Vicimus
Joined: Nov 16, 2005
|
Posted:
Jan 04, 2006 - 08:30 |
|
Okay I made a quick JavaScript page to prove that Java random number generation is totally okay. The source code is there to be scrutanised if anyone doesn't believe me.
Second, there are some interesting statistics I figured out using probability mathematics, check it out...
(Edit: Peke pointed out a problem with my math below that has been fixed. Nothing to do with Java's random number generation)
http://home.kooee.com.au/issima/random.htm
Cheers,
Vic. |
Last edited by Vicimus on %b %04, %2006 - %09:%Jan; edited 1 time in total |
|
peke
Joined: Jun 04, 2004
|
  Posted:
Jan 04, 2006 - 08:43 |
|
Well. It generates 120% test results. How ya explain that? |
_________________ People who are organized are just too lazy to look for things. |
|
Wizard
Joined: Jul 09, 2004
|
  Posted:
Jan 04, 2006 - 08:46 |
|
The total percentage is 120% on 6 dice - the totals for each if distrubted evenly should be 16.6% yeah? |
_________________ "As long as one person lives in darkness then it seems to be a responsibility to tell other people."
Last edited by Wizard on %b %04, %2006 - %09:%Jan; edited 1 time in total |
|
Vicimus
Joined: Nov 16, 2005
|
  Posted:
Jan 04, 2006 - 08:49 |
|
Ooops...I'm sure that's an error with my arrays ... they always confuse me starting at 0 instead of 1 ... hold on I'll fix it up and re-upload. |
|
|
Rynkky
Joined: Aug 03, 2004
|
  Posted:
Jan 04, 2006 - 08:53 |
|
peke wrote: | Well. It generates 120% test results. How ya explain that? |
This proofs that Java RNG is flawed1!11!!1!!! |
|
|
Vicimus
Joined: Nov 16, 2005
|
  Posted:
Jan 04, 2006 - 09:00 |
|
No, it proves my maths was flawed. Nothing wrong with the Java RNG, just my brain.
The numbers where all right, I was just configuring the percentages incorrectly. I mean, do you really think the people at Java are so stupid they can't even create a "consistently inconsistent" random number between 1 and 6? ... gimme a break.
Anyway, the argument is closed. Look now. |
|
|
rainbowsoldier
Joined: Dec 08, 2005
|
  Posted:
Jan 04, 2006 - 09:02 |
|
Its not the number generator that needs to be adjusted. The luck % in the top bar needs to be modified to include un rerollable dice, as it only counts those. Things like armour and KO wake up rolls would be nioce to reflect on your luck aswell. |
_________________ [url=http://www.sloganizer.net/en/style3,Nuffle.png][/url] |
|
tautology
Joined: Jan 30, 2004
|
  Posted:
Jan 04, 2006 - 09:03 |
|
Quote: |
44% - the probability of an unskilled AG3 player throwing a successfully caught short pass to another unskilled AG3 player without the use of a reroll (including possibility of catching an innacurate pass)
93% - the probability of an unskilled AG3 player throwing a successfully caught short pass to another unskilled AG3 player with the possible use of a reroll (including possibility of catching an innacurate pass).
|
Without stooping to doing the math, I feel certain that these at least are wrong.
Just on a common sense check, the possibility of passing 2 GFIs with rerolls (81.0185% by my calculations) , MUST exceed the possibility of making a 4+ pass, 3+ catch with rerolls, even factoring in the 7% or so chance that a scatter rolls back to the catcher on an inaccurate pass (which you list as 93%).
In general however, I agree with your sentiment |
|
|
Vicimus
Joined: Nov 16, 2005
|
  Posted:
Jan 04, 2006 - 09:06 |
|
rainbowsoldier,
While it might be possible to make it right in theory, that luck thing will never seem right. GFI and failing on turn 8 or the second half to win the game will not been seen as unlucky as GFI and failing to put an extra assist on an inconsequential block when you're up 3-0. While in theory they're both as "unlucky" as each other, one obviously matters a whole heap more than the other! |
|
|
Vicimus
Joined: Nov 16, 2005
|
  Posted:
Jan 04, 2006 - 09:11 |
|
tautology wrote: | Quote: |
44% - the probability of an unskilled AG3 player throwing a successfully caught short pass to another unskilled AG3 player without the use of a reroll (including possibility of catching an innacurate pass)
93% - the probability of an unskilled AG3 player throwing a successfully caught short pass to another unskilled AG3 player with the possible use of a reroll (including possibility of catching an innacurate pass).
|
Without stooping to doing the math, I feel certain that these at least are wrong.
Just on a common sense check, the possibility of passing 2 GFIs with rerolls (81.0185% by my calculations) , MUST exceed the possibility of making a 4+ pass, 3+ catch with rerolls, even factoring in the 7% or so chance that a scatter rolls back to the catcher on an inaccurate pass (which you list as 93%).
In general however, I agree with your sentiment |
Man, that one blow me away too! I double and triple checked the math (which by no stretch means that it is right), but it kept coming up with that. You have to take into consideration all possible methods of success. Even, for example...
Roll an innacurate pass (2/6 chance) times
reroll innacurate pass (2/6 chance) - i.e. reroll failed times
chance of ball scattering back to target square (6.25%) times
chance of catcher catching innaccurate pass (50%)
This is just one of about 8 different ways it could work. And you have to add them all together to take them into consideration as a boolean OR is reflected by the sum of possibilities in probabality math (or as far as I know it is!). |
|
|
Curro
Joined: Jun 07, 2005
|
  Posted:
Jan 04, 2006 - 09:17 |
|
Quote: |
0.5% - the probability of the same ogre failing his bonehead role three times in consecutive turns.
98.7% - the seeming probability of an ogre failing his bonehead role when he has to do something really important.
|
Ok, this is Absolutely correct. |
|
|
tautology
Joined: Jan 30, 2004
|
  Posted:
Jan 04, 2006 - 09:27 |
|
I think it might be more expedient in this case to simply calculate the possibilities of success:
Pass is accurate =50% chance of catching it with a reroll =88.89% total chance of success =44.4%
Pass is not accurate, but reroll works = 25% Chance of catching it without a reroll=66.7% total chance of success = 16.67%
We now have a total chance of success = 61.1%
Add in the chance of an inaccurate pass scattering to you (assume reroll was used to attempt accuracy, which is the best strategy for success)-- maybe 6.5%, with a 50% of catching the scatter, so add another 3.25% to our total and we get a final success chance around 64.35%, give or take on the scatter equation which I did once but refuse to do again |
Last edited by tautology on %b %04, %2006 - %10:%Jan; edited 3 times in total |
|
tobardin
Joined: Apr 08, 2005
|
  Posted:
Jan 04, 2006 - 09:29 |
|
I checked the math and probabilty for a fair catch with two unskilled AG3-players with RR is 61,1%!!
With RR and 6.5% Scatter-Back-Chance it is 62.7% (assuming the 6.5% Scatter-Back is right) |
|
|
Vicimus
Joined: Nov 16, 2005
|
  Posted:
Jan 04, 2006 - 09:35 |
|
Maybe, it was the most complex probability I worked out. You guys could well be right (although you've both come up with different numbers). Are you sure you're taking into consideration all the possibilities? If you only take into consideration half the ways it could go right, your number could be way off...
I did it like this...
Chance of throwing an accurate pass that is caught (no reroll needed) PLUS
chance of throwing an inaccurate pass that is caught (no reroll needed) PLUS
chance of throwing inaccur. pass that is rerolled to an accurrate pass and caught PLUS
chance of throwing accurate pass that is dropped and rerolled to be caught PLUS
chance of throwing inaccur. pass that is dropped and then rerolled to be caught PLUS
chance of throwing inaccur. pass that is rerolled and is still inaccurate but is caught PLUS
chance of fumbling then rerolling to an accurate pass that is caught PLUS
chance of fumbling then rerolling to an inaccurate pass that is caught
Not to say I am right, as it's a complex sum and I'm very tired. But I want to get this right eventually with your help! |
Last edited by Vicimus on %b %04, %2006 - %09:%Jan; edited 1 time in total |
|
tobardin
Joined: Apr 08, 2005
|
  Posted:
Jan 04, 2006 - 09:40 |
|
I added to the 61.1% only 1.625% for the Scatter because :
First it was not thrown accurate to the catcher=50%
Reroll the throw was a failure=50%
Scatter to target sqaure=6.5%
Fair Catch=50%
All multiplied resukts in 1.625% |
|
|
|