/sci/ - Science & Math

>>11372225
you sound liek an addict

>>11372226
Addicted to what, basic math?

>>11372225
that's Martingale: https://en.wikipedia.org/wiki/Martingale_%28betting_system%29
it doesn't work over the long run if you have finite bank roll.

>>11372225
You'll still lose money on average no matter what strategy you have.

>Unless you lose every single 49/100 roll for eternity,
Oh so you have infinite money to bet? Then why gamble at all?

>>11372225
you mean something like that? I still lose after 10 bets on average, please tell if i made some error

import random

def roulete(money):
bet=1
sumOfBet=1
i=0
while money>=0:
money-=bet
result=random.uniform(0,1)
print(result)
if result<=0.5:
bet=sumOfBet*2
sumOfBet+=bet
else:
money+=(2*bet)
sumOfBet+=bet
bet=1
i+=1
if i%100000: print(money)
print(i)
roulete(500)



>>11372279
https://pastebin.com/8xfKY5g9

>>11372279
> I still lose after 10 bets on average
That's because you are calculating the EV of roulette, which of course is negative, instead of considering the strategy

>>11372257
>>11372263
Since the strategy is sound, anyone who understands basic math should lend you money to execute it.

>>11372291
no, the strategy is not sound. and no sensible person with a finite bank roll would lend you money for it. just read the wikipedia article I linked. it explains how you're making a reasoning error.

>>11372297
The article literally says the strategy is sound with an infinite bankroll, which functionally exists since humans have concepts of credit and lending. Unless someone risks all of humanities worth and loses, everyone will gain by pooling their money to one person making the bets, with the promise of getting their money back with small interest.

>>11372263
You're wrong
Using OPs strategy, with an upper limit of 15 fails in a row before you reset:
You start with 1$
You earn on average 2.55 dollars per round of increasing bets
Your last bet in a round is on average 278.36 dollars
One of the highest last bets for a round would be 9.6million

Do the same with an upper limit of 10:
start 1 dollar
earn 24c on average
average last bet for a round 38 dollars
one of the highest last bets is 39k

Do the same with an upper limit of 5:
start 1 dollar
earn 0.2c on average
average last bet for a round 5 dollars
one of the highest last bets is 162 dollars

Essentially the system should work, but you'd be better off doing a legitimate profession if youre poor
Or being limited by the roulette table bet limit if you can afford it

>>11372304
kek no. and even if that would work, it still wouldn't work because you just described inflation.

>>11372279
>debug mode enabled.

>>11372297
If you believe otherwise, model the bets. What are the chances of reaching a bet, if you are starting with a $1 bet, that requires you to bet all of the world's combined GDP, and losing? On average, how much money does one make? This is not that challenging to do. The world GDP is $80,934,771,028,340

>>11372312
just imagine what would happen.
someone starts betting that way, and wins like $10k or whatever. because he won, he wants to repeat that, over and over. but what happens when he's real "unlucky" x times in a row? everyone's money's gone? everybody just dies?

also, look up inflation.

>>11372225
Great advice. Remember to take out student loans and home equity loans to fund it, 500k should be doable. Assuming the improbable happens and you cannot double your money but lose it, bankruptcy laws exist. Let's face it, unless you were born rich, you have nothing to lose by trying this. Let's face it, none of us are making out of this world alive, so wrecking yourself financially doesn't really matter.

>>11372321
The chances of losing all the worlds money are FAR lower than any number of apocalypse scenarios, so it's not a valid consideration.

>>11372329
the chances are only lower if you're going for a very low amount of money to win. that is, if you limit yourself to 10 runs or whatever.
if you increase the number of runs, you increase the chance of ruin.

seriously, dude, look up why the martingale betting system doesn't work, and look up inflation.

>>11372291
I simulated it again with improved strategy: if i lose more than twice in a row i change bet to 1$, i stop playing i earn twice amount of money i started with. I am doing 1000 games with 500k starting money, results Wins: 482, Loses: 518
Thoughts?
https://pastebin.com/LGDeJDPN

>>11372225
Basically this is Martingale but you triple your bets after the second instead of doubling?

No progressive betting system can be profitable.

Each bet on its own has a negative return. You can't sum a group of negative values and end up with a positive.

>>11372336
>seriously, dude, inflation.
How stupid can you be? Debt and inflation are man-made concepts driven purely by expectation. Nobody has perfect information and exploiting the system is not going to result in inflation. The theories related to inflation have been debunked anyways, because money is purely fiat in 2020 anno domini, meaning it's value is purely the result of might enforcing it. That is why Banana Republics experience massive inflation whereas the United States, having the largest military and reserves of nuclear warheads, experiences inflation almost exclusively in a predetermined way so as to effectuate monetary policy for economic purposes. Regardless, in this case it would be a zero-sum game, meaning the casino is not printing money but losing it.

>>11372344
>Each bet on its own has a negative return.
You are ignoring regression to the mean.

I don't know why people struggle to grasp this, it literally says on the wikipedia article linked above the strategy is valid unless you have a finite bankroll.

>>11372350
>How stupid can you be?
how ironic, coming from a fool who thinks he's found a way to free money by misunderstanding ev and the differences between finite and infinite values.

>>11372351
>regression to the mean.
You can't actually be this stupid. The mean return is still negative.

infinite money is worth 0 anywais

>>11372356
Why the ad hom, no need to get riled up by basic math that you don't understand. I am really perplexed that people can't grasp this on a board ostensibly for people interested in math.

The ONLY time the strategy loses is if you lose every single roll forever. Barring that, every single time you roll, you have a chance to finish ahead and cash out. Since we have access to computers, we can model a curve on the typical earnings. Since the chances of burning through all the money in the world are lower than any number of other scenarios that would have even worse consequences (meteor strike, disease, etc.) it's not even worth entertaining that possibility.

>>11372363
ad hominem is justified in this case. i sincerely hope you are trolling

my final post on this matter

>>11372367
Shit, the guy who couldn't understand basic math is taking his ball and going home, what are we going to do thread?!

>>11372344
>sum a group of negative values and end up with a positive.

orly?

-(1+2+3+4+5...)=-1/12

>>11372225
Go to a casino and try this. You will lose all your money.

>>11372407
>I don't care if it's mathematically sound I am going to dismiss it casually because I can't understand it

>>11372353
I'm not the same poster, moron. I disagree with his math and its sustainability but to pull the DUDE INFLATION card is absolutely retarded. You might get away with it in humanities but not here.

>>11372393
Come on, this result is derived from an analytical extension of the zeta function, and not literally the result of the infinite sum.

>>11372431
he's talking about increasing the total amount of money in the world without creating anything of value. that is inflation.

>>11372225
Bet limits exist.

>>11372421
The reason it doesnt work is because of exponentials. Your gain rate will be extremely slow and it only takes one bad streak to lose all your capital. Your odds of being wrong in roulette are higher than a coinflip due to zeros. Go flip a coin for 5 minutes and see what the highest streak you get to is.

>>11372443
And also outside minbet will be ~10$ and your maxbet will probably only allow a losing streak of 6-8. You will lose all your money.

It's all fun and games until you lose 17 times in a row, which WILL happen.
been there, done that. don't be retarded

One of my favorite expression is "One test is worth a thousand opinions"

Go to a casino, bring some money and a notepad. Keep track of your profit and losses over a long period. Post your results.

>>11372439
If nobody knows, inflation cannot ensue.

>>11372610
depends on the size of the inflation.

>>11372526
>>11372605
>you can disprove easy to simulate logic applied to huge samples with a sample size of one
Why are you on this board if your pea brain comes up with posts like that?

>>11372624
>implying your simulation is guaranteed to be accurate
>t. failed theorist

>>11372624
>100 instance graph shows 40% profit while being 1 loss away from losing everything once and 2 losses away from losing everything at another point
Im gyessing you also failed to account for zeros in this sim? I trust youre not too dumb to understand that this graph represents exactly why the martingale doesnt work.

>>11372605
>a coinflip is 50/50
Heh, guess that's what you "scientist" types say. Well, I gotta say, over my long life, I've seen many scientists come and go, along with their ever changing "theories." One thing I've never seen disproved though is a piece of wisdom my grandpa gave me. You never really know til you try yourself.
>flips coin
>it's heads
Where's the 50/50 anon? Looks like it's 100% heads to me.

>>11372225
This doesn’t work because the negative value of the potential loss in the next round of betting keeps the expected value negative.
If you had infinite money, you could continue infinitely if you get an infinite string of loses, with the expected infinite loss associated balancing this out whatever finite gains you’d get.
There’s no such thing as infinite money though, and so you just go bankrupt after the bet inflates to the size of your bank account (or whatever the max bet limit is).

>>11372617
Depends on the size of the military, control over media and public perceptions actually.

>>11372702
that just sounds like slavery with extra steps

>>11372363
If you had infinite money, you can win everything even the slots.
But why would anyone care about gambling if they already have infinite money?

>>11372959
You already have infinite money, because you have the knowledge and ability to prove the efficacy of the Martingale system. You can borrow against that when you need more liquid capital.

So while you maggots argued about theory and other bullshit, I've spent the past hour placing $1 bets. Started with $1000 and I now have $1585.

>>11372714
Sounds like real life, and what actually determines monetary value.

>>11372225
>he thinks he's figured out how to beat a casino
Martingale betting doesn't work because there are table minimums and maximum bets

Bet on the 1/50 one that multiplies your money by 50.

>>11373174
Keep posting updates. I want to laugh at you when you lose it all.

>>11373174
So whats the table max on that game? Eventually you will hit a run that hits the max & results in you losing a ton.

>>11373210
Let him keep playing please, he will learn on his own
Youre doing great op

>>11373205
>>11373210
>>11373230
>get handed a mathematically sound way to make money (and simple at that)
>NO, YOU CAN'T JUST DO THAT! YOU HAVE TO LOSE!
If anything, this says a lot about the culture we live in. You, on a board that supposedly is for people interested in math, not only deny the math, but wish misfortune on anyone that uses it for general benefit. You want to be hopeless victims of the financial system, for anyone not to be actually offends you. Take a step outside yourselves and breathe for a second.

>>11373205
That won't be happening because I cashed out. Guess you'll have to find some other way to amuse yourself whilst I decide what to do with the $585 I scored.

>>11373210
It would have let me bet the whole $1000, not sure on the maximum though. Longest losing streak was 6, which meant putting down $486 on the 7th. If that had missed I wouldn't have had enough to cover the next bet of $1458, but it won. Most losing streaks came good on the 4th bet.

>>11372225
>What is the law of big numbers

You are addicted.

>>11372225
this has a name, people came up with it 50+ years ago, it doesn't work and you're a retard

>>11373174
Based. Let the good times roll.

>using millions of dollars to make $1
fucking retard, no wonder you're a poor nigger. what happens when you hit the table limit you spastic retard? what about when you reach 10 million dollars? you got 10 million dollars all converted into chips and ready to play?

>>11373348
>to make $1
Where exactly did that come from? Do you struggle with match concepts? This system makes infinite money. It's pointless to consider a single bet, only the entire function.

>>11373361
>fucking retard, no wonder you're a poor nigger. what happens when you hit the table limit you spastic retard? what about when you reach 10 million dollars? you got 10 million dollars all converted into chips and ready to play?

>>11373251
>most losing streaks come good on the 4th bet

>>11373365
Who are you quoting? If you're asking about that anon's post, it's mostly ad hom, with a few meaningless fallacies appended on the end.
>the table limit
There is nothing binding you to a single table or even a single game.
>10 million in chips
In those dollar amounts ($1 mil+) they are typically plaques, or table vouchers, at least that's how they do it in Macau (in HKD) when I have played for those amounts.

>>11373368
Statistically, this is true if you do a limited montecarlo simulation.

>>11373388
You can’t just ignore the outliers where you get 6+ losses. Just one bad streak is required to make you go bust. >>11373251 even admitted to coming within a 50/50 chance of losing most of his money. The strategy has existed for a couple of centuries now and is widely known as a losing strategy.

>>11373249
It’s demonstrably bad when you look at the math without being a retard. We explained to OP numerous times why he shouldn’t do it; laughing at his misfortune when he inevitably burns himself after being warned is fair game.

>>11373249
>If anything, this says a lot about the culture we live in.
>You, on a board that supposedly is for people interested in math, not only deny the math, but wish misfortune on anyone that uses it for general benefit.
>You want to be hopeless victims of the financial system, for anyone not to be actually offends you.
>Take a step outside yourselves and breathe for a second.

OP I believe in you, fuck the haters.

>>11373411
The longer you play, the more likely you are to bust out. The higher the stakes, the sooner you will hit the table limit and not be able to cover the losses. So no, it's not a get rich scheme, but if you play small stakes with money you won't miss and walk away after landing a 4 or 5, you can easily make $100 in less than an hour, but you are risking many times that amount. When you land a first bet you are only doubling a small stake, but they add up. You actually want to hit a run of losses. But yes, you would be hard pushed to find a table that lets you eat 9 losses and put $13122 on the 10th. Find tables that let you bet 10c and the 10th bet is $1312 + $656 you've already bet to pay out $2624 + a bunch of 10c wins. Now ask yourself how likely it is to miss ten bets in a row. If you can't stomach that, don't do it. Simple as.

>>11373427
So do I.

gg y'all, i got stuff to do. nice playing with you

>>11372225
>all these retards arguing
>nobody simply proves their argument using expected value

>>11373798
>thinking the retard who can't even conceptualize such basic concepts as being unlucky and exponentials would understand such a value

>>11372683
>doesn't notice the y axis starts at 7000
KYS brainlet

>>11372225
>>11372225
>Roulette is beatable, even with losing odds.
I had an idea involving a neural net that predicts roulette numbers based on video footage/ball tracking. Something as simple as mapping the distance between the ball and the 0 slot at every fraction of a second would be interesting. A more advanced model could also track the deflector positions, ball speed, etc.
If it actually worked you could setup a 2 man team were one person aims a camera at the wheel and then somehow communicate the predictions to the player at the table.

It's better to just know the outcome in advance and bet on that

>>11373852

I thought bets were placed before the ball dropped?

t. Casinolet

>>11373361
Look at the chart dumpass, on any bet you are risking your entire winnings for a 1$ gain,

>>11373852
too late faggot. some mad lad already did it decades ago by counted the rotations of the roulette wheel and rotations of the ball. he got promptly banned from the casinos before he could even make his money back that he spent on buying a roulette wheel to get the calculations.

>>11373855
You can bet while the ball is still spinning in the top grove after the dealer has chucked it with his fingers, but they freeze the bets before it drops, you would have to predict it off something like pic related, also the lower part of the wheel has bumps and other ‘randomizers’. It wouldn’t work, I doubt even a google server bank could do it in real time

>>11373798
His strategy isn’t even clear, it’s some modification of martain gale, but his English is retard.
When op says “sum of all your previous bets x2” not clear if he means all your losing bets or all your bets. Martingale you’re just double down every bet, which is different than “sum of your bets times 2x.”
>>11372257
It’s not Martiangale.
>>11372225
Yes, that would work.

>>11372225
Let’s say you won your 4th bet. You go back to $1 right? It would only take 3 losses to wipe out your gains.
The question isn't “lose a 49/100 forever” the question is “lose a 49/100 3 times”. Still only a 11.6 percent chance. So I approve your system
>>11372257
It’s not martiangale

>it works because sometimes I win
Oh /sci/...

>>11373857
you are the retard, the strategy is literally doubling your total spent each time.

so:
bet 1 = 1$
b2 = 2$
b3 = 6$
b4 = 18$
etc

learn reading comprehension

>>11373887
Let’s say you won your 4th bet. You go back to $1 right? It would only take 3 losses to wipe out your gains.
The question isn't “lose a 49/100 forever” the question is “lose a 49/100 3 times”. Still only a 11.6 percent chance. That’s just the odds of losing 3 out of three, but what are the odds of losing 3 in a row out of 30 bets? It approaches 100%
The math is not simple because streaks are not independent events, it’s Discrete-time Markov chain.
>>11372257
It’s not martiangale

>>11373927
You are right, I found ops wording confusing.

>>11372225
>mentally ill zoomer thinks he figured out what people professionally gambling all their lives haven't
this is why nobody likes you, and also part of the reason you can't get laid. it's time to change it up

>>11373927
it would work then

>>11372225
1. Take $1000 to casino to play this out
2. Post vid of you losing $1000
3. Stfu

>>11372441
>>11373183
>>11373210
these are the REAL reasons OPs otherwise flawless strat would not work in real life (except if a casino was retarded enough to not have ceilings)
the autists sperging about "muh infinite money" are missing the mark.
Infinite money was a given condition in OPs strat

>>11373474
For clarity in this explanation, a game is a streak of losses until a win OR 10 losses in a row. A loss is hitting the 10 losses, anything else is a win.

Using the example of 10c starting bet and an upper limit of 10 bets, your expected value is slightly higher than -$0.25.
Around 0.13% of games (one streak of losses until a win) you will go bust. This means you have almost a 50% chance of going bust after playing 544 games.
Your expected winnings at this point is $1231.61 (assuming you win every game and the winnings are in proportion to the weighting of each win percentage), with the loss in the final round being $1968.30. You’ll need to win 870 games to reach the point where your winnings will protect you from this loss, which will only happen around a third of the time.

I can post a pic of the excel sheet I did these calculations on tomorrow, or some statfags can check these numbers. You’re advocating for what is a gamble that isn’t in your favour (I know, shocking at a roulette table). This ‘strategy’ is just a way of blindingly yourself against the odds being stacked against you.

>>11372225
Fact: No negative EV game can ever be won through any conceivable finite strategy. This is a known result since game theory was first conceived and you can prove it yourself using Kakutani's fixed point theorem.

math druggies are so hilarious
all they do is cook in their head mothods to gain the the system and make money out of thin air

obsessing over betting games is what genetic trash does

>>11372235
to gambling dumbshit

>>11373184
>not betting on all of them so you always win

>>11374109
>the autists sperging about "muh infinite money" are missing the mark
no, because gambling strategy only makes sense in terms of finite bank. if you have an infinite bank you don't need to gamble.

>>11374656
this