BuzzerBeater Forums

BB Global (English) > Game Engine Discussion

Game Engine Discussion

Set priority
Show messages by
This Post:
00
259783.21 in reply to 259783.20
Date: 7/2/2014 10:13:42 AM
Overall Posts Rated:
766766
ok i agree with that also.

Dunks are definitely non-driving based shots.

So just to debunk a few things ive said previously.

- There are no dunks off of a drive.
- DR + IS is most likely used as a combination in the formula for calculating shot % during a drive. (Although JoeMaverick is providing some evidence to counter this, however my own anecdotal evidence suggests that IS will increase the chances of a driving-based shot going in. )

and just to backtrack a bit as well regarding the Jump Range Jump shot formula, my initial formula suggested
(JS + <distance_factor>*JR) - I dont think this is wrong actually? maybe a multiplication than an addition? Doesn't really matter if its applied as a factor in the end formula?

After listening to Nachtmahr, there is probably a slight change to that. something like:
(JS * <distance_factor>) * ((JR)*<distance_factor>)
That is, Jump shot is affected by distance, regardless.... then this result is applied against the JR which is also affected by distance.
There might be some weighting applied as well, to reduce the actual effectiveness that the JR skill applies here. But thats always going to be impossible to tell.

The end result of a guy with 15JS and 15JR is a 'score' of 56.25, compared to a guy who has atrocious JR, who would get a score of 3.75

That actually seems about right. Listed below are the scores, for a guy with 15 JS and then JR from Atrocious through to 15
1(Atroc) - 3.75
3 - 11.25
5 - 18.75
7 - 26.25
9 - 33.75
11 - 41.25
13 - 48.75
15(marv) - 56.25

Seems not too bad really!


This Post:
00
259783.24 in reply to 259783.22
Date: 7/2/2014 11:25:53 PM
Overall Posts Rated:
766766
IS hasn't more impact on a driving layup than JR. The reason behind that is that it will be easier for a good outside shooter to score a driving layup than for a player who is only able to drive (for example).


I think i know what you are saying and I kind of agree with you somewhat.

What you are suggesting is - that if a player has lots of different skills, and lots of different abilities, this gives him a vast array of basketball weaponary to use and as such, he will be a better player at everything.
eg: Because he can shoot the 3point shot really well, his defender will defend him really close, which gives him a better chance of driving past him and scoring on driving layups. Or because he acn hit the 3pt shot but is playing at PF, he will hit more 3pt shots because the defender will have poor OD etc etc.

I think logically, this happens in real life. and i recently suggested this somewhere, and i think that, from a 'shot selection' perspective, the GE will try and pick this player more because he will have more mis-matches to expose. And i think we can discuss this later on, because the way that the GE 'selects' which player is going to shoot will be a very complicated thing.

But if we were to just talk purely about the isolated situation of, two players, performing layups, with the exact same defender, then surely IS makes a difference to % that one will make over the other?

From: jonte

This Post:
00
259783.28 in reply to 259783.27
Date: 7/3/2014 9:42:41 AM
Overall Posts Rated:
925925
just one small question that I asked myself recently:

if:

a) driving layups are mostly based on driving and dunks are mostly based on IS. and
b) it is possible that a drive is followed by a dunk (instead of a layup) (although you said it is not?)

Does a Player with 15 driving and 1 IS take less dunks than a Player with 15 driving and 10 IS? and an (outside) Player with 10 Driving and 15 IS?

based on the fg% percentages you posted it doesn't seem so, but i wonder if you have any data on the shot distrubution of Players like thos three i mentioned above.

Last edited by jonte at 7/3/2014 9:43:17 AM

This Post:
00
259783.30 in reply to 259783.28
Date: 7/3/2014 10:32:15 AM
Overall Posts Rated:
766766
b) it is possible that a drive is followed by a dunk (instead of a layup) (although you said it is not?)


I think ive pretty much established that dunks are not based off of a drive. Ever.

So your next questions
Does a Player with 15 driving and 1 IS take less dunks than a Player with 15 driving and 10 IS? and an (outside) Player with 10 Driving and 15 IS?


So logically you would think that IS is the major skill influencing dunks. IS = 1 would mean less dunks thatn IS = 10.

if the player is set to play outside, then i think we are starting to talk about how the Game engine takes into account the position, and that position in the offensive engine, which we haven't really covered yet, but in brief, i believe that these other factors may influence number of dunks.

This Post:
00
259783.31 in reply to 259783.29
Date: 7/3/2014 10:44:10 AM
Overall Posts Rated:
766766
im still not sure about that. Does JS really influence a dunk? even if its only 1%
Does IS really affect a 3pt shot? even if its 1%?

This would suggest that there is a standard formula for calculation of all shots - which if this is the case, then we can use maths to possibly reverse engineer something that resembles a formula.
Using previously mentioned stats:
driving layup:
FG%: ~35%
assisted: 40-45%
defended: ~55%
shooting foul: 20-25% of the defended shots


we can use those stats and the 4 skills of JS, JR, DR and IS to possibly work on a formula. a bit like a simultaneous equation, but with 4 unknowns..ill have a think about that actually... a bit late in the evening for my brain to start thinking about quadratically simulatenously coehsively molecularly equations.

Advertisement