BuzzerBeater Forums

Hello, I spent some time and made a little research about Freethrow skill correlation to accuracy. For each skill level I used 5 different test subjects in Transfer list. Of course using more players, stats would be more accurate.

http://photosliv.ru/9297.jpg

Please note that:

Always exists a tolerance of 2-3%
All numbers are taken at the end of season, so it is impossible to tell if player increased during that time in FT skill.
In the second box first part of numbers represent made shots and second part attempted - this means that same player is same color - for example first test subject with atrocious FT skill made 14/41 freethrows during this season. You can ignore that box because it is made for me to count easier.
Test subjects remain anonymous.
Legendary skill is usually 20+ so It is impossible to determine if it is 20 or higher skill.

I made a conclusion that profilic FT level is enough for my players. Hope you have something to add and hope this will help you. If anyone willing to make better research using more test subjects, I can sent original excel file.

Great work indeed.

Great, great work. It´s very interesting.
Thanks for sharing this.

Very interesting indeed! Thank you!
Now I just wait for something with more players in order to correct the strange numbers (prolific being better than the levels just ahead)

Very interesting. What it tells me is that inept is really the bare minimum that you should have in the FT department. Optimum %/training value.

I did something similar a month or so ago, and came up with pretty much the same numbers. I tried to get at least 500 free throw attempts for every skill level, and made a best fit curve.

I (Excel) came up with FT%=0.2585*ln(x)+0.1849, where x=numerical skill level (atrocious=1, pitiful=2, etc.), and for all you stat people, this formula had an R-squared value of 0.977 (pretty high correlation).

Hope this helps.

0.977 wow. And remember you don't know if players skill increased during that period. The best way for more accurate calculations is using only players who are not trained FT skill entire season. More attempts the more accurate info. Also there is no info about how stamina affects FT. Maybe if stamina is low and it is late period of game - chances to hit FT decrease? In addition it is unclear about enthusiasm, home court and other factors influence to this skill. Judging only by 5 test subjects (which is very low in any scientific research) I can see this skill could really be a constant with only very tiny random factor or even none.

Interesting that in transfer market there are more legendary or legendary+ skilled players than colossal or phenomenal. I guess that it is very hard to stop psychologically to train if you already trained a player to that level.

If someone will do more accurate study it will be great, but I don't think it is worth that time. Basically you can already imagine how your player will perform. More interesting thing what I want to know is free throw shooting tendency for each position. For example C and PF imo should have at least 10-11 FT in you want normal results. SF SF and PG could "afford" lower skills. Then we will be able to see which skill minimum all position players need to have then this information will be useful.

Last edited by Jack Bauer at 04/04/2011 18:15:38

Getting the percentage of the upper skill levels is problematic as was mentioned, because if he has something like "phenomenal" in Free Throws, there is a good chance that the GM was training free throws during the season, the solution to minimise the impact of this problem would be to increase the age of the player as much as possible, making it more likely that the skill increase would've occurred when he was younger.

I would say that Respectable is really the minimum that a decent player should have because anything under 70% is terrible, especially at the end of close games where the opposing team will foul the player intentionally. There is a high chance that the most important games of the season that you play will be close games decided at the end of the game and you don't want your Point Guard, who is Awful at free throws, to be taking the deciding shots from the line and missing them, throwing away a game that should have been won.

As was said before, if you really can't afford to have decent free throw shooters, at least look for Inept, because anything less than that is no better than Shaq. I also agree that Power Forwards and Centers could benefit from having an even higher skill, somewhere in the ranks of Prolific because they are the ones taking the most shots and even though the percentage only rises by about 13% when compared to Respectable, during the whole course of a season that would make a considerable difference.

Anyway, good work, this table should be seen by all respectable GM's of Buzzerbeater in order to have an idea about the importance of the Free Throw skill.

this indeed confirms what I have felt for a long time.

once respectable is reached, it's not realy profitable to train FT anymore.
If a player has 7/10 or 8/10 chance to hit the FT, it will only make you win or loose a game in very rare cases, and at that time, the stamina and GS of the player, and the amount of minutes he already played, or if was rested recently, will be of far greater impact on the fact he will make the shot.

What this study indicates is that 'inept' is good enough, and that 'respectable' is good, meaning ideal FT skill for a player is somewhere between inept and respectable.
I think it's too narrow.

I have been asking about FT and why higher skill doesn't result in clearly better results (my PG for instance scored more % FT when he was respectable, then when he was proficient, most likely due to the other factors GS, stamina and the timepoint where he takes the FTs during games).
They also made statistics and improvement of FT skill also improved % scored. But that is also what this study indicates.

I'd would like to see a better dividing. For instance it will need to take an average skill to get 50% chance of scoring.
My dividing of skills would look more like this:
atrocious: around 5% chance of scoring
2: around 10%
3 around 20%
4 around 30%
5 around 40%
6 50%
7 55%
8 60 %
9 65%
10 70%
11 74%
12 78%
13 82
14 85
15 88
16 91
17 94
18 96
19 98
20 around 100%

It would also not be pushing managers to train to real high levels, but at least managers will be motivated to train up to 11 or 12 at least, where now many are happy with level 6...
It wouldn't be good to force managers to train FT to at least 12 before they get 60%change either, so to make it linear would go to far. But as it is now, I think the diffrence between level 8 and level 14 is not big enough, and manager will not be pursuaded to train FT beyoond level 7 or 8.

Did you see if experience has any impact on FT?

Because i also have thought that experience has an impact on FT...

Last edited by Marot at 04/05/2011 06:36:18