Parrondo's Paradox applied to Roulette

[toby]

Hello forum
@Toby... hhhummm. you simple don't know a jota over what are you talking about.
Another member (thanks @tangram) has answer your "mental" problem in this thread.
You simple couldn't apply this discovery on casino games.
Punto.
But you can't stand with it, and you, full of arrogant ignorance, you prefer to shot some mug over a respectable professor that his only "mistake" is not to reply some of your "bright" questions to him...

You look so stupid.

<<< "if he answers in english" >>>

what do you think a brillant investigator works on...
which is his normal language to publish his discoverys, to work every day...?
you look so racist just because he is Spanish (he born and works and lives in the kingdom of Spain)
or because the idea that you have about a "spanish" person
Wich is your level in spanish = 0.
But then again you go with the fan of s**t...

<<<"I doubt of Parrond's Paradox investigation.">>>

I've built a simple excel (I would not attach it this time because I've lost it... it comes from 8 or 10 ago...) and applying strictely Parrond's rules I'can reproduced the movement of this brownian motor.
But unfortunately you can't apply the idea on any casino game... is that simple.

Furthermore, even if I've found it I wouldn't attached in this post, just because I'would never help to an arrogant stupid person has you.

Your worst friend  
Betatester.

I'm from Argentina, we speak only Spanish here.

The letter sent to the professor were from other spanish forumers and I to debank the paradox.

If you are able to read spanish read this 7 pages written by 2 hard boiled spanish friends. nolinks://nolinks.grupojoker.com/sobre-la-paradoja-de-parrondo-t1854.html?hilit=%20parrondo

I'm not interested in argueing.

[Bugsy]

I have been studying Parrondo's Paradox.   If you are not familiar with it you can Google the subject and find many many articles that explain it and confirm that a physicist found that by alternating between two long term losing systems (no shortage of those), a long term winning system was produced.   The invention, research and applications of Parrondo's Paradox go way beyond gaming (involving controlling randomness by creating a "ratchet" effect on a molecular level) but for our purposes I will explain the gaming application in simple terms. 

We have two simple coin tossing games.   Player will bet "Heads" and win one unit when Heads comes up and lose one unit when Tails comes up.

The coins are not fair coins and each of the two games has a negative expectation.

GAME 1

One coin (Coin A) is tossed that will deliver Heads slightly less than 50% of the time, so player loses long term.

GAME 2

Two coins are tossed, one coin (Coin B) will deliver Heads 75% of the time and one coin (Coin C) will deliver Tails 90% of the time.   Coin B the favorable coin can only be tossed when bankroll is not divisible by 3, and Coin C the unfavorable coin has to be tossed when bankroll is divisible by 3.   When bankroll is not multiple of 3 player has 3 in 4 chance to win and 1 in 4 chance to lose; when bankroll is multiple of 3 player has 1 in 10 chance to win and 9 in 10 chance to lose.    

Example:  Starting bankroll is 100 units, player tosses Coin B and wins, bankroll 101, player tosses Coin B again and wins, bankroll 102 divisible by 3 so player must now toss Coin C and loses, bankroll 101, toss Coin B and win, bankroll 102, toss Coin C and lose (eventually this pattern gets broken and if Game 2 is played long enough it becomes no longer true that bankroll is divisible by 3 just a third of the time and Game 2 also becomes a long term loser. )

Parrondo found and apparently proved as confirmed by many studies, that by alternating between these two games (even randomly, but optimally by playing each game twice in a row back and forth) that over the course of 50,000 runs of 100 tosses each, the simulations showed a trend for substantial gains, because a "ratchet" like effect is achieved by alternating between the two games that keeps the profits moving higher.

It was said in many of the articles that unfortunately these games cannot be duplicated at the casino, mainly because we don't have the bets that give the player a 75% advantage (no you cannot use two dozens because you can only win or lose one unit each bet) or a bet that gives the player a 90% disadvantage (the two bets required for Game 2. )  However it seems to me we can come close, as follows.

Obviously for Game 1 we can use any of the even chances.   For Game 2 a bet that favors the player by 75% is used about every 3 times, and the unfavorable bet where the player only has a 10% chance is played the other times.   So if we add 75 + 75 + 75 -90 and divide by 4 we come up with an average bet where the player has a 33. 75% chance to win, and certainly this is pretty close to betting one dozen (33. 33%).

So my proposal for a system that duplicates Parrondo's Paradox is to bet one of the evens twice in a row and then bet one of the dozens twice in a row.

I did test this on 450 real spins (choosing to stick with just one color and one dozen throughout) and the results were, shall we say, less than spectacular.

Now I have a question for anyone who feels they have a handle on Parrondo's Paradox-- I want to know why the way I have changed it a little for our purposes at Roulette won't work, perhaps not as well, but at least slightly well (when his similar gaming strategy did)?  And/or, does anyone see a way to tweak the system better for Roulette, so it more closely duplicates Parrondo's games?



Not in roulette, is there a game in casino where Parrondo Parodox can be appllied?

[bikemotorman]

How about craps would it work with craps.
Very low house edge on pass and don't pass, 6 and 8 also very low house edge maybe it will work maybe not.

Guys do they Craps in Europe, Craps very crazy game people scream shout cry laugh.

Stuart

[betatester]

OK. I'll do it for the moderator and for the forum.
I never will show me or act in a wrong way, but every one has his limits. Toby put me in one of these.
=========================================================================
Mira compadre, decía Wittgenstein que de lo que uno no sabe no debe hablar. Aplicaté la regla.
Por los e-mails veo que tu problema es que "no te sale"... no puedes recrear esta simple simulación.
Bien aquí la tienes. No cuentes conmigo para nada más.
O corrige tus palabras y reconoce que tu tenías ni p... idea de lo que estabas hablando.
Por cierto he hecho la hoja mientras se calentaban los tortellinis.
A tu salud.
========================================================

nolinks://nolinks.eleceng.adelaide.edu.au/Groups/parrondo/link.html

Here winning means that the player receives one euro and losing means that the player loses one euro.

(this sentence comes from Parrondo's home page: nolinks://seneca.fis.ucm.es/parr/) << anda leeteló otra vez, viejo...

Let's gives some significance transportating the bets into a roulette game.

18/37 < 19/37
e=0

e=2,72%

In Parrondo'x paradoxical games [e=0,005 = 0,5%]

For game A:
Coin nº1
wining 18/19 < [1/2-e] We are going to play (and win 1 Unit) if hit in = Odd
losing 19/18 > [1/2+e] We are going to lose 1 Unit if hit in = Even + zero

For Game B
Coin nº2 or Coin nº 3

If residual k/3 <> 0

Then
Coin nº2
winning 27/37 < [3/4-e] We are going to play (and win 1 Unit) if hit in = 9 Streets (1-2-3-5-6-7-9-10-11)
losing 10/37 > [1/4+e] We are going to lose 1 Unit if hit in = 3 (4-8-12) Streets + zero

If residual k/3 = 0
Then
Coin nº 3
winning 3/37 < [1/10-e] We are going to play (and win 1 Unit) if hit in = 1 Street (the last that has hit)
losing 34/37 > [9/10+e] We are going to lose 1 Unit if hit in = 11 Streets + zero

Both games once transported to roulette bets (but always with a +1 or -1 result for the wager) gives still worse results than those predicted by Parrondo.[see the figures in the brackets]

But do you any casino with those payouts...?

Remember:
When flipping coin 1 we bet in 18 number (odd) and we are payed 1 Unit if win or lose 1 Unit
When flipping coin 2 we bet 1 Unit in 27 numbers (9 streets) and we are payed 1 Unit if win or lose 1 Unit.
When flipping coin 3 (k/3 =0) we bet 1 unit on 1 street and we are payed 1 Unit if win or lose 1 Unit.
Let's going to choose "the best" combination (alternating) btw both:

A (k +/-1)
A (k +/-1)
A (k +/-1)
B (k/3 = 0 ?) and then coin 2 or 3 (k +/-1)
B (k/3 = 0 ?) and then coin 2 or 3 (k +/-1)
A (k +/-1)
A (k +/-1)
A (k +/-1)
B (k/3 = 0 ?) and then coin 2 or 3
B (k/3 = 0 ?) and then coin 2 or 3
...
and so on...

Now let's simulate (in a very, very basic form) by the means of a spread sheet.


Download and press F9

Salud.
Betatester.

[toby]

No doubt you know much more about Parrondo than I know.

What we were looking for is this matter applied to roulette, no way.

Nice links you sent, I knew only the spanish one(seneca).

I´ll check the excel spreadsheet.

[toby]

I checked the excel and only found hits, I didn't see balances. You might have missed them.

When you play on 4 double streets is not playing 50% of the numbers, they pay like an EC but you play 24 numbers.

The right way to built the excel spreadsheet is to pay what is due, then we can add and sustract and see the balance.

[toby]

-