Advanced Statistics for Dummies: True Shooting Percentage
Jul 11, 2013 14:25:24 GMT -6
London likes this
Post by Bayi on Jul 11, 2013 14:25:24 GMT -6
Most of us are familiar with the game of basketball. You shoot the ball. You make baskets. You score points. One side of the ball is dedicated to putting the ball IN the basket, while the other side is dedicated to keeping the ball OUT of the basket.
Over our sim careers, this is as much as we needed to know. Percentages and ratings were important, but we focused on raw numbers - not of our own accord, but because this is all Fast Break Basketball gave us.
Now? Now we're entering into a new era. Now we have access to advanced stats. Now we can get all Sloan up on our teams and our trades. And now we need to figure out what exactly they mean.
This is the first in a series of articles I'm going to put together looking deep into the world of FBB3's advanced statistics. What do these stats mean? How can we use them? What can an advanced stat noob like me take away from those extra percentages and weird rates? The articles are part research into real-life use, and they are part look at the league's current players.
(Note: WITH ALL OF THESE I'm dealing in vast generalities.)
The Schedule
July 11 - True Shooting Percentage (TS%)
July 16 - Assist and Turnover Rate (AstR/ToR)
July 18 - Usage (Usg)
July 23 - Rebounding Rate (ORR/DRR/RR)
July 25 - Value Added and Estimated Wins Added (VA/EWA)
July 30 - Player Efficiency Rating (PER)
---
True Shooting Percentage
According to Wikipedia, true shooting percentage measures a player's efficiency at shooting the ball. The formula is PTS/(2*[FGA+.044*FTA])
We're all used to using raw percentages to gauge a shooter. For example, field goal percentage is the percentage of baskets a person makes, not counting free throws. If a player makes 42% of his baskets, he's a really bad shooter. A jump shooting guard who can make 50% or more of his shots might become the stuff of legends.
Yet, field goal percentage does not accurately show a player's ability to score. For example, 42% is a GREAT percentage for making three-point shots. Alternatively, it would qualify as one of the worst in league history for free throw shooting. The fact that the three important percentages are not parallel makes it difficult to gauge the value of a player's ability to score. Is a three-point shooter who makes 42% of his shots more valuable than a center who shoots 60% in the paint but only makes 57% of his free throws?
Enter true shooting percentage, which takes away the variable of shot selection and places the focus on efficiency in scoring points. Remember: teams don't win games because they shoot better - they win games because they score more points.
In other words, true shooting percentage allows us to more accurately compare a guard who shoots 45% from the floor but draws 8 fouls a game (and makes seven) to a center who shoots 10-foot jump shots but can't manage to make his free throws. It's not perfect - it never will be - but it's better than what we had.
Context
The difficulty in advanced stats is that we often don't understand them within the context of the league or corresponding talent. What's a good TS% look like? Does it matter from position to position? Can I get away with a player who has a weak TS%, or is it a fatal flaw in a player's talent level?
The answer, as always, is "it depends." Let's start by looking at current real NBA players and how true shooting percentage correlates with talent.
2012-2013 Season
1. 67.0% - Tyson Chandler
2. 64.6% - Kevin Durant
3. 64.0% - Lebron James
4. 63.7% - Kyle Korver
5. 61.6% - Jose Calderon
6. 61.2% - Serge Ibaka
7. 60.8% - Taigo Splitter
8. 60.7% - Kevin Martin
9. 60.5% - Deandre Jordan
10. 60.55 - Carl Landry
The top ten is a combination of players who are defense-oriented centers (scoring mostly on putbacks or assisted layups/dunks), sharp-shooting snipers, or legitimate superstars, the exception being Kevin Martin, who makes it onto the list because he draws fouls at a rate higher than one might expect, and Carl Landry who makes it in based on who knows what.
Not a lot of scorers on this list, really, and that's by design. High scorers take more shots. More shots means less efficiency. There's a reason the only two scorers on this list happen to be the two best players in the league - and there's a reason those two are considered the two best in the game.
But that doesn't answer our question. What makes a good TS%?
We can look at it one way: what does it take to be top 10 TS% in the NBA? Over the past fifteen years, the #10 position in TS% has been somewhere between 57.8% (Brad Miller, 2004-2005) and 61.3% (twice: Jose Caleron, 2008-09, and Mikki Moore, 2007-08). Give or take a year here and there, you need a 60% TS% to be in the top 10.
According to an article from ESPN Fantasy Basketball's John Cregan, he considers players for his fantasy teams based on the following simplified analysis: "A .500 TS% is about average, .550 is very good, and anything over .600 is exceptional."
If 60% is good, then what is GREAT? Looking at league history, the league leader for the past 15 years has been typically somewhere around 62-67%. Tyson Chandler had a 70.8% last season (2011-12) - a league record - and he holds two of the top five single season TS% spots, along with Artis Gilmore (two) and Wilt Chamberlain (one). The top 25 single-season TS% seasons are all above 65.7%, so we'll use that as our benchmark.
A note on using TS% solely to identify a great player: number seven on the all-time TS% single season list is Tim Legler, which is evidence that there's not a direct correlation between TS% and superstardom. What we DO know that it can be valuable in finding an efficient player to stick into the rotation.
In Our League
So, who's shooting well in our league?
It's early in the season, so our season stats are going to be skewed. As an example, Mexico City's Andy Johnson leads the league in TS% with 78.7%.
But we do have a full season to work with already, so instead of looking at this season alone, we're going to look at "career" TS%, which in this case means all of last season plus these last three sims. What we see is a similar pattern to what the NBA exhibits:
Top 10 Career TS% (Minimum 60 total games)
1. 64.2% - Bill Russell (C, Cairo)
2. 63.1% - Bennie Borgmann (SG, Madrid)
3. 61.4% - Gene Conley (PF, Beijing)
4. 61.1% - Ray Felix (C, Brazil)
5. 60.9% - Wilt Chamberlain (C, London)
6. 60.1% - Andy Phillip (SG, Tel Aviv)
7. 59.8% - Janus Krumins (C, Sydney)
8. 59.4% - Dolph Schayes (C, New York)
9. 59.3% - Neil Johnston (PF, Chicago
10. 59.3% - B. McCracken (SF, Berlin)
Lots of big men, though in our case a lot of our big men are higher usage players - with Borgmann, Chamberlain and Schayes all in the top 15 in team usage (which we will cover next week.) Still, it's the same pattern. Big men and sharpshooters.
In Closing
True Shooting Percentage - like the traditional shooting percentage categories - does not dictate a specific level of talent overall. However, a good TS% can help determine which players are most efficient, which comes into play both in creating super-duper-stars, big man role players and in finding great bench pieces.
Over our sim careers, this is as much as we needed to know. Percentages and ratings were important, but we focused on raw numbers - not of our own accord, but because this is all Fast Break Basketball gave us.
Now? Now we're entering into a new era. Now we have access to advanced stats. Now we can get all Sloan up on our teams and our trades. And now we need to figure out what exactly they mean.
This is the first in a series of articles I'm going to put together looking deep into the world of FBB3's advanced statistics. What do these stats mean? How can we use them? What can an advanced stat noob like me take away from those extra percentages and weird rates? The articles are part research into real-life use, and they are part look at the league's current players.
(Note: WITH ALL OF THESE I'm dealing in vast generalities.)
The Schedule
July 11 - True Shooting Percentage (TS%)
July 16 - Assist and Turnover Rate (AstR/ToR)
July 18 - Usage (Usg)
July 23 - Rebounding Rate (ORR/DRR/RR)
July 25 - Value Added and Estimated Wins Added (VA/EWA)
July 30 - Player Efficiency Rating (PER)
---
True Shooting Percentage
According to Wikipedia, true shooting percentage measures a player's efficiency at shooting the ball. The formula is PTS/(2*[FGA+.044*FTA])
We're all used to using raw percentages to gauge a shooter. For example, field goal percentage is the percentage of baskets a person makes, not counting free throws. If a player makes 42% of his baskets, he's a really bad shooter. A jump shooting guard who can make 50% or more of his shots might become the stuff of legends.
Yet, field goal percentage does not accurately show a player's ability to score. For example, 42% is a GREAT percentage for making three-point shots. Alternatively, it would qualify as one of the worst in league history for free throw shooting. The fact that the three important percentages are not parallel makes it difficult to gauge the value of a player's ability to score. Is a three-point shooter who makes 42% of his shots more valuable than a center who shoots 60% in the paint but only makes 57% of his free throws?
Enter true shooting percentage, which takes away the variable of shot selection and places the focus on efficiency in scoring points. Remember: teams don't win games because they shoot better - they win games because they score more points.
In other words, true shooting percentage allows us to more accurately compare a guard who shoots 45% from the floor but draws 8 fouls a game (and makes seven) to a center who shoots 10-foot jump shots but can't manage to make his free throws. It's not perfect - it never will be - but it's better than what we had.
Context
The difficulty in advanced stats is that we often don't understand them within the context of the league or corresponding talent. What's a good TS% look like? Does it matter from position to position? Can I get away with a player who has a weak TS%, or is it a fatal flaw in a player's talent level?
The answer, as always, is "it depends." Let's start by looking at current real NBA players and how true shooting percentage correlates with talent.
2012-2013 Season
1. 67.0% - Tyson Chandler
2. 64.6% - Kevin Durant
3. 64.0% - Lebron James
4. 63.7% - Kyle Korver
5. 61.6% - Jose Calderon
6. 61.2% - Serge Ibaka
7. 60.8% - Taigo Splitter
8. 60.7% - Kevin Martin
9. 60.5% - Deandre Jordan
10. 60.55 - Carl Landry
The top ten is a combination of players who are defense-oriented centers (scoring mostly on putbacks or assisted layups/dunks), sharp-shooting snipers, or legitimate superstars, the exception being Kevin Martin, who makes it onto the list because he draws fouls at a rate higher than one might expect, and Carl Landry who makes it in based on who knows what.
Not a lot of scorers on this list, really, and that's by design. High scorers take more shots. More shots means less efficiency. There's a reason the only two scorers on this list happen to be the two best players in the league - and there's a reason those two are considered the two best in the game.
But that doesn't answer our question. What makes a good TS%?
We can look at it one way: what does it take to be top 10 TS% in the NBA? Over the past fifteen years, the #10 position in TS% has been somewhere between 57.8% (Brad Miller, 2004-2005) and 61.3% (twice: Jose Caleron, 2008-09, and Mikki Moore, 2007-08). Give or take a year here and there, you need a 60% TS% to be in the top 10.
According to an article from ESPN Fantasy Basketball's John Cregan, he considers players for his fantasy teams based on the following simplified analysis: "A .500 TS% is about average, .550 is very good, and anything over .600 is exceptional."
If 60% is good, then what is GREAT? Looking at league history, the league leader for the past 15 years has been typically somewhere around 62-67%. Tyson Chandler had a 70.8% last season (2011-12) - a league record - and he holds two of the top five single season TS% spots, along with Artis Gilmore (two) and Wilt Chamberlain (one). The top 25 single-season TS% seasons are all above 65.7%, so we'll use that as our benchmark.
A note on using TS% solely to identify a great player: number seven on the all-time TS% single season list is Tim Legler, which is evidence that there's not a direct correlation between TS% and superstardom. What we DO know that it can be valuable in finding an efficient player to stick into the rotation.
In Our League
So, who's shooting well in our league?
It's early in the season, so our season stats are going to be skewed. As an example, Mexico City's Andy Johnson leads the league in TS% with 78.7%.
But we do have a full season to work with already, so instead of looking at this season alone, we're going to look at "career" TS%, which in this case means all of last season plus these last three sims. What we see is a similar pattern to what the NBA exhibits:
Top 10 Career TS% (Minimum 60 total games)
1. 64.2% - Bill Russell (C, Cairo)
2. 63.1% - Bennie Borgmann (SG, Madrid)
3. 61.4% - Gene Conley (PF, Beijing)
4. 61.1% - Ray Felix (C, Brazil)
5. 60.9% - Wilt Chamberlain (C, London)
6. 60.1% - Andy Phillip (SG, Tel Aviv)
7. 59.8% - Janus Krumins (C, Sydney)
8. 59.4% - Dolph Schayes (C, New York)
9. 59.3% - Neil Johnston (PF, Chicago
10. 59.3% - B. McCracken (SF, Berlin)
Lots of big men, though in our case a lot of our big men are higher usage players - with Borgmann, Chamberlain and Schayes all in the top 15 in team usage (which we will cover next week.) Still, it's the same pattern. Big men and sharpshooters.
In Closing
True Shooting Percentage - like the traditional shooting percentage categories - does not dictate a specific level of talent overall. However, a good TS% can help determine which players are most efficient, which comes into play both in creating super-duper-stars, big man role players and in finding great bench pieces.
