About this "new stats era"

[chips93]

TS is arbitrary and doesnt tell the whole story. Simply listing how well they shot from 2,3 and ft is better.

And by arbitrary I mean a made up metric that the creator could change if they wanted to.

what is arbitrary about TS%

it follows very simple logic, and theres nothing subjective about it

All I care about is wins.....

not really concerned how you get them.

knowing how you get them helps you understand how you get some more ...

I'm not saying that we shouldn't look at the basic statistics (FG, TO, ASS, REB, etc), but when people start arguing about X player vs Y because of Win Shares, Stats with Adjusted Pace and other stuff like that it can kills the "eye test" that we all should have.

For me, there's no way that we can mantain a correlation between players that played different brands of basketball (60's vs 00's, for ex.) and it's not worthy providing stats to compare players.

i agree that some stats like PER, that are based on subjective weighting of some stats over others, can be misleading, but what is the argument against pace adjusted stats?

all they do is level the playing field

[Maga_1]

i agree that some stats like PER, that are based on subjective weighting of some stats over others, can be misleading, but what is the argument against pace adjusted stats?

all they do is level the playing field

In my opinion we can't adjust stats acording to pace, but i'm not very familiar with that work so maybe i could be wrong.

I don't think you can adjust Jordan's stats in the early 90's with Durant's stats right now, there is way too many variables in the game.
Basketball evolves, a lot. It's like a new brand of game that comes.

[chips93]

In my opinion we can't adjust stats acording to pace, but i'm not very familiar with that work so maybe i could be wrong.

I don't think you can adjust Jordan's stats in the early 90's with Durant's stats right now, there is way too many variables in the game.
Basketball evolves, a lot. It's like a new brand of game that comes.

no basketball statistician would claim that stats are perfect, but if everybody uses the basic stats ppg, rpg, etc, why not make them more accurate if you can, while not introducing any subjectivity.

obviously you cant adjust for every variable, but when you can adjust for one, and make theses basic stats better reflections of what is going on on the court, then why not?

[Joyner82reload]

In my opinion we can't adjust stats acording to pace, but i'm not very familiar with that work so maybe i could be wrong.

I don't think you can adjust Jordan's stats in the early 90's with Durant's stats right now, there is way too many variables in the game.
Basketball evolves, a lot. It's like a new brand of game that comes.

Of course you can adjust the stats, but there is bias

I.E. if Jordan shot 50% in an era where the average was 48%, then you could assume he was +2% above average. Kobe on the otherhand shot 45% in an era where the average was 45%, 0% above average. So you could extrapolate that Jordan would shoot 47% in today's era, however that's assuming players today are at equal skill levels to players in the past. For all we know, an average NBA player in the 80's might be in the d-league today or a star today might be a role player in the 80's

There are a lot of assumptions made when adjusting stats, and there will be some error. The absolute value of error is impossible to find across era's and we must assume that is insignificant when comparing players.

But it's no different from the eye test. I.E. a player playing against college players will look much better than a player playing against professionals. If you switched a player from todays game and put him in a different era, he could either be better, worse, or the same. When comparing players from different generations by the eye test, we generally assume their competition was the same. It's no different from adjusting stats and assuming the error is miniscule.

[navy]

what is arbitrary about TS%

it follows very simple logic, and theres nothing subjective about it

Explain why they use .44 instead of .45 then.

[game3524]

Advance stats(in general) are useful and have been great for the game IMO. However, PER is complete junk and there is a special place in hell awaiting Hollinger when his time is up on earth.

[TheMarkMadsen]

you are on a website where half the posters think scoring is overrated but that PER, WS/48, Drgt are not overrated

[Maga_1]

The points i'm trying to prove is:
- Stats are good, no matter what. But the way you use them is what can make them important or not.
- People should care of why that stat is giving you that number, not using the number to say that the X player is better than the Y player because of a 0.01% when there is some many variables that can be englobed.
- (This one i'm a little bias) but people can't compare NBA stats with other basketball stats around the world, it's way too different. Believe me.

[Maga_1]

you are on a website where half the posters think scoring is overrated but that PER, WS/48, Drgt are not overrated

Exactly.
I can make a point that the guy in my avy was one of the best passers when he was in NBA purely basing on stats. But if you ask about who he is, 80% would not know him and the other 20% will say that he's a scrub because he's not in the NBA anymore

[BankShot]

An intelligent fan should know the advantages and limitations of both qualitative and quantitative analysis.

If you ask me, anyone that thinks statistics have no place in understanding the outcome in sports probably either hasn't taken the time or doesn't have the capacity to understand what the metrics are attempting to convey.

But then again, I'm sure the average stat-hating dope on ISH has a better handle on observing basketball than those paid by a billion-dollar industry to use advanced metrics to gain an edge......

[The Iron Sheik]

what is arbitrary about TS%

it follows very simple logic, and theres nothing subjective about it

the guy who made the formula decided that free throws are worth 0.44 points or whatever it is.

it's just hard to see the reason for ts%. if it's just an amalgam of fg%, 3p% and ft%...things we already have, then why does it exist? What can it tell me that those 3 percentages don't already?

[Anaximandro1]

Raw numbers are distorted by multiple factors including pace and stat - padding.

- In 1967, the league average was 121.6 possessions per 48 minutes

- In 1983, the league average was 103.1 possessions per 48 minutes

- In 1999, the league average was 88.9 possessions per 48 minutes

- In 2014, the league average is 94.0 possessions per 48 minutes

In other words, points produced per 100 possessions, or the percentage of available rebounds a player grabbed while he was on the floor are better performance indicators than points/rebounds per game.

Context is important. Higher usage rates and more playing time correlate with a decrease in efficiency.

For instance, Tony Parker produces 110 points per 100 possessions while Tiago Splitter produces 117 points per 100 possessions.

However Parker leads the Spurs in usage rate and minutes played per game while Tiago ranks very low in those categories, so the comparison makes no sense. That said, Parker vs Duncan makes sense since their usage rates and minutes played per game are similar.

Moreover, stat-padders are exposed in the playoffs. So, playoff advanced stats are good indicators of success.


Typical statline for superstars, regardless of position

WS/48 0.20

PER 25.0

ORtg 110

USG% 30.0

TS% 55.0

DRtg < 105

MP 40

- Prime Jordan has the best set of results over the last 30 years. No one else comes close.


- 2nd Tier -> prime Duncan/Shaq/Hakeem/Lebron. It's too close to call.

Prime Shaq has the higher PER, LeBron the higher WS/48 and ORtg, Duncan has a phenomenal DRtg ...

Of course, you might want to check rebound, assists and block percentages in order to compare prime Duncan/Shaq/Hakeem.


- 3rd Tier -> prime Barkley/Dirk/Kobe/Robinson/Malone/Wade/Garnett. Their stats appear flawed in one way or another.


It seems to me that advanced stats work pretty well.

[BankShot]

the guy who made the formula decided that free throws are worth 0.44 points or whatever it is.

it's just hard to see the reason for ts%. if it's just an amalgam of fg%, 3p% and ft%...things we already have, then why does it exist? What can it tell me that those 3 percentages don't already?

This is a classic example of someone that doesn't even try to understand what a particular metric (True Shooting %) is trying to convey.

If this poster looked into how the TS% is calculated (or though critically for a second about how statistics are formed) he'd hopefully realize its not

"[some] guy who made the formula decided that free throws are worth 0.44 points or whatever it is"

[ZenMaster]

This is a classic example of someone that doesn't even try to understand what a particular metric (True Shooting %) is trying to convey.

If this poster looked into how the TS% is calculated (or though critically for a second about how statistics are formed) he'd hopefully realize its not

"[some] guy who made the formula decided that free throws are worth 0.44 points or whatever it is"

haha that's funny as hell

[STATUTORY]

no one in real life discusses basketball through those lenses. it's exclusively an online phenomenon