Deeper Dive Into a Player’s Height
Posted: Sat Dec 23, 2017 11:31 pm
The league has now gone through four drafts with the new college draft system, and the new system has had its fair share of praise and critiques. The new college draft system has injected the league with some serious talent with the likes of Buddy Pedraza already making a name for himself, being the most accomplished player of the group as a two time All Star Team selection and All NBA 3rd Team honor this past season. However, there’s been some adjustments the league has had to make. Sometimes the players are oddly skilled, such as a PG having higher post defense potentials than passing and perimeter defense, a big man that can’t rebound or defend with decent handles, or sub-40 stamina ratings that we weren’t used to. We also noticed these players come into the league very raw. Buddy Pedraza is an exception to this, but the league is finally starting to experience what Jeffrey Duren, Ben Simmons, Karl Anthony Towns and many others can be capable of. These issues can be fixed or resolved with either time and small investments of player trainings if you’re not willing to take the time to wait it out.
There’s one more issue that has been noted by other general managers with these players, however, that cannot be fixed and is set from the moment the draft class is imported: height. It all started when the first college season took place. Once the first index was uploaded to be examined, most of us flocked to team pages and noticed that big men, especially centers, were shorter than what we expected. It hasn’t changed much from year to year for the college index once a new season is updated. The big men did not seem so big.
With my team in a rebuild, I was curious to see what the new freshman class would look like for the upcoming season. The college index has not been updated for the new season, so I went over to the Top 100 Recruits link and, out of curiosity, checked to see the height of the upcoming big men. Of the Top 100 Recruits, there are 33 listed PFs and Cs, and below is the table for the count by height and position.
You may be asking “Is this listing even worth considering? Are we drafting higher ranked players?” As a quick test, I went through the last four draft classes and for each of the first 15 projected draft picks I checked their class ranking. Of the 60 players, 53 of the top 15 projected players were ranked in the Top 100 of their class, or 88% of the total players. I believe this list is worth looking at in terms of which group of players will make it into the league (As a fun side note, Ben Simmons was not ranked in the Top 100, one of the most talented players to be drafted from the system).
I was a little surprised by some of the numbers listed in the above table. First, to see such a heavy weight towards 6’8 centers was disheartening, because selfishly I was hoping for an opportunity to land a true freshman big man center in this next draft. There’s still the possibility of the other classes having players declare, I know. The other surprising count was a total of 8 power forwards at a listed 6’6 height or shorter, and the fact that the highest count for power forwards was 6 listed at 6’9, compared to the centers highest count of 6 listed at 6’8. It’s the first time I’ve counted each from the Top 100 listing, so I don’t know if this is just a strange class coming into the college system or if this was the norm for prior years.
Another limitation I ran into from not performing any true height testing or observations were that I don’t know if players grew from the recruited listed height. With no prior year college index history to go to like we do with our PBSL league, I cannot tell if anyone has grown from the time they were recruited to the time they were drafted. We know players do not grow in the PBSL league, but I don’t know if they do during college or from the transition of high school recruit to freshman year. Judging by the % of top 15 rated prospects in each of the draft that were ranked in the Top 100, and the uncertainty of not knowing if these players grow, it’s entirely possible the best PFs and Cs will be 6’8 or shorter.
After looking at those players’ height, I started to wonder if big men were shorter with this new system. If I was going to look into height, I decided to look into all positions to get a better grasp to see if it was just big men or all players in general, because I started to feel SGs were getting shorter as well.
I thought it would be easy to import and figure out in just a few minutes, but I ran into a few problems:
• The biggest problem was that I couldn’t use the All Signed Players list to get the average height by position. The most important reason why is that players listed positions are dependent on most minutes played at a certain position. For example, Gordon Hayward is listed as a PG, but he was originally drafted as a SF.
• Another minor reason is that I am missing out on all the free agents which limits myself to a smaller population size to compare data.
• Finally, it would be inefficient to separate manually the new college drafted players from the historical classes.
How did I get around these obstacles? I decided to import every draft class into one table by going through each archived year in our index history! It took a little bit of time, but once everything was in a table it was set and done. I had all listed players by their original position, and I classified each class by its index year.
There were only two small problems I ran into when I did this, one I took care of and the other that’s small and shouldn’t influence the results. First, for each draft class the computer filled out those classes with a bunch of red and orange potential players. I removed those players from consideration, as it was very highly unlikely they ever played a minute in the league. I wanted to see only players that either did or could have played in the league. The other problem, which is related to the color potential problem, is that during the early years of the league, the draft file was not finalized until after the season had ended. For example, in the 1996-1997 index season, the draft class shows Tracy McGrady as a purple potential player, but he was not a purple potential player when I drafted him. Draft files were usually finalized after the Finals and when the new reason rolled over in the index, meaning after the offseason, there is no way of viewing the final draft file from the early years because it gets replaced with the new seasons’ upcoming draft class. The reason I am certain there is probably zero effect on these results is that the draft file was tinkered only for the top prospects in the class, and I do not believe any yellow, green, or blue potential players were updated to be orange or red potential players. Unfortunately, the college coaches can play players at any position of their choosing. I saw with Harland Ellenger, who was a PG, get switched over to the SG position during his time at Oklahoma State. In the draft file when transferred over, he was sure enough a SG, what he was listed in the college index. There’s not much I can do to fix this, especially for players that played in college in the prior seasons. I’m going to have to live with the small error on this.
(tl;dr version: everything is fine, nothing effected the results in a major way I believe but several of the drafted college players may not be classified exactly at the position they were generated by the game)
Below are the results of the average height (in inches) by position for each of the draft classes.
I do want to add that the years listed are one year behind if you’re thinking of actual draft class years. The years listed are based on the index year I had to type in for the links to pull the pages needed. For example, the 2002 year is the draft class of Lebron James, Dwayne Wage, and Chris Bosh, not 2003. I didn’t realize until after I was near done with the tables, but I think it makes it easy to know which index year to go to if you’re curious.
The last 4 points for each of the positions listed in the graph are the new draft classes average height, and you can notice a slight drop from the 2014 year onward compared to the previous seasons, but this is such a big picture overall view, and we cannot make comparisons using this.
I decided to dig a little deeper by position, but wanted to add another element in the mix if we were going to break it down: the average height for all historical draft classes by position. This was easy to do, all I did was take the average position from the 1990-2013 draft class for each position and made it a benchmark to compare each season. Now we can not only ask “Is the college system smaller” but also “did the players grow or get shorter as the season continued?”
I had to list everything in inches to be able to sort and average out the values, so for those who are too lazy or just want a reference for inches to what we normally use as feet-and-inches, I'm providing the following table:
RESULTS BY POSITION
Point Guard
Shooting Guard
Small Forward
Power Forward
Center
ANALYSIS
I want to start with the center position, because this is the reason why I even started looking into the height of the players in the first place. The average height of a center from 1990-2013 was 83.5 inches, or just under 6'11, and each year of the new college system the centers were slightly shorter, but not as much as what we perceived it to be. You can take it how you want it, because I can see both sides of an argument here. In 3 of the 4 draft classes, the centers were on average the shortest the league has experienced, if only ever so slightly (1992 and 2008 were comparatively close). However, the height average difference was one at most 1.5 inches from the average of the 1990-2013 seasons. We can say that yes, the centers are shorter than average compared to the historical draft classes, but it's not shaking the foundation of the league.
What I got the most out of these graphs were obviously at the SG position. The college draft system SGs are nearly 2 inches shorter than the average of the historical draft classes and the height of these SGs are consistent with the 4 seasons we’ve had for the college system. There’s been a significant drop off for these player’s height, and I don’t think that will change with any of the upcoming draft classes. Another interesting trend I saw was for PGs, the average height of these players was growing from about the 2008 season onward, only to fall back down with the new college players that reverted them back to the height of PGs from the 1990s.
SUGGESTION
I do have a suggestion on how to remedy this situation. Before the shouts of training height and human growth hormone get thrown out, I’d like to point out that I do not think the players’ decrease in average height is necessarily a bad thing. The Hawks with Daniel Randle are starting to show that height doesn’t exactly restrict what position your player can play, if the player has sufficient ratings to play where you put him.
I did, however, prepare a few more charts by position for you to consider. I used the same tables to generate a % distribution to see how heights were being allocated towards the league. I have presented the following graphed out data:
• College (bar chart): this is the distribution % by height for those very same players in the draft class system referenced above in the tables I imported.
• Pros (bar chart): this is the distribution % by height for those very same players in the historical draft files referenced above in the tables I imported.
• College Distribution (line graph): this is new. Thanks to wombataholic being curious about the game, he found a Configuration file that lists out the Height distribution odds of generated players (i.e. all college created players). I imported this file into my Excel file and calculated out the odds of height by position and put them on this graph.
Not only are the players slightly shorter than average, but they are more uniformly generated at several heights. You can see it best at the SG position in the below graph where we have had a whopping 57.5% of SGs turn out to be 6’3. From all the graphs, you can see the college players currently drafted are lumped into several heights without much deviation at all, whereas the Pro players’ distribution is more spread out that gave us variety to choose from, which I am in huge favor of.
You can really begin to see how the College Distribution line graph reflects the increase/decrease in height by position for the College bar graph. Of the thousands of players that get generated in the college system, I’m guessing only about 52 players a season make the cut to be considered a possible PBSL player (using the at least yellow potential assumption), meaning we are most likely not going to see a true representation of the College Distribution from the Configuration file, but it should be reflective. I think we see that here.
I personally like to see more spread in this distribution just like we had with the historical draft classes, it adds to the randomness of the game and the league can have players like Shaun Livingston, Penny Hardaway, and others. My only suggestion would be to edit the Configuration file, which is setup in a way that we could do such a thing, in the college game and make small changes to several positions and monitor the results. Again, only so many players make it to the pros just like in real life, but making small changes to a college game none of us have any handle on where results don’t matter does not seem like decision that could shake the league up, as long as the changes were small and infrequent.
I’m interested to hear the league’s thoughts if this has sparked any ideas?
There’s one more issue that has been noted by other general managers with these players, however, that cannot be fixed and is set from the moment the draft class is imported: height. It all started when the first college season took place. Once the first index was uploaded to be examined, most of us flocked to team pages and noticed that big men, especially centers, were shorter than what we expected. It hasn’t changed much from year to year for the college index once a new season is updated. The big men did not seem so big.
With my team in a rebuild, I was curious to see what the new freshman class would look like for the upcoming season. The college index has not been updated for the new season, so I went over to the Top 100 Recruits link and, out of curiosity, checked to see the height of the upcoming big men. Of the Top 100 Recruits, there are 33 listed PFs and Cs, and below is the table for the count by height and position.
You may be asking “Is this listing even worth considering? Are we drafting higher ranked players?” As a quick test, I went through the last four draft classes and for each of the first 15 projected draft picks I checked their class ranking. Of the 60 players, 53 of the top 15 projected players were ranked in the Top 100 of their class, or 88% of the total players. I believe this list is worth looking at in terms of which group of players will make it into the league (As a fun side note, Ben Simmons was not ranked in the Top 100, one of the most talented players to be drafted from the system).
I was a little surprised by some of the numbers listed in the above table. First, to see such a heavy weight towards 6’8 centers was disheartening, because selfishly I was hoping for an opportunity to land a true freshman big man center in this next draft. There’s still the possibility of the other classes having players declare, I know. The other surprising count was a total of 8 power forwards at a listed 6’6 height or shorter, and the fact that the highest count for power forwards was 6 listed at 6’9, compared to the centers highest count of 6 listed at 6’8. It’s the first time I’ve counted each from the Top 100 listing, so I don’t know if this is just a strange class coming into the college system or if this was the norm for prior years.
Another limitation I ran into from not performing any true height testing or observations were that I don’t know if players grew from the recruited listed height. With no prior year college index history to go to like we do with our PBSL league, I cannot tell if anyone has grown from the time they were recruited to the time they were drafted. We know players do not grow in the PBSL league, but I don’t know if they do during college or from the transition of high school recruit to freshman year. Judging by the % of top 15 rated prospects in each of the draft that were ranked in the Top 100, and the uncertainty of not knowing if these players grow, it’s entirely possible the best PFs and Cs will be 6’8 or shorter.
After looking at those players’ height, I started to wonder if big men were shorter with this new system. If I was going to look into height, I decided to look into all positions to get a better grasp to see if it was just big men or all players in general, because I started to feel SGs were getting shorter as well.
I thought it would be easy to import and figure out in just a few minutes, but I ran into a few problems:
• The biggest problem was that I couldn’t use the All Signed Players list to get the average height by position. The most important reason why is that players listed positions are dependent on most minutes played at a certain position. For example, Gordon Hayward is listed as a PG, but he was originally drafted as a SF.
• Another minor reason is that I am missing out on all the free agents which limits myself to a smaller population size to compare data.
• Finally, it would be inefficient to separate manually the new college drafted players from the historical classes.
How did I get around these obstacles? I decided to import every draft class into one table by going through each archived year in our index history! It took a little bit of time, but once everything was in a table it was set and done. I had all listed players by their original position, and I classified each class by its index year.
There were only two small problems I ran into when I did this, one I took care of and the other that’s small and shouldn’t influence the results. First, for each draft class the computer filled out those classes with a bunch of red and orange potential players. I removed those players from consideration, as it was very highly unlikely they ever played a minute in the league. I wanted to see only players that either did or could have played in the league. The other problem, which is related to the color potential problem, is that during the early years of the league, the draft file was not finalized until after the season had ended. For example, in the 1996-1997 index season, the draft class shows Tracy McGrady as a purple potential player, but he was not a purple potential player when I drafted him. Draft files were usually finalized after the Finals and when the new reason rolled over in the index, meaning after the offseason, there is no way of viewing the final draft file from the early years because it gets replaced with the new seasons’ upcoming draft class. The reason I am certain there is probably zero effect on these results is that the draft file was tinkered only for the top prospects in the class, and I do not believe any yellow, green, or blue potential players were updated to be orange or red potential players. Unfortunately, the college coaches can play players at any position of their choosing. I saw with Harland Ellenger, who was a PG, get switched over to the SG position during his time at Oklahoma State. In the draft file when transferred over, he was sure enough a SG, what he was listed in the college index. There’s not much I can do to fix this, especially for players that played in college in the prior seasons. I’m going to have to live with the small error on this.
(tl;dr version: everything is fine, nothing effected the results in a major way I believe but several of the drafted college players may not be classified exactly at the position they were generated by the game)
Below are the results of the average height (in inches) by position for each of the draft classes.
I do want to add that the years listed are one year behind if you’re thinking of actual draft class years. The years listed are based on the index year I had to type in for the links to pull the pages needed. For example, the 2002 year is the draft class of Lebron James, Dwayne Wage, and Chris Bosh, not 2003. I didn’t realize until after I was near done with the tables, but I think it makes it easy to know which index year to go to if you’re curious.
The last 4 points for each of the positions listed in the graph are the new draft classes average height, and you can notice a slight drop from the 2014 year onward compared to the previous seasons, but this is such a big picture overall view, and we cannot make comparisons using this.
I decided to dig a little deeper by position, but wanted to add another element in the mix if we were going to break it down: the average height for all historical draft classes by position. This was easy to do, all I did was take the average position from the 1990-2013 draft class for each position and made it a benchmark to compare each season. Now we can not only ask “Is the college system smaller” but also “did the players grow or get shorter as the season continued?”
I had to list everything in inches to be able to sort and average out the values, so for those who are too lazy or just want a reference for inches to what we normally use as feet-and-inches, I'm providing the following table:
RESULTS BY POSITION
Point Guard
Shooting Guard
Small Forward
Power Forward
Center
ANALYSIS
I want to start with the center position, because this is the reason why I even started looking into the height of the players in the first place. The average height of a center from 1990-2013 was 83.5 inches, or just under 6'11, and each year of the new college system the centers were slightly shorter, but not as much as what we perceived it to be. You can take it how you want it, because I can see both sides of an argument here. In 3 of the 4 draft classes, the centers were on average the shortest the league has experienced, if only ever so slightly (1992 and 2008 were comparatively close). However, the height average difference was one at most 1.5 inches from the average of the 1990-2013 seasons. We can say that yes, the centers are shorter than average compared to the historical draft classes, but it's not shaking the foundation of the league.
What I got the most out of these graphs were obviously at the SG position. The college draft system SGs are nearly 2 inches shorter than the average of the historical draft classes and the height of these SGs are consistent with the 4 seasons we’ve had for the college system. There’s been a significant drop off for these player’s height, and I don’t think that will change with any of the upcoming draft classes. Another interesting trend I saw was for PGs, the average height of these players was growing from about the 2008 season onward, only to fall back down with the new college players that reverted them back to the height of PGs from the 1990s.
SUGGESTION
I do have a suggestion on how to remedy this situation. Before the shouts of training height and human growth hormone get thrown out, I’d like to point out that I do not think the players’ decrease in average height is necessarily a bad thing. The Hawks with Daniel Randle are starting to show that height doesn’t exactly restrict what position your player can play, if the player has sufficient ratings to play where you put him.
I did, however, prepare a few more charts by position for you to consider. I used the same tables to generate a % distribution to see how heights were being allocated towards the league. I have presented the following graphed out data:
• College (bar chart): this is the distribution % by height for those very same players in the draft class system referenced above in the tables I imported.
• Pros (bar chart): this is the distribution % by height for those very same players in the historical draft files referenced above in the tables I imported.
• College Distribution (line graph): this is new. Thanks to wombataholic being curious about the game, he found a Configuration file that lists out the Height distribution odds of generated players (i.e. all college created players). I imported this file into my Excel file and calculated out the odds of height by position and put them on this graph.
Not only are the players slightly shorter than average, but they are more uniformly generated at several heights. You can see it best at the SG position in the below graph where we have had a whopping 57.5% of SGs turn out to be 6’3. From all the graphs, you can see the college players currently drafted are lumped into several heights without much deviation at all, whereas the Pro players’ distribution is more spread out that gave us variety to choose from, which I am in huge favor of.
You can really begin to see how the College Distribution line graph reflects the increase/decrease in height by position for the College bar graph. Of the thousands of players that get generated in the college system, I’m guessing only about 52 players a season make the cut to be considered a possible PBSL player (using the at least yellow potential assumption), meaning we are most likely not going to see a true representation of the College Distribution from the Configuration file, but it should be reflective. I think we see that here.
I personally like to see more spread in this distribution just like we had with the historical draft classes, it adds to the randomness of the game and the league can have players like Shaun Livingston, Penny Hardaway, and others. My only suggestion would be to edit the Configuration file, which is setup in a way that we could do such a thing, in the college game and make small changes to several positions and monitor the results. Again, only so many players make it to the pros just like in real life, but making small changes to a college game none of us have any handle on where results don’t matter does not seem like decision that could shake the league up, as long as the changes were small and infrequent.
I’m interested to hear the league’s thoughts if this has sparked any ideas?
