BuzzerBeater Forums

Spoko, jedni i drudzy są potrzebni buzzerbeater... I wytrawni weterani, i świeża, jurna krew głodna sukcesów.

W tym cały urok.

Last edited by Bolesław III Krzywousty at 7/28/2011 3:41:11 PM

A co Wy na to?

As announced previously, during this offseason, all of BuzzerBeater’s trainers have been studying hard at BuzzerBeater University, getting their Ph.D.s in cross-training. As a result, they expect to produce more well rounded athletes. In the past, coaches would teach a new skill (such as driving) using related things players already knew how to do (such as handling), so that players with a higher handling skill would learn driving more quickly, and players using this method to improve their driving ability would also learn better handling, at a slower rate.

With cross-training, these new techniques emphasize each skill as part of becoming a more complete basketball player, and make reference to every skill. Thus, improvement in driving is related not just to related skills (like handling), but also more weakly to unrelated skills (like shot blocking). For the average player in the game, this will result in approximately 10% slower training in their primary skill than before, and additional training in other skills approximately corresponding to that 10% loss. A particularly well-rounded player will continue to receive cross-training but will see a much lower reduction in primary skill training, while a particularly one-dimensional player will see a larger loss. Although players will always receive a set amount of cross-training, it is difficult to predict which unrelated skills will improve on a weekly basis.

czytajac o takich pomyslach zastanawiam sie jeszcze kieedy wprowadza dodatkowy rzut kostka, ktory bedzie decydowal jakie skille pojda w gore i o ile :)

It looks like there have been a bunch of different types of reactions to why this might be a bad idea. Let me see if I can try and answer some of the criticisms:

1) Why is this change needed?

Several reasons that add up to this being the most elegant solution, which is part of why it's hard to explain. Part of it was trying to deal with the problem of players whose salaries exceeded their utility, causing them to become nomads. We'd like to ensure that a player making $400k/week will actually provide close enough to $400k/week of value that they're useful to somebody who can afford them, and this helps us get closer to that goal. At the same time, it's a bit more realistic, it makes the training mini-game more interesting, etc. But it's not going to stop you from making unique players or anything like that. It's also in part a reaction to transfer prices and trying to stabilize both the salary curve (the change for this season was smaller than in previous seasons) and transfer prices for extraordinary players (of many types, both smart and not-so-smart). To put it another way, if you have a favorite player whom you'd like to make mono-skilled, it would be nice to be able to keep training him forever without his really becoming a liability to your team. A change like this is a gentle nudge in the direction of making that possible, too; the problem with mono-skilled players is NOT that they are too effective, but rather that they are not effective enough given their salaries, yet no pool of comparable multi-skilled players exists. Our hope is that this change will bring those back into better balance, giving teams more of a choice of strategies for how to build.

It's a bit complicated to answer this one, because there really were several things we reacted to, not just one overriding concern, and we picked a solution that attempts to solve them all simultaneously. The plurality answer would be the high-salaried nomads.


2) Isn't this going to kill players because they run out of potential?

Let me tell you a little bit more about how salaries and potential work. Both of them are based (in slightly different ways) on the idea of looking at all of the different ways a player could be used, picking the one in which they will be most effective, and giving them a salary / calculating their remaining potential on that basis. The "best position" shown on your player is roughly (again, not quite exactly) an indicator of how the player is currently viewed. If you have a point guard, and he improves in shot blocking, he'll almost certainly still be evaluated as a point guard, which means that the shot blocking will be a neglible difference in how effective he appears to be as a point guard, and thus have a negligible effect upon his salary/potential. To put it another way, what we're doing is helping your trainers to make your players better in a way that increases the ratio of their effectiveness to their salary / remaining potential, and that should generally be merely a good thing.

3) This invalidates my strategy!

There's a good reason we announced this one well in advance, and it's in order to let teams try and plan properly. We try to do that every time the rules might impact a strategy. I don't actually expect an effect on this level is really going to have a major impact, but it's certainly a change, and there are some strategies that are hurt and some others that are helped by any change. That's always true, and larger changes are announced in advance in order to help prevent teams from being surprised by the change.

4) But...I don't believe you!

I'm not sure how I'd answer this one, but I will ask: if you were given the option to avoid skill increases from cross-training (all other effects would continue to apply; you'd just be giving up on increases with zero compensation), would you want it? If enough people actually would prefer that, it would be easy for us to implement...

Assumptions-
THe 10% is random. The post doesn't really say it is, we are assuming. It doesn't even commit to 10%. Its based on how obsurdly imbalanced the player is. The more imbalanced the more slow, the less the less.

Issues with my analysis which can't be solved, but are acceptible I feel.
-We can't see sublevels
-We don't know order of training
-We don't know the highend effect of this new elastic. 10% was given as the average, what is the max? If its like 50% then my analysis is way off, as I will go with 10%, the average over a players life I assume
-We don't know the starting skills

My method.
Since players can only start strong, and usually 7 at best. I will just assume any skill over 7 has been trained for all points above 7. All skills below I will just assume have not been trained. Obviously this is not the true case, but its the best we can do. For each point trained I will subtract 10%. I will pool the subtracted points together and then divide them amongst all skills.

This is VERY ruff math obviously.

Current players

Jump Shot: 7 Jump Range: 3
Outside Def.: 3 Handling: 6
Driving: 2 Passing: 4
Inside Shot: 18 Inside Def.: 18
Rebounding: 18 Shot Blocking: 10
Stamina: 8 Free Throw: 8

(6 skills above 7, total of 38 points to distribute 3.7 points and subract 3.7)


Jump Shot: 10 Jump Range: 3
Outside Def.: 2 Handling: 6
Driving: 1 Passing: 2
Inside Shot: 17 Inside Def.: 20
Rebounding: 17 Shot Blocking: 6
Stamina: 6 Free Throw: 3

(4 skills above 7, total 36 total, 3.6 points to distribute and 3.6 to subtract)

Post new system=

Jump Shot: 7~8 Jump Range: 3~4
Outside Def.: 3~4 Handling: 6~7
Driving: 2~3 Passing: 4~5
Inside Shot: 16~17 Inside Def.: 16~17
Rebounding: 16~17 Shot Blocking: 9~10
Stamina: 7~8 Free Throw: 7~8

Jump Shot: 9~10 Jump Range: 3~4
Outside Def.: 2~3 Handling: 6~7
Driving: 1~2 Passing: 2~3
Inside Shot: 15~16 Inside Def.: 18~19
Rebounding: 15~16 Shot Blocking: 6~7
Stamina: 6~7 Free Throw: 3~4


As we can see, with training to level 20, assuming we start at 0 even is a 20 point gain, at -10%, assuming that is top end, you can lost 2 points. If those 2 points are randomly distributed over the other 11 skills.... its like less than .2 pops.

ITS NOT GOING TO TURN YOUR C INTO AN SF....

People are overracting. We are talking about taking a 300k salary player with atrocious stats, maybe giving him a pitiful instead and lowering his primary skills an average of about ONE POP each, which will actually lowers the game effectiveness for those specific skills about 10% or whatever but is gonna lower the salary, because of the way it is calculated, by as much as 30%. That is a big change.

So you are gonna have cheaper monostat monsters in terms of salary who will need about 2~6 weeks MORE monoskill training to reach legendary, depending on the skill, since we are talking about a pop or two here.


Trochę spokojniej? ;)

No to pierwsze zwycięstwo wyrwane ;)

No to pierwsze zwycięstwo wyrwane ;)

Zespół New York Knickerbockers przyłożył się do gry bardziej niż Radogoszczanka.

No to teraz masz u chłopaków przejebane...

;-)))

Last edited by tomek de Mello_ELIT at 7/30/2011 4:54:30 PM

Ostatnie miejsce w konferencji.
Przedostatnie w lidze.
Pięknie zacząłem.
Jak zwykle ;)

Gratulacje dla Ryśka z klanu

Spokojnie Hawkeyed jeszcze się odgryziemy, rewanże u nas.

Pozdro

Last edited by Głuchy at 7/31/2011 3:42:53 AM

Ostatnie miejsce w konferencji.
Przedostatnie w lidze.
Pięknie zacząłem.
Jak zwykle ;)

Gratulacje dla Ryśka z klanu

No, no... Tak patrze to najmocniejszego składu to nie wystawiłes ;-)

A ja fart jakich mało;-) Hegemonia to prawdziwe Lwy!

Zaczęło się niewesoło, ale nie ma co wpadać w paranoję. Dopiero pierwszy mecz. Na 17-5 się nie nastawiam.