DukeTyrion
Joined: Feb 18, 2004
|
 Posted:
Apr 20, 2011 - 15:59 |
 
|
| maysrill wrote: | | DukeTyrion wrote: | Which reminds me of another puzzle;
There are 3 cups, one of them has a pea underneath it. After you choose a cup you will be shown the lack of contents on one of the other cups and asked if you wish to keep your current cup, or swap to the last remaining cup.
The answer is of course, that you should always swap ... but that's a whole different puzzle. |
The easy way to explain that one:
-After they've eliminated one of the cups, the remaining cup is 66% likely to have the pea under it. The one you chose originally only had a 33% chance.
-If you chose the correct cup in the first place (33%) you lose by switching. If you chose the incorrect cup in the first place (67% chance), then they've eliminated the only other wrong answer, and it is guaranteed to be under the remaining cup. |
Indeed, but it can be quite interesting to see number of people who try to argue it has become 50/50 (which indeed it is for anyone entering the room who hasn't seen the previous part) |
|
|
freak_in_a_frock
Joined: Aug 02, 2003
|
| Posted:
Apr 20, 2011 - 16:14 |
|
As far as i am concerned everything is 50/50. Either it will or it won't.
It makes stats a lot easier to accept, and also makes for a far more interesting way of playing blood bowl. Need a pow on a -2dice block, 50% chance of it happening... either it will or it won't... |
|
|
DukeTyrion
Joined: Feb 18, 2004
|
| Posted:
Apr 20, 2011 - 16:25 |
|
| freak_in_a_frock wrote: | As far as i am concerned everything is 50/50. Either it will or it won't.
It makes stats a lot easier to accept, and also makes for a far more interesting way of playing blood bowl. Need a pow on a -2dice block, 50% chance of it happening... either it will or it won't... |
Unless you are PeteW, in which case there is a 98% chance of it happening. |
|
|
Nelphine
Joined: Apr 01, 2011
|
| Posted:
Apr 20, 2011 - 16:30 |
|
the comment about 11 goblin teams is actually relevant, and it does change the probability.
Specifically, your pool is 'all other teams in the box'; but the chance of you activating at the same time as team A is not the same chance as you activating at the same time as team B. If the chances are the same, then yes, the low number of goblin teams becomes 10/(n-1). But if for instance 9 of those goblin teams are owned by the same coach, and he only ever plays once per month, the 10th team is yours, and the 11th team is a coach with normal activation chances, then the chance becomes 1/(n-1)+(1/30)*(9/n-1). (The 1/30 is a completely arbitrary number that assumes if he activates on a given day, you will activate at the same time. The number should actually be much lower).
Thus, teams played by extremely active coach's get a higher weighting than teams that are run by very casual coaches. However, to get accurate numbers for this, you would need to look at the information for a significant number of coaches, which is probably a task that is outside most of our scope.
The information that murker only played 1 chaos team in his first 34 games is quite interesting though, and seems to show the common bias of seeing a pattern in a small selection and erroneously extrapolating it to the larger world. |
|
|
licker
Joined: Jul 10, 2009
|
| Posted:
Apr 20, 2011 - 16:43 |
|
If you have to caveat your statement so heavily then it's really not worth debating.
I think people have become too fixated on Murker personally, rather than the question many are really asking.
Why should [B] have any bias at all?
Meaning, just make the schedule based purely off of TV. Do not take any other factors into consideration.
Now that's not to say it would be 'better' or 'worse', it would just be different. But this gets to the question of what is the point of [B] in the first place? The current system does indeed prefer certain matchups over others, and coaches who grok that will meta it for their own benefit. This does not seem exactly intended, and this kind of meta gaming is really outside of team design, it's only input is which race do you chose (races even).
And for the issue of 'get rid of ranked', well... it does have some merit (though I wouldn't personally advocate it) in as much as it would produce larger pools of teams to draw from and would get away from some of the small sample size issues which no doubt are an issue at certain times because I'm sure there is clumping of teams at certain times since the player base is not as large as everyone would like  |
|
|
Mr_Foulscumm
Joined: Mar 05, 2005
|
| Posted:
Apr 20, 2011 - 16:55 |
|
| FreeRange wrote: | | Mr_Foulscumm wrote: | | Some people seem to be under the impression that Blackbox was made for everyone. It's not. Use another method to play. This concept shouldn't be that hard to grasp should it? |
You're right, I should go play League since I can't get a game in Ranked...
oh wait...
As of right now there are 69 users online and only one team in the gamefinder. What kind of choice does that leave? |
Not to sound harsh or anything... but your argument is the dumbs. |
_________________
Everybody's favorite coach on FUMBBL |
|
Grod
Joined: Sep 30, 2003
|
| Posted:
Apr 20, 2011 - 17:11 |
|
Betting on someone moving cups around is idiotic and you deserve to lose all your money if you actually believe it is fair, irrespective of your knowledge of probability. I saw some guys in Barcelona once doing this. They had people planted in the crowd who would come in and make "easy money". Then idiot tourists would step up and lose. Then the crowd plants would come back afterwards and give the money back discreetly so they could do it again. |
_________________
I am so clever that sometimes I don't understand a single word of what I am saying.
Oscar Wilde |
|
PurpleChest
Joined: Oct 25, 2003
|
| DukeTyrion wrote: | | maysrill wrote: | | DukeTyrion wrote: | Which reminds me of another puzzle;
There are 3 cups, one of them has a pea underneath it. After you choose a cup you will be shown the lack of contents on one of the other cups and asked if you wish to keep your current cup, or swap to the last remaining cup.
The answer is of course, that you should always swap ... but that's a whole different puzzle. |
The easy way to explain that one:
-After they've eliminated one of the cups, the remaining cup is 66% likely to have the pea under it. The one you chose originally only had a 33% chance.
-If you chose the correct cup in the first place (33%) you lose by switching. If you chose the incorrect cup in the first place (67% chance), then they've eliminated the only other wrong answer, and it is guaranteed to be under the remaining cup. |
Indeed, but it can be quite interesting to see number of people who try to argue it has become 50/50 (which indeed it is for anyone entering the room who hasn't seen the previous part) |
one factor so far missing from this puzzle is the knowledge of the person showing you the empty cup. If they KNEW it was empty and showed you it follows the rules above, and you should switch.
If it is a blind draw that just happens to produce a no pea result (the chooser had a 33% of picking the pea and showing you it), the odds are then 50-50. |
_________________
Barbarus hic ego sum, quia non intelligor illis -Ovid
I am a barbarian here because i am not understood by anyone |
|
pythrr
Joined: Mar 07, 2006
|
| Posted:
Apr 20, 2011 - 17:15 |
|
| The_Murker wrote: | | Forgive me, I started a poll very similar to a recent one. It is a last ditch attempt to move my personal argument away from the data and towards what the community desires. It does not appear Christer is prepared to defend the merits of the suitability modifier, the existence of which logically requires the additional mirror match modifier. I am against both, and feel perhaps many others are. The current formula and scheduler "works", and the effort was and is appreciated, but I think it could be better to fall more in line which what Box could/should be. A free for all. One that does not attempt to make games "better." Just provides numerically similar match-ups. Don't like TV? It's the rules. Don't like CLAWPOMB? It's the rules. We'll see how those interested enough to vote feel. Thanks for all the comments. |
Dude, sorry to break it to you, but this is not a democracy. |
|
|
Astarael
Joined: Aug 14, 2005
|
| Posted:
Apr 20, 2011 - 17:25 |
|
Those argueing that it should be a 'free for all' and nothing should be prejudiced for or against... Shouldn't that mean there should be no 15% TV rule? It's an extreme, but in a purely random no bias box that'd have to be removed.
The stuff implemented is aimed at keeping things both competitive and fun. There needs to be an implemented bias against mirror matches, otherwise it's prejudiced towards them. This has all been said before but some people seem to still be acting like they think it's to avoid mirror matches. It's not, it's to avoid the natural bias towards them. |
_________________
Oh my. |
|
licker
Joined: Jul 10, 2009
|
| Posted:
Apr 20, 2011 - 17:33 |
|
| Astarael wrote: | Those argueing that it should be a 'free for all' and nothing should be prejudiced for or against... Shouldn't that mean there should be no 15% TV rule? It's an extreme, but in a purely random no bias box that'd have to be removed.
The stuff implemented is aimed at keeping things both competitive and fun. There needs to be an implemented bias against mirror matches, otherwise it's prejudiced towards them. This has all been said before but some people seem to still be acting like they think it's to avoid mirror matches. It's not, it's to avoid the natural bias towards them. |
Well you could remove the 15% rule, but it's not necessary. I'd think when people say 'make TV the only factor' that implies that teams are still matched to the closest TV team available.
No one to my recollection has advocated a completely random draw where the only criterion is having a team on the list.
And I'm not sure what you're trying to say in your 2nd paragraph... the bias against mirror matches is not there to avoid them? Ok, it's to balance them out vs. the racial suitability score, but the proposal is not to just axe the mirror match modifier alone, it's to axe the entire inclusion of bias altogether. Other than to do the draw based on TV, which isn't biased towards which races play which other races more or less frequently. Which is the point. |
|
|
Grod
Joined: Sep 30, 2003
|
| Posted:
Apr 20, 2011 - 17:38 |
|
Using coach ability (say BWR) in the formula would probably improve the evenness of matchups more than any other quantity (including possibly TV). So the point is moot about whether matchups are even or not. |
_________________
I am so clever that sometimes I don't understand a single word of what I am saying.
Oscar Wilde
Last edited by Grod on %b %20, %2011 - %17:%Apr; edited 1 time in total |
|
Garion
Joined: Aug 19, 2009
|
| Posted:
Apr 20, 2011 - 17:38 |
|
 Is this thread going to beat the ClawPOMB one? |
_________________
 |
|
licker
Joined: Jul 10, 2009
|
| Posted:
Apr 20, 2011 - 17:42 |
|
It might be the only thing which can beat ClawPOMB...
 |
|
|
pythrr
Joined: Mar 07, 2006
|
| Posted:
Apr 20, 2011 - 17:47 |
|
please, for the love of all that is unholy, lock this steaming pile of .... |
_________________
 |
|
|
|
| |
NBFL Season 32: The ...
Search
