Amen Thompson is the Rockets' REAL franchise player

It's not just his 3 ball, it's literally any non layup shot taken in the flow of the game. He's not a ****ing max player with this CBA, idgaf what r****ded metric people throw out, that's team suicide.

 

Dean Oliver invented TS% and compares it to efg% as:

“True shooting percentage provides a measure of total efficiency in scoring attempts, while effective field goal percentage isolates a player’s shooting efficiency from the field.”​

So, imo, best way to understand the TS% ratio is to first understand what eFG% shows us.

eFG% can mathematically be rewritten as

eFG% = FG points / (2 * FGA)
​

with PBP info, the TS% ratio can be expressed as

TS% = Total points / (2 * shots)​

Doesn’t that make it clearer?!

try it. That above eFG% is exactly equal to (fgm + .5 * 3s) / FGA

The 2 is only there to make the % visually easy to compare to FG%. Albeit, perfect = 150% not 100%

Thus, ts% adds foul points to eFG% to get overal shooting efficiency. Read spoiler for why a really true TS% should be (Total points from scoring attempts) only — per Oliver’s wording, not total points. But the ratio above is the identical numenerator for the TS% formula that has the (2 * (FGA + .44 * FTA)) denominator, whereby PBP eliminates the need for the .44.

Spoiler: How .44 estimator was determined

Per Oliver’s wording, as a measure “total efficiency from scoring attempts) ts% is arguably:

(Points from shots) / (2 * shots)

this would eliminate points not from scoring attempts — ie subtract points from technical fts and non-shooting fouls fts

This wording comes from papers written about the intent of ts% and how the .44 estimator was measured.

the .44 estimator is defined by Oliver in papers as an empirical estimate of “possession-ending free throws.” So, it could be said “total points” in the numerator of TS% shouldn’t count technical ft points, since technical fouls shots don’t usually end possessions. Regardless the intent of TS% is about shooting, thus “end possession free throws” is merely a technical way of measuring fts from shooting fouls. That’s where he got .44.

 

It will never be done, bc pbp didn’t exist prior to ~1996. So they need the estimator to compare across eras.

keep in mind though, Dean Oliver came up with .44, by using actual pbp with years worth of data. So, that tells you, he would not have calculated the .44 if pbp had always existed. We’d just scrape pbp per game

 

Thanks for the explanation. Just to be clear, what does "shot" mean? FGA + FTA?

 

Yeah you got it, maybe think of it as "total scoring attempts" as an expanded way of saying "shots." eFG% obviously leaves out the FTA's

 

Shots = scoring attempts. You can’t express scoring attempts from the boxscore, only pbp.

so, it would be FGA + FTAs from shooting fouls.

again, I’m using Oliver’s definition of TS% being about “scoring attempts”, thus “shots” is just my shorthand word.

 

FWIW, I'm much less "against" things like TS%/"advances STATS" than I am against "catch all metrics." I just think people use it for the wrong thing. TS% heavily skews towards low volume snipers and big men, it's trash for evaluating overall scorers outside of the crazy anomaly players like Harden. (That's why if you compare people in the same era Reggie Miller is MASSIVELY higher in TS% compared to Michael Jordan, but nowhere near the scorer MJ is - volume scorers tend to look pretty bad in TS% as a general trend.) Catch all metrics are all subjective to the author, they're so easy to use to push narratives, not great for accurate portrayal of reality. In truth, the best scorers score the most points. The best shooters have the highest percentages - choose the correct stat for the attribute you're recruiting for. Where the metrics/advanced stats REALLY shine is in the money ball (or Morey-ball pre-Harden years) situation where you're trying to put together a team either on a low budget, or in a less attractive destination (surely people remember the crapping out we did in FA year after year after year?) where you need to use budget talent to a greater effect. If you use those types of stats to find statistical anomalies where one player does a unique thing significantly above league average, and then build around those crazy niche skills you can over-succeed based on the "on paper" talent and actually get great results. If you overdo it though you can hamper the successes and make it easy to counter because you're so min/maxed that just taking the niche talent away removes the edge (see: later years Harden era when all GS had to do was tire Harden out and the refs swallow their whistle and you've taken away the one thing that enabled us to truly compete.)

But yeah, I'm the casual (literally watch almost every game, and typical tip off time is 1am for me, just saying) because I just don't swallow the metric bullshit wholesale like the rest of y'all. Cool beans.

 

And I believe Morey also strongly advocated boxscore metrics for drafting. Your scouting team isn’t big enough to watch everyone, so it helps to narrow the list down.

This need for scout-based player evaluation also sheds light on why NBA teams have far better stats than are available to the public. Hence Stone scoffing at Jackson Gatlin’s question about player Net ratings (even despite them being based upon pbp).

Teams have years of video data of NBA games. Plus they have scouts and coaches analyzing the video data who actually know when a player made a mistake vs what they were supposed to do. Teams are training computer analysis to learn to spot their specific play flow and offensive/defensive assignments. So, their “advanced metrics” can be precisely tailored to real game situations generating stats that can truly measure effectiveness vs what’s supposed to happen. Not just end of possession “play types” like current nba.com stats, but stuff like did initial actions that were denied successfully flow into secondary options. (e.g., did a snuffed PnR convert to secondary action, did our Princeton split action denied successfully covert to weakside DHO, etc). Plus minute measurable stuff like did he cut, screen, pass, shoot properly or did he disrupt flow, causing a broken possession.

that’s why I say, eventually, the public will have access to video metrics that will put to bed current adv metrics … albeit they will likely be generalized. As an example, to your point about TS% favoring certain players, right now the video-based “Gravity” stat is flawed. It’s really about gravity for perimeter players, I can demonstrate that with current rankings.

 

So it's FGA + FTA - technical foul shots. I assume that it includes and-1 FTs.

I often glance at points/FGA to get a rough notion on a player's efficiency after a game.

 

“Shooting foul” is a real term in the officiating rules.

I’m not of fan of describing it based on what it isn’t. But, if you insist , FTs from “non-shooting fouls” aren’t from shot attempts either.

 

Oh, I see. I forgot the over-the-limit fouls that result in free throws but not from scoring attempts.

 

Yeah everything you just said lines up with my understanding of how analytics/metrics would be used in a front office vs the "catch alls" which seem a lot more tailored to pundits/fans to discuss in a less recruitment focused way. It makes total sense that they have their own scouting metrics tailored/keyed in to the specific desired traits they're targeting to fill team needs. I'd imagine Morey's own metrics/benchmarks are mich more effective at finding what he's looking for than the fad ones we see all over the net which definitely seem more aimed at pushing the authors narrative on "what makes a player good?"

 

Yeah the fixed-position Synergy cameras was a game-changer for NBA teams, including consulting companies paid to tailor their video analytics. We do see X accounts that are clearly paying the big bucks for Sportsradar access, but even that is just generalized analytics for popular consumption and engagement.

 

These stat clusters always feel kind of random -- like I picture a guy using sliders on various stats until he gets an impressive peer group.
But, that said, that's a damn impressive season and nice company.

 

Yeah let's pull up all the hundreds of those posts that came out for Jalen Green where it was him, KD and LeBron etc lol? They're genuinely worthless bullshit made to pump up the fanbase.

 

and still has 50 games before he turns 24

 

I get your point and completely agree that a max contract is a particularly precious resource to manage in the new CBA. You really can't stay competitive very long in the newest CBA by outspending. Completely agree on the point.

Where I'm coming from is - a player like Amen will cost close to if not the full max because DPOY candidates with less usage and less efficiency are already getting those types of paychecks in the current financial climate. Those numbers I was pointing to above represented two season of Age 23 Kawhi Leonard(Player A and C) while he was an MVP and DPOY candidate and a Scottie Barnes' All Star season(Player D). Whether you think he will one day be as impactful as Kawhi is irrelevant - the fact is, the statistical case for Amen's numbers(Player B) compared to age 23 Kawhi and Scottie is real and there WILL be teams betting on that case just for the chance of being right.

If It were my choice - YES, we are giving a max contract to someone who can hit 3s at least at 35% clip and clearly Amen is not that guy but I just don't know what our options are as a team if you aren't giving Amen a max deal...I do know you 100% can't risk losing him for nothing so you have 3 options on the risk/reward scale:
- match whatever a team offers - which could be the max and Amen is our new max guy. Maybe there is a 10% chance he turns out to be young Kawhi/complete player, a 3e0% chance he ends up being a 'Giannis level limited but efficient despite that' type of player, and a 60% chance he becomes 'a walking triple double on poor efficiency volume gunner like Westbrook(but with better defensive impact)'.
- trade Amen for a 100% no arguments max guy - but it would not likely be a 'not in their prime' max guy because those guys are rarely available - Giannis is the obvious one here - so what, 2ish years of remaining prime Giannis of which you ALSO have to rebuild a team that allow you to pace and space with him(bye, Alpi)...so you 100% get "the guy" but you also need to trade for his brothers AND you also need to surround him with a ton of shooters(which is this team's achilles heel already) - basically you are gutting the team to optimize for Giannis but you are doing it under the timeline of the end of his era as a do it all player. There is probably a 95%+ chance we don't win a title with him and we are again a lotto team in 5 years.
- OR trade Amen for picks that allow you a shot at drafting a 100% max guy 4 years from now. This really depends on which pick you get - I would think teams willing to talk about a trade would not be the ones drafting a Wemby/Flagg type prospect so you probably have a 5-10% chance of picking the right player(who grows into the value 4+ years from now) in that scenario because the generational prospects won't even be available to you.

Bottom line I think the risk of losing Amen is far more damaging to the team than signing him to a max deal and regretting the decision. He is at very least a very effective player and is a likely upgrade for most teams out there so you would think he could be tradeable while at the peak of his athleticism even if we did sign him to a max deal and struggled to find a legit 3 level scorer to pair him with. I think the odds are much better that you keep Amen and you try to find value trading Alpi and Tari for someone else. We do need one of our max guys to be a 3 level scorer - that is true. Amen is not that guy but then again neither is Alpi and Alpi's defense can be a liability at times for what? A bunch of two pointers? Amen gets the line more and makes them at a higher clip so where is the value in Alpi's near max contract for us?

 

He's not a DPOY candidate though, I'll be surprised if he gets top 5 votes, definitely not top 3. Great defender, not league beating at all though. I never said I wanted Giannis so I'm not wasting my time reading all that paragraph tbh. The problem is y'all are glazing him to the point it's completely skewing your perception. He may make all-defensive teams if he's lucky, he's not the best defender in the league and he's unlikely to receive that accolade compared to his contemporaries unless there's a lot of injuries. He's also absolute trash on offensive fundamentals. It's a bad move paying him that much more money. If you have to give a max contract out, give it to somebody who can put the team on their back in scoring, if not, let some other team make the idiotic mistake, or accept that you're sacrificing your success chances to appease a completely flawed player.