SRL Archive - Discussions & Debates

With an increasing general level of education across the population, i'd expect gambling activities to slowly die down as more educated people realize how gambling is as good as dumping money down the drain.

Gambling hosts always get an unfair deal at their end, eg. 55% vs 45%, even with a basic literacy in math/common sense, the gamblers should realize that they are bound to lose in the long run.

The net profit is negative by nature of the biased deal, and the more you gamble, the more you are dragging the standard deviation down (Central limit theorem), that is, the more certain your loss will be.

Even in the case of a fair gamble, such as when the gambling is 'hosted' among 2 gamblers of a 50 50 chance, the expected return would be zero. That is, you are wasting precious time for absolute zero return. And since time=money, you are practically still losing money :p

It's fine if you enjoy it, (consider it a leisure expenditure), but the truth is, gambling often adversely affects the happiness of individuals. Even if you have a lucky day, the joy would only be temporary because eternal happiness can only be achieved when your success is achieved with immense effort.

It is, therefore disappointing/surprising that gambling activities can remain so prevalent (both in RS&RL) even after centuries of lesson learnt from history. Gambling is even more illogical than for instance, taking drugs because you have everything to lose but nothing to gain. (at least you have almost a 100% certainty to gain temporary gratification from drugs)

How many mathematical proofs are needed to convince the population that the Gambler's fallacies are simply flawed math/logical thinking? Would gambling continue to flourish even as literacy rates skyrocket? Discuss.

Gambling is fine if it's done in moderation, if you're good at math, you can predict the chances of what it will turn out to be, greatly wavering the odds in your favor.

Well said.

i'm with NKN on this one. he has a good point if you can manipulate the odd's the chance of a positive outcome is higher. but still it will only be worthwhile in moderation.

I guess the reasoning would be that they just need to be lucky once to be able to be up more than they started and then quit, but I know that isn't the reality. I guess you could sum up the reason as impulsiveness/bad judgement.

From: NKN
Gambling is fine if it's done in moderation, if you're good at math, you can predict the chances of what it will turn out to be, greatly wavering the odds in your favor.

No, simply not possible. If a gambling host does a 55 x 2 for 2 event that has an equal chance of happening, no matter what kind of tricks you try you can't possibly overcome the law of probability. I specifically mentioned Gambler's fallacies. Future events can never be affected by past occurrence, that's in fact common sense yet so many gamblers fail to see it.

From: BMWxi
I guess the reasoning would be that they just need to be lucky once to be able to be up more than they started and then quit, but I know that isn't the reality. I guess you could sum up the reason as impulsiveness/bad judgement.

Almost no gambler actually quit after winning. Human psychology dictates that a gambler's greed will haunt him until he goes from win to lose to eventual bankrupt.

From: riwu
No, simply not possible. If a gambling host does a 55 x 2 for 2 event that has an equal chance of happening, no matter what kind of tricks you try you can't possibly overcome the law of probability. I specifically mentioned Gambler's fallacies. Future events can never be affected by past occurrence, that's in fact common sense yet so many gamblers fail to see it.


Almost no gambler actually quit after winning. Human psychology dictates that a gambler's greed will haunt him until he goes from win to lose to eventual bankrupt.

If he rolls 100 100 times in a row, the chances are in your favor that he won't roll a 100.

I don't know who you are gambling with, but every host I've seen does H/C X 2, FF X 2, or FP X 2. All 3 of these games favor the host. In H/c, you pick hot colors(red, orange, yellow) or cold colors(blue, purple, pastels), and in FF, you are forced to choose cold. There is also one other flower, mixed, which results in auto-host-win. The odds, because of this "mixed flower" become:

Your Pick = 42%
Other color + Mixed = 58%

I know a few host friends, and they make hundreds of millions per day doing this. Yes, they could lose 20 in a row and lose it all, but the odds are so heavily in the hosts favor that this is unlikely to happen.

E: Dice are still in the game? Thought you guys were talking about flowers.

E2:

From: riwu

Even in the case of a fair gamble, such as when the gambling is 'hosted' among 2 gamblers of a 50 50 chance, the expected return would be zero. That is, you are wasting precious time for absolute zero return. And since time=money, you are practically still losing money :p

When I diced, host always took 5% commission.

From: NKN
If he rolls 100 100 times in a row, the chances are in your favor that he won't roll a 100.

No. If the probability of him rolling 100/100 is 1% (assuming the range is 1-100), then even if he roll 100/100 for 1 million consecutive times, the probability of him rolling another 100/100 is still 1%.

Assuming the dice is unbiased, past and future events are completely independent . Probability is a static concept, it doesn't change simply because of what happened in the past. It is a logical fallacy to think otherwise.

@Footy we are just constructing hypothetical situations here, and this discussion is not solely aimed at RS (though i started it because its prevalence in RS is worth investigating)

http://www.youtube.com/watch?v=_fICHWZZcns&feature=youtu.be
I was bored. Don't hate.

From: KeepBotting
http://www.youtube.com/watch?v=_fICHWZZcns&feature=youtu.be
I was bored. Don't hate.

:confused: private video

From: Footy

When I diced, host always took 5% commission.

I was hinting at staking actually. Kind of laughable that there is actually a popular staking bot at pb. Have they considered what happened when the staking bots play against each other? It's a zero sum game, there are no product of any economic value generated. One person's win is the other's loss. Or are they thinking that a staking bot can actually do better than a legit (who would check for pid etc)..?

It's really hysterical how gullible pb users are. But that may be quite hypocrite considering how many users here would also use a staking script if it were to be made…

From: riwu
:confused: private video


I was hinting at staking actually. Kind of laughable that there is actually a popular staking bot at pb. Have they considered what happened when the staking bots play against each other? It's a zero sum game, there are no product of any economic value generated. One person's win is the other's loss. Or are they thinking that a staking bot can actually do better than a legit (who would check for pid etc)..?

It's really hysterical how gullible pb users are.

Sorry, try it now. Accidentally hit priv instead of unlist.

This was a good read, and I agree with it all.

I just wrote a quick script to demonstrate. Make sure you are using lape (script-> Interpreter -> lape).

Law of averages people. Sure you might have your joyous win streak, but you will lose in the long run.

Run the script. It simulates your 45% chance of winning when playing the old 55x2 rs dice game, and can be easily edited to fit the probabilities of most gambling games. I can guarantee that your chance of winning is within .1% of 45%. Also, your highest win streak will be lower than your highest lose streak.

It is also important to note that every event is independent from all previous (and future) events. Just because someone loses 1000 times in a row, that doesn't mean that person has a better chance of winning on their next try.

program Gamble;
var
  loop, zeros, streak, bestStreak, lose, loseStreak : Integer;

begin
  for loop := 0 to 10000000 do
    if random(10) <=4 then
    begin
      lose := 0;
      inc(zeros);
      inc(streak);
      if streak > bestStreak then
        bestStreak := streak;
    end else
    begin
      streak := 0;
      inc(lose);
      if lose > loseStreak then
        loseStreak := lose;

    end;

  cleardebug;
  writeln('wins percent: ' + ToStr(zeros / 10000000 * 100));
  writeln('highest win streak:' + ToStr(beststreak));
  writeln('highest lose streak:' + ToStr(losestreak));
end.             

tl;dr don't gamble.

From: Nebula
This was a good read, and I agree with it all.

I just wrote a quick script to demonstrate. Make sure you are using lape (script-> Interpreter -> lape).

Law of averages people. Sure you might have your joyous win streak, but you will lose in the long run.

Run the script. It simulates your 45% chance of winning when playing the old 55x2 rs dice game, and can be easily edited to fit the probabilities of most gambling games. I can guarantee that your chance of winning is within .1% of 45%. Also, your highest win streak will be lower than your highest lose streak.

It is also important to note that every event is independent from all previous (and future) events. Just because someone loses 1000 times in a row, that doesn't mean that person has a better chance of winning on their next try.

program Gamble;
var
  loop, zeros, streak, bestStreak, lose, loseStreak : Integer;

begin
  for loop := 0 to 10000000 do
    if random(10) <=4 then
    begin
      lose := 0;
      inc(zeros);
      inc(streak);
      if streak > bestStreak then
        bestStreak := streak;
    end else
    begin
      streak := 0;
      inc(lose);
      if lose > loseStreak then
        loseStreak := lose;

    end;

  cleardebug;
  writeln('wins percent: ' + ToStr(zeros / 10000000 * 100));
  writeln('highest win streak:' + ToStr(beststreak));
  writeln('highest lose streak:' + ToStr(losestreak));
end.             

tl;dr don't gamble.

I won on that 7/10 times. I win

From: KeepBotting
http://www.youtube.com/watch?v=_fICHWZZcns&feature=youtu.be
I was bored. Don't hate.

Does no one care?

From: NKN
I won on that 7/10 times. I win

I really do hope you're joking…

From: KeepBotting
Does no one care?

From: Nebula
I really do hope you're joking…

From: NKN
I won on that 7/10 times. I win

Binomial distribution.
X~B(10000001, 0.4)
P(X > (10000001 / 2) ) = 0.

Your chance of winning on that is 0 corrected to 13 decimal places lol.
The theoretical chance of winning may be 0.000000000000000000000001

Try some binomial calculators online.
https://www.google.com.sg/webhp?sourceid=chrome-instant&ie=UTF-8#hl=en&safe=off&output=search&sclient=psy-ab&q=binomial%20calculator&oq=&gs_l=&pbx=1&fp=6876a2dc191226e3&bpcl=38625945&bav=on.2,or.r_gc.r_pw.r_qf.&biw=1366&bih=643


Sorry KeepBotting we prefer written arguments :p
It's easier/faster to read than to listen.

It's impossible to predict your chance of winning. It's completely random.

From: KeepBotting
It's impossible to predict your chance of winning. It's completely random.

Completely possible. The chance/probability of landing head on a fair coin is a static 50%. Nothing else.

I'm assuming what you are referring to is the actual outcome. Yes, that is completely random. This is where central limit theorem comes in. As sample size approaches infinity, actual outcome approaches expected outcome. In Nebula's script Random is rolled 10000001 times which therefore makes the probability of winning more than half out of 10000001 games very close to 0.

I'd think our gauss mouse inventors would understand this too.

So many numbers…tiny brain…hurt…

Anyway, I gamble, I win. My logic. Boom :/

From: riwu
Completely possible. The chance/probability of landing head on a fair coin is a static 50%. Nothing else.

In a perfect world, yes. But go ahead and flip a coin 100 times and average it out. I guarantee it won't be exactly 50%.

From: Rezozo
So many numbers…tiny brain…hurt…

Anyway, I gamble, I win. My logic. Boom :/

Unless you have gambled for an infinite period of time, actualizing logic based on experiences is like saying an increase in total murder causes an increase in population, just because throughout history number of murders have increase, and so has total world population.

Also note that correlation does not imply causation.

From: KeepBotting
In a perfect world, yes. But go ahead and flip a coin 100 times and average it out. I guarantee it won't be exactly 50%.

Then toss another 1000 times, it will be closer to 50 than before.
Then another 1million times, it will be even closer than 50.
Another 1 trillion times, it will be really close to 50.
Another infinite times, it will be 50.

Of course infinity is a concept, nothing can ever reach infinity. But in mathematical proofs it is a satisfactory analogy.

Also in my first statement, i was referring to you using the term chance, which is actually a mathematical concept that is indeed static at 50%.

From: riwu
Unless you have gambled for an infinite period of time, actualizing logic based on experiences is like saying an increase in total murder causes an increase in population, just because throughout history number of murders have increase, and so has total world population.

Also note that correlation does not imply causation.



Then toss another 1000 times, it will be closer to 50 than before.
Then another 1million times, it will be even closer than 50.
Another 1 trillion times, it will be really close to 50.
Another infinite times, it will be 50.

Lol, and yet I still win :D Im happy enough with that, math makes my head hurt :(

On the other hand, my logic also includes that if you get 3 E's together on the SAT's…You know you're screwed.

From: riwu
Unless you have gambled for an infinite period of time, actualizing logic based on experiences is like saying an increase in total murder causes an increase in population, just because throughout history number of murders have increase, and so has total world population.

Also note that correlation does not imply causation.



Then toss another 1000 times, it will be closer to 50 than before.
Then another 1million times, it will be even closer than 50.
Another 1 trillion times, it will be really close to 50.
Another infinite times, it will be 50.

Of course infinity is a concept, nothing can ever reach infinity. But in mathematical proofs it is a satisfactory analogy.

Also in my first statement, i was referring to you using the term chance, which is actually a mathematical concept that is indeed static at 50%.

program new;

var heads, tails, timesflipped:integer;

procedure flipcoin;
begin
case random(2) of
0:
begin
WriteLn('The coin landed on heads.');
Inc(Heads);
Inc(TimesFlipped);
end;
1:
begin
WriteLn('The coin landed on tails.');
Inc(Tails);
Inc(TimesFlipped);
end;
end;
end;

procedure proggy;
begin
ClearDebug;
Writeln('********************************************');    //Sexy progress report is sexy.
Writeln('*                 coin.simba               *');
Writeln('********************************************');
Writeln('Heads: ' + IntToStr(Heads) + '.');
Writeln('Tails: ' + IntToStr(Tails) + '.');
WriteLn('********************************************');
end;

begin
repeat
flipcoin;
until Timesflipped = 100;
proggy;
end.  

Run that a couple hundred times. Tell me how often it's exactly 50.

EDIT: I'm running this until Timesflipped = 10000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000; (a googol) and I'll get back to you guys in the morning.

You know, in programming there's something call a loop that does the job for you ;)

program new;
var
  i, n: Integer;
  win, lose: Extended;

begin
  n:= 1000000;

  for i:=1 to n do
    if Random(2) = 1 then
      win:= win + 1
    else
      lose:= lose + 1;

  win:= Round((win/n)*100);
  lose:= Round((lose/n)*100);
  writeln('% wins: ' + ToStr(win));
  writeln('% loses: ' + ToStr(lose));

  if win = lose then
    writeln('Wow now i finally understand what''s call a large sample size');
end. 

From: riwu
You know, in programming there's something call a loop that does the job for you ;)

program new;
var
  i, n: Integer;
  win, lose: Extended;

begin
  n:= 1000000;

  for i:=1 to n do
    if Random(2) = 1 then
      win:= win + 1
    else
      lose:= lose + 1;

  win:= Round((win/n)*100);
  lose:= Round((lose/n)*100);
  writeln('% wins: ' + ToStr(win));
  writeln('% loses: ' + ToStr(lose));

  if win = lose then
    writeln('Wow now i finally understand what''s call a large sample size');
end. 

What's the advantage of a for…do loop over a case statement? Lol I just picked the first way I could think of.

From: KeepBotting
What's the advantage of a for…do loop over a case statement? Lol I just picked the first way I could think of.

for loop has nothing to do with a case statement and neither is it a replacement for it. I replaced your case statement with a simple Boolean check because a case statement is only meaningful when checking for 3 or more conditions.

From: riwu
for loop has nothing to do with a case statement and neither is it a replacement for it. I replaced your case statement with a simple Boolean check because a case statement is only meaningful when checking for 3 or more conditions.

Oh. Duh.

I'm going to bed >.< I'll let you know how those 10,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000 coin flips turn out.

From: riwu
for loop has nothing to do with a case statement and neither is it a replacement for it. I replaced your case statement with a simple Boolean check because a case statement is only meaningful when checking for 3 or more conditions.

I have to go with riwu on this one. If's are more efficient too.

KeepBotting, you would literally have to run the script that many times, remembering the number for each, as the script will delete the memory after each run. Using the for loop will keep that.

God my head hurts. Google is so long!

From: KeepBotting
Oh. Duh.

I'm going to bed >.< I'll let you know how those 10,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000,*000 coin flips turn out.

Huh unless i missed out something your script only does 100 flips.
Mine does 1000000 flips and it takes few seconds to execute so you won't have to wait too long ;)

From: riwu
Huh unless i missed out something your script only does 100 flips.
Mine does 1000000 flips and it takes few seconds to execute so you won't have to wait too long ;)

I changed it to a googol. I want to see what the ratio will be.

From: KeepBotting
I changed it to a googol. I want to see what the ratio will be.

Uhm if it's already at 50% with just 1000000, are you expecting it to suddenly deviates from 50% as n increases? : o

Also how did you change it to googol? i'm extremely interested in that amazing processor of yours that utilize some 512-bit integer type. :rolleyes:

From: riwu

Also note that correlation does not imply causation.

And back to rs gambling, the gambler is at a further disadvantage because of the possibility that the host will take their money regardless of the outcome.

Lol who cares.

From: riwu
No. If the probability of him rolling 100/100 is 1% (assuming the range is 1-100), then even if he roll 100/100 for 1 million consecutive times, the probability of him rolling another 100/100 is still 1%.

Assuming the dice is unbiased, past and future events are completely independent . Probability is a static concept, it doesn't change simply because of what happened in the past. It is a logical fallacy to think otherwise.

@Footy we are just constructing hypothetical situations here, and this discussion is not solely aimed at RS (though i started it because its prevalence in RS is worth investigating)

What riwu says is true, these are independent events, even if it feels like the probability is in your favor, it's not.

Is this in relation to H/C x2/3/4/5/6/7/8/9/10 hosts on RuneScape?

From: riwu
Uhm if it's already at 50% with just 1000000, are you expecting it to suddenly deviates from 50% as n increases? : o

Also how did you change it to googol? i'm extremely interested in that amazing processor of yours that utilize some 512-bit integer type. :rolleyes:

Maybe.

And I just did this

begin
repeat
flipcoin;
until Timesflipped = 10000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000;
proggy;
end.

And wtf? I have a 32-bit computer.

From: KeepBotting
Maybe.

And I just did this

begin
repeat
flipcoin;
until Timesflipped = 10000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000;
proggy;
end.

And wtf? I have a 32-bit computer.

32-bit signed integer can't go over 2^32/2-1 so that would only result in integer overflow. Good luck with your experiment anyway :rolleyes:

From: riwu
32-bit signed integer can't go over 2^32/2-1 so that would only result in integer overflow. Good luck with your experiment anyway :rolleyes:

It actually crashed Simba overnight :C

If you don't have sufficient RAM it will.

From: riwu
If you don't have sufficient RAM it will.

Must have been it :X back to hosting! :D

Aww i thought you have more interesting theories to argue with me…

From: riwu
Aww i thought you have more interesting theories to argue with me…

Well if one loses a bet, one should always rebet twice what they lost, no?

From: KeepBotting
Well if one loses a bet, one should always rebet twice what they lost, no?

No. That's another common fallacy. Unless you have infinite wealth and time.
You will very likely get cleaned before you can even double your wealth.

So you're saying there's a chance…..

But really it's the same reason why people gamble in general. It's not about the odds it's about the chance of winning big

The chance is winning big is smaller than the chance of losing big. It's an unfair deal.

From: riwu
No. That's another common fallacy. Unless you have infinite wealth and time.
You will very likely get cleaned before you can even double your wealth.

No risks = no rewards though.

I really don't understand why people simply don't box with flat 99s and voluntarily turn their 50/50 to 45/55 or in some cases 40/60. Simply dumb.

I'm sure someones gonna pull the PID card over here but truth is that PID matters little in boxing, weps on is a different story.

Let's calculate using a googol!

procedure inc_str(var num : string);
var
  i9, s : string;
  I : Integer;
begin
  s := num;

  while s[Length(s)] = '9' do
  begin
    i9 := i9 + '9';
    Delete(s, Length(s), 1);
    if Length(s) = 0 then
    begin
      s := '0';
      Break;
    end;
  end;

  I := StrToInt(s[length(s)]) + 1;
  Delete(s, Length(s), 1);
  s := s + IntToStr(i);

  while Length(i9) > 0 do
  begin
    Delete(i9, Length(i9), 1);
    s := s + '0';
  end;

  num := s;
end;

procedure div_str(var num1, num2 : string);
begin
  // Don't bother.. -.-
end;

var
  number, wins, losses : string = '0';

begin
  repeat
    inc_str(number);
    if random(2)=1 then inc_str(wins)
    else inc_str(losses);
  until(number = '10000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000');
  WriteLn([number, wins, losses]);
end.

E: Calculates ~200k flips a second. Limit is when the length of the string reaches 2147483647 (a 9, followed by 2147483646 9's).
E2: Will have done calculated to a googol in about 9.51 * 10^88 years :>
E3: Let it run for some minutes, I'm now at a 50.01% win rate with 118667252 flips! Maybe Simba's Random() procedure isn't fully random…

From: Zyt3x
Let's calculate using a googol!

procedure inc_str(var num : string);
var
  i9, s : string;
  I : Integer;
begin
  s := num;

  while s[Length(s)] = '9' do
  begin
    i9 := i9 + '9';
    Delete(s, Length(s), 1);
    if Length(s) = 0 then
    begin
      s := '0';
      Break;
    end;
  end;

  I := StrToInt(s[length(s)]) + 1;
  Delete(s, Length(s), 1);
  s := s + IntToStr(i);

  while Length(i9) > 0 do
  begin
    Delete(i9, Length(i9), 1);
    s := s + '0';
  end;

  num := s;
end;

procedure div_str(var num1, num2 : string);
begin
  // Don't bother.. -.-
end;

var
  number, wins, losses : string = '0';

begin
  repeat
    inc_str(number);
    if random(2)=1 then inc_str(wins)
    else inc_str(losses);
  until(number = '10000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000');
  WriteLn([number, wins, losses]);
end.

E: Calculates ~200k flips a second. Limit is when the length of the string reaches 2147483647 (a 9, followed by 2147483646 9's).
E2: Will have done calculated to a googol in about 9.51 * 10^88 years :>
E3: Let it run for some minutes, I'm now at a 50.01% win rate with 118667252 flips! Maybe Simba's Random() procedure isn't fully random…

Wow. My coin.simba being way slower that that, I'd say my grandkids would be dead by the time it finished.

From: Zyt3x
Let's calculate using a googol!

E: Calculates ~200k flips a second. Limit is when the length of the string reaches 2147483647 (a 9, followed by 2147483646 9's).
E2: Will have done calculated to a googol in about 9.51 * 10^88 years :>
E3: Let it run for some minutes, I'm now at a 50.01% win rate with 118667252 flips! Maybe Simba's Random() procedure isn't fully random…

A perfect 50.00000000… is only 100% certain when number of flips reaches infinity, as you can never reach infinity you can never achieve 100% certainty of a perfect 50% :p But you can see how the standard deviation lowers as n increases.

Ignore the green line. The yellow, red and blue line all have a mean (i.e. expected outcome) of 0. But they differ in their variance. As variance decreases, the graph becomes 'slimmer', and when variance approaches 0, the graph approaches a straight vertical line, that is, approaches a 100% certainly of the mean.

Honestly does not matter what you say about it, if people believe they have a chance to "make it" they will go all in; eyes closed.

You profit by walking when you're up. Expecting to be up at some point is reasonable, thus people do it.

Gambling in general never leads to any good, but if you are an adult in a legal setting and you want to, go ahead. I won't stop anyone. But I believe gambling in RS is plain silly because the odds are against you and older hosts are simply taking advantage of younger players thereby ruining their RS experience.

From: vaporak10
Gambling in general never leads to any good, but if you are an adult in a legal setting and you want to, go ahead. I won't stop anyone. But I believe gambling in RS is plain silly because the odds are against you and older hosts are simply taking advantage of younger players thereby ruining their RS experience.

Time for me to start a gambling clan and take advantage of the young. Lol.

they thrive because some people just love the feeling of getting their bank cleaned.

From: KeepBotting
In a perfect world, yes. But go ahead and flip a coin 100 times and average it out. I guarantee it won't be exactly 50%.

:bart: You shouldn't guarantee that it won't be 50% since it still could happen however unlikely it is.