Poll

Should we stop using FG% as a statistic in the current (3-point)-era?

yes
5 (50%)
no
5 (50%)

Total Members Voted: 10

Voting closed: July 12, 2020, 06:09:41 PM

Author Topic: Should we stop using FG% as a statistic in the current (3-point)-era?  (Read 2688 times)

0 Members and 1 Guest are viewing this topic.

Offline RodyTur10

  • Jim Loscutoff
  • **
  • Posts: 2757
  • Tommy Points: 291
  • Always offline from 9pm till 3am
Especially impressive was how much he got to the line, and his clutch FT shooting (19/21).

I think we really ought to retire the FG% stat. In today’s game it just confuses the real picture; Jaylen’s FG% in this series is a prime example. He shot .594 from 2, which is red-hot.
Absolutely, FG%, PER, EWA and probably some others that I've forgotten should be retired. The craze about statistics somehow just being those numbers so we should disregard every statistic in favour of accolades and narratives is simply insane: those statistics are incredibly poor at capturing what they aim to approximate.

This deserved a topic.

When we talk about shooting percentages let's agree to use: 2P%/ 3P%/FT% (or just TS%?), instead of FG%, since for the majority of players FG% doesn't give an accurate picture.

Offline Jvalin

  • Ray Allen
  • ***
  • Posts: 3737
  • Tommy Points: 737
I tend to look for the Effective Field Goal Percentage (eFG%). This statistic adjusts for the fact that a 3-point field goal is worth one more point than a 2-point field goal.

Personally speaking, I prefer eFG% from TS%.

The formula for TS% is the following



where
PTS = Total points scored
FGA = Field Goals Attempted (2pt + 3pt)
FTA = Free Throws Attempted

There are 2 potential problems here:

1.
Quote
The coefficient 0.44 is related to ratio in FTAs from 2 point or 3 point possessions in NBA and and-1 attempts. If all FTAs are from 2 point shooting foul possessions, the coefficient should be 0.5, 0.333 for all 3 point shooting foul possessions, and 0 for all and-1 possessions. For different leagues or in different era, the FTAs ratio is different. It is questionable to use same true shooting percentage formula on those leagues unless the coefficient 0.44 is being adjusted. This raises another question that the true shooting percentage is not comparable for different leagues, age, gender, etc, even for different era.

2. The formula doesn't differentiate between 3-point shots and 2-point shots.
« Last Edit: June 11, 2020, 06:51:05 PM by Jvalin »

Offline gouki88

  • NCE
  • Red Auerbach
  • *******************************
  • Posts: 31552
  • Tommy Points: 3141
  • 2019 & 2021 CS Historical Draft Champion
It’s pretty much useless with TS% and EFG%, alongside the fact that 2P% has been around for a while. Not sure how you go about getting braindead mainstream analysts on the big networks
from dropping it though. Any analysis that requires effort seems to frighten them
'23 Historical Draft: Orlando Magic.

PG: Terry Porter (90-91) / Steve Francis (00-01)
SG: Joe Dumars (92-93) / Jeff Hornacek (91-92) / Jerry Stackhouse (00-01)
SF: Brandon Roy (08-09) / Walter Davis (78-79)
PF: Terry Cummings (84-85) / Paul Millsap (15-16)
C: Chris Webber (00-01) / Ralph Sampson (83-84) / Andrew Bogut (09-10)

Offline cgange20

  • Jordan Walsh
  • Posts: 24
  • Tommy Points: 11
  • "That Ref looks like he started shaving yesterday"
Cmon now. I will give you that the analytics do not tell the whole story, but I do not see how you can totally abolish the whole stat. It still is good to know what a guy shoots.
cartergange

Offline mmmmm

  • NCE
  • Rajon Rondo
  • *****
  • Posts: 5308
  • Tommy Points: 862
I tend to look for the Effective Field Goal Percentage (eFG%). This statistic adjusts for the fact that a 3-point field goal is worth one more point than a 2-point field goal.

Personally speaking, I prefer eFG% from TS%.

The formula for TS% is the following



where
PTS = Total points scored
FGA = Field Goals Attempted (2pt + 3pt)
FTA = Free Throws Attempted

There are 2 potential problems here:

1.
Quote
The coefficient 0.44 is related to ratio in FTAs from 2 point or 3 point possessions in NBA and and-1 attempts. If all FTAs are from 2 point shooting foul possessions, the coefficient should be 0.5, 0.333 for all 3 point shooting foul possessions, and 0 for all and-1 possessions. For different leagues or in different era, the FTAs ratio is different. It is questionable to use same true shooting percentage formula on those leagues unless the coefficient 0.44 is being adjusted. This raises another question that the true shooting percentage is not comparable for different leagues, age, gender, etc, even for different era.

2. The formula doesn't differentiate between 3-point shots and 2-point shots.

The two (eFG & TS) are measuring two very different things.

eFG = effective shooting efficiency from the floor.   I.E., how well the player puts the ball in the basket on field goal attempts, weighted to account for both the increased difficulty and value of 3PT shots.

TS = scoring efficiency.   How efficiently the player converts scoring attempts into points.  This takes into account the ability to get to the FT line, which is a critically important, but completely different skill from shooting efficiency.

I see no reason to "prefer" one over the other.  Both are very, very important measures.  They just tell us very different things about a player.

As to the ".44" number being an approximation, it actually so far still correlates well with actual reality.  It is most certainly at least close enough to not really worry about it.  The expression "FGA+.44*FTA" is simply an approximation of the number of true scoring attempts.   A "true scoring attempt" is either a FGA or an attempt to score the resulted in the player getting to the line (with no FGA charged due to a foul being called).  The .44 coefficient is simple a first-order approximation based on studying the data and is useful since the majority of box scores give you FGA & FTA but do not give actually observed TSA.

I do anticipate now that we have better and better data tracking that at some points most sites will start to report actual  "true scoring attempts" which will convert the TS equation to just

TS = points/TSA

, which is what it is actually conceptually measuring.   Yes, that does not distinguish between 2PT & 3PT shots because it is not measuring skill at shooting either.  It is measuring scoring efficiency.
NBA Officiating - Corrupt?  Incompetent?  Which is worse?  Does it matter?  It sucks.

Offline mmmmm

  • NCE
  • Rajon Rondo
  • *****
  • Posts: 5308
  • Tommy Points: 862
On the topic of whether to "stop using FG%" I would caution that it still is a very useful state because it measures something critical that eFG%, TS% & 3PT% do not:  Miss-rate.

Misses are a critical 'anti-stat'.  A missed FGA is, on average, going to be grabbed by the defense about ~75% of the time.   Not quite as bad as a turnover, but pretty close.  It ends your possession and the other team has a much higher efficiency at scoring on their possession if it starts with a defensive rebound than if they have to in-bound the ball from under their basket.

Team FG% (And defensive FG%) still are highly correlated with winning titles.

So miss rate (i.e., 1.0 - FG%) is something we do need to track.   That said, it is probably much more important as a _team_ stat in this day and age than as a measure of an individual's skill.

For measuring an individual's skills at shooting & scoring, eFG%, TS% & 3PT% are much more useful.
NBA Officiating - Corrupt?  Incompetent?  Which is worse?  Does it matter?  It sucks.

Offline gouki88

  • NCE
  • Red Auerbach
  • *******************************
  • Posts: 31552
  • Tommy Points: 3141
  • 2019 & 2021 CS Historical Draft Champion
On the topic of whether to "stop using FG%" I would caution that it still is a very useful state because it measures something critical that eFG%, TS% & 3PT% do not:  Miss-rate.

Misses are a critical 'anti-stat'.  A missed FGA is, on average, going to be grabbed by the defense about ~75% of the time.   Not quite as bad as a turnover, but pretty close.  It ends your possession and the other team has a much higher efficiency at scoring on their possession if it starts with a defensive rebound than if they have to in-bound the ball from under their basket.

Team FG% (And defensive FG%) still are highly correlated with winning titles.

So miss rate (i.e., 1.0 - FG%) is something we do need to track.   That said, it is probably much more important as a _team_ stat in this day and age than as a measure of an individual's skill.

For measuring an individual's skills at shooting & scoring, eFG%, TS% & 3PT% are much more useful.
Yeah, that is something I definitely had forgotten. about. Good point about it being a pretty important team stat. However, in terms of a measure for individual scoring and shooting efficiency it’s not too great. TP for that
'23 Historical Draft: Orlando Magic.

PG: Terry Porter (90-91) / Steve Francis (00-01)
SG: Joe Dumars (92-93) / Jeff Hornacek (91-92) / Jerry Stackhouse (00-01)
SF: Brandon Roy (08-09) / Walter Davis (78-79)
PF: Terry Cummings (84-85) / Paul Millsap (15-16)
C: Chris Webber (00-01) / Ralph Sampson (83-84) / Andrew Bogut (09-10)

Online Roy H.

  • Forums Manager
  • James Naismith
  • *********************************
  • Posts: 58549
  • Tommy Points: -25636
  • Bo Knows: Joe Don't Know Diddley
Stop using?  No.  But looking at it secondarily to eFG% is a good start.


I'M THE SILVERBACK GORILLA IN THIS MOTHER——— AND DON'T NONE OF YA'LL EVER FORGET IT!@ 34 minutes

Offline Somebody

  • Tiny Archibald
  • *******
  • Posts: 7816
  • Tommy Points: 560
  • STAND FIRM, SAY NO TO VIBE MEN
On the topic of whether to "stop using FG%" I would caution that it still is a very useful stat because it measures something critical that eFG%, TS% & 3PT% do not:  Miss-rate.

Misses are a critical 'anti-stat'.  A missed FGA is, on average, going to be grabbed by the defense about ~75% of the time.   Not quite as bad as a turnover, but pretty close.  It ends your possession and the other team has a much higher efficiency at scoring on their possession if it starts with a defensive rebound than if they have to in-bound the ball from under their basket.

Team FG% (And defensive FG%) still are highly correlated with winning titles.

So miss rate (i.e., 1.0 - FG%) is something we do need to track.   That said, it is probably much more important as a _team_ stat in this day and age than as a measure of an individual's skill.

For measuring an individual's skills at shooting & scoring, eFG%, TS% & 3PT% are much more useful.
Oh yes I'm aware of that use for FG%. The problem with the stat is that you have people still being addicted to the classic shooting splits to measure efficiency for individuals and having no clue that FG% is really only useful as a team stat like what you pointed out. It should absolutely be retired for any evaluation of an individual's scoring efficiency (or shooting efficiency in the case of eFG%).
Jaylen Brown for All-NBA

Offline Somebody

  • Tiny Archibald
  • *******
  • Posts: 7816
  • Tommy Points: 560
  • STAND FIRM, SAY NO TO VIBE MEN
Cmon now. I will give you that the analytics do not tell the whole story, but I do not see how you can totally abolish the whole stat. It still is good to know what a guy shoots.
The whole point of this thread is to try and get rid of the lazy narrative in your post lol, analytics give you a great idea of what's going on when you know what statistics to use.
Jaylen Brown for All-NBA

Re: Should we stop using FG% as a statistic in the current (3-point)-era?
« Reply #10 on: June 12, 2020, 12:06:19 AM »

Offline Somebody

  • Tiny Archibald
  • *******
  • Posts: 7816
  • Tommy Points: 560
  • STAND FIRM, SAY NO TO VIBE MEN
Speaking of retiring stats, I would like to suggest the retirement of stats like PER, EWA, game score or any composite box metric created by John Hollinger :laugh:
Jaylen Brown for All-NBA

Re: Should we stop using FG% as a statistic in the current (3-point)-era?
« Reply #11 on: June 12, 2020, 12:09:22 AM »

Offline Hoopvortex

  • Bill Walton
  • *
  • Posts: 1243
  • Tommy Points: 164

2. The formula doesn't differentiate between 3-point shots and 2-point shots.


The formula accounts for 3 vs. 2 by factoring in points scored, so it doesn't need to specify how many of each were taken. Another way to think about it is that a missed two is the same number of points as a missed three.

I didn't respond to your first point because I don't see anything to disagree with.




'I was proud of Marcus Smart. He did a great job of keeping us together. He might not get credit for this game, but the pace that he played at, and his playcalling, some of the plays that he called were great. We obviously have to rely on him, so I’m definitely looking forward to Marcus leading this team in that role.' - Jaylen Brown, January 2021

Re: Should we stop using FG% as a statistic in the current (3-point)-era?
« Reply #12 on: June 12, 2020, 12:19:13 AM »

Offline Somebody

  • Tiny Archibald
  • *******
  • Posts: 7816
  • Tommy Points: 560
  • STAND FIRM, SAY NO TO VIBE MEN
I tend to look for the Effective Field Goal Percentage (eFG%). This statistic adjusts for the fact that a 3-point field goal is worth one more point than a 2-point field goal.

Personally speaking, I prefer eFG% from TS%.

The formula for TS% is the following



where
PTS = Total points scored
FGA = Field Goals Attempted (2pt + 3pt)
FTA = Free Throws Attempted

There are 2 potential problems here:

1.
Quote
The coefficient 0.44 is related to ratio in FTAs from 2 point or 3 point possessions in NBA and and-1 attempts. If all FTAs are from 2 point shooting foul possessions, the coefficient should be 0.5, 0.333 for all 3 point shooting foul possessions, and 0 for all and-1 possessions. For different leagues or in different era, the FTAs ratio is different. It is questionable to use same true shooting percentage formula on those leagues unless the coefficient 0.44 is being adjusted. This raises another question that the true shooting percentage is not comparable for different leagues, age, gender, etc, even for different era.

2. The formula doesn't differentiate between 3-point shots and 2-point shots.

The two (eFG & TS) are measuring two very different things.

eFG = effective shooting efficiency from the floor.   I.E., how well the player puts the ball in the basket on field goal attempts, weighted to account for both the increased difficulty and value of 3PT shots.

TS = scoring efficiency.   How efficiently the player converts scoring attempts into points.  This takes into account the ability to get to the FT line, which is a critically important, but completely different skill from shooting efficiency.

I see no reason to "prefer" one over the other.  Both are very, very important measures.  They just tell us very different things about a player.

As to the ".44" number being an approximation, it actually so far still correlates well with actual reality.  It is most certainly at least close enough to not really worry about it.  The expression "FGA+.44*FTA" is simply an approximation of the number of true scoring attempts.   A "true scoring attempt" is either a FGA or an attempt to score the resulted in the player getting to the line (with no FGA charged due to a foul being called).  The .44 coefficient is simple a first-order approximation based on studying the data and is useful since the majority of box scores give you FGA & FTA but do not give actually observed TSA.

I do anticipate now that we have better and better data tracking that at some points most sites will start to report actual  "true scoring attempts" which will convert the TS equation to just

TS = points/TSA

, which is what it is actually conceptually measuring.   Yes, that does not distinguish between 2PT & 3PT shots because it is not measuring skill at shooting either.  It is measuring scoring efficiency.
Play-by-play data has been a boon for analytics tbh, it's a pretty big component of modern +/- metrics such as RAPM/RPM. So much for analytics not telling you the whole story when it uses data that literally tracks entire games :laugh:. Obviously those metrics have their issues (eg. collinearity, sample size, etc), but they provide a much fairer approximation of value over a large sample size than most box metrics, ntm they shed light on which players make an impact in the non-box phases of the game (Kevin Garnett says hello).
Jaylen Brown for All-NBA

Re: Should we stop using FG% as a statistic in the current (3-point)-era?
« Reply #13 on: June 12, 2020, 08:23:39 AM »

Offline Vermont Green

  • Ed Macauley
  • ***********
  • Posts: 11225
  • Tommy Points: 860
Kind of like looking at on base percentage or OPS in baseball vs. straight batting average.  With two different kinds of shots (2 point and 3 point) there is a difference and a value in looking at a stat that makes some adjustment for this.  But that doesn't mean you don't need overall FG% at all.  That still is useful or has some relevance just like straight batting average still has some relevance.

As to TS, I think free throw percentage should be kept out of it.  It is something entirely different and should stand on its own.  I don't want to see that mixed in with shooting percentage.

Another stat that I like is to look at is total points per attempted FG.  This is a form of efficiency but captures the effect of made 3 point shots and also both how much you get to the line and the rate that you convert free throws into points.

https://www.foxsports.com/nba/stats?season=2019&category=SCORING&group=1&sort=16&time=0&pos=0&team=0&qual=1&sortOrder=0&opp=0&page=1

Re: Should we stop using FG% as a statistic in the current (3-point)-era?
« Reply #14 on: June 12, 2020, 09:04:15 AM »

Offline nickagneta

  • James Naismith
  • *********************************
  • Posts: 48120
  • Tommy Points: 8794
  • President of Jaylen Brown Fan Club
Kind of like looking at on base percentage or OPS in baseball vs. straight batting average.  With two different kinds of shots (2 point and 3 point) there is a difference and a value in looking at a stat that makes some adjustment for this.  But that doesn't mean you don't need overall FG% at all.  That still is useful or has some relevance just like straight batting average still has some relevance.

As to TS, I think free throw percentage should be kept out of it.  It is something entirely different and should stand on its own.  I don't want to see that mixed in with shooting percentage.

Another stat that I like is to look at is total points per attempted FG.  This is a form of efficiency but captures the effect of made 3 point shots and also both how much you get to the line and the rate that you convert free throws into points.

https://www.foxsports.com/nba/stats?season=2019&category=SCORING&group=1&sort=16&time=0&pos=0&team=0&qual=1&sortOrder=0&opp=0&page=1
TS% and Total Points/FG are extremely similar stats that tell the same story, player scoring efficiency. They are just in different formats.