forward to taking chances with young players and letting them make mistakes on their way to development. The days of eye test are over though. Analytics is part of basketball whether you like it or not.
Bingo. Once people learn how to be informed, it's hard to take that away. The old, "Well this guy is a lights out shooter!" because you remember a time when he hit 4/5 won't work when people can go see his eFG% and see that he is actually a poor shooter.
Depends on what stats you are using. Many of the ones being used are simple counting stats others are formulas for projecting performance. You can argue about the accuracy of the projecting stats but you can't really argue about the ones based on actual measurements. You mentioned TS% and that's just a counting stat. It's no different than FG% which has been used forever. The reason to use TS% is because it takes into account three point shooting and free throws which straight FG% doesn't. Three pointers and free throws are such a big part of the game, quoting a stat that doesn't account for them is pointless. TS% is basically showing you how many shots a player needs to get their points. It's not a projection, it's actual numbers. Sure you can stick to straight FG%, but then people can claim that a guy shooting 46% on mostly 2 point shots is a better scorer than Steph Curry because he has a higher FG%. If I told someone that John Wall dribbled the ball more per touch than Harden this year then I'm guessing that quite a few people would say "no way". Fortunately there's a stat for that and it would show that Wall indeed took more dribbles per touch than Harden. Those are numbers recorded by cameras in the arena. There's no debating the accuracy of a formula, it's simple counting. The problem with the eye test is that people a) don't see everything and b) only remember certain things. Analytics started in baseball. Through the use of high speed video, what they found out was that several of the fundamentals that had been taught by coaches for decades were actually untrue. Youth players have always been taught that when hitting, always keep you weight on your back foot and "squish the bug". In reality, most every high level hitter actually transfers their weight to their front foot as they strike the ball. Pitchers were taught to lift their leg and reach a "balance point" before moving forward. Now we find that most every successful pitcher's first move is their hips going forward and is one of the major reasons for the huge jump in velocity that we see with today's pitchers. For basketball, the best example that I can think of is Kobe at the end of games. Kobe has a reputation as being the guy that you want to take the final shot in a game. In reality, he's right at league average and certainly not elite at it. He has the reputation because we tend to remember the famous game winners that he's made but not the misses. Here's a great ESPN article that lays out his actual performance at the end of games and discusses why people's perception is so different that what happened. https://www.espn.com/blog/truehoop/post/_/id/24200/the-truth-about-kobe-bryant-in-crunch-time As for all of the small things you mentioned, those actually get included in analytics. NBA teams have many more stats that are team related to determine a players effect on others on the floor. All you have to do is go read the NY Times well publicized "No Stats All Star" article on Shane Battier. It talks about how Battier doesn't put up a lot of standard box score stats but he helps the team greatly through little things and things that you wouldn't pick up with the eye test. It mentions that even though he's not racking up stats, his teammates actually rebound and defend better when he's on the floor. All of those small indirect contributions to the team are brought to light by analytics. https://www.nytimes.com/2009/02/15/magazine/15Battier-t.html Every single team in the NBA uses analytics. That's because it works. That's why teams shoot so many more threes now and so few mid range shots. If you don't agree with any projecting stat such as PER then fine. On the other hand, it would be ridiculous to argue against stats that are direct measurements. I guess it comes down to your priority. Are you trying to state the truth or are you trying to prevent people from calling b.s. on what you said? If it's the latter, then the eye test is the way to go, you can't ever be disproved. If you want to know what is actually happening then you use the actual numbers that are measured.
