Monte Carlo methods – Why it's a bad idea to go to the casino • Hacker News Board

Re: Monte Carlo methods – Why it's a bad idea to go to the casino

#71

Post by rocqua » Fri, Jul 05, 2019, 6:39 AM UTC

virgilp wrote: post #58

RomanBob wrote: post #47

Earlier quoted context omitted.

This is a common misconception about why casinos win. That 'casinos have infinite bankroll so they cannot lose' isn't actually an argument because casinos have limited bankroll. If they would allow me to play a +EV game I can crush them even if they had infinite bankroll by varying my betsize. The 'random walk' argument only works for fixed bet size.

I actually made a simulation of this because it seems obvious - you can actually double your bet each time you lose, and this basically guarantees that you eventually win each time! Turns out, you have to go surprisingly high for this to be true/effective on the long run. And the casino will cut you off way below the limit (i.e. there's of course a maximum bet). Plus, with this strategy, each win is a small one... bu…

The expected value of that strategy is still negative. You get a high chance of winning 1 dollar, with a low chance of losing everything. Only the chance of losing everything isn't low enough.

Re: Monte Carlo methods – Why it's a bad idea to go to the casino

#72

Post by rocqua » Fri, Jul 05, 2019, 6:41 AM UTC

mongol wrote: post #69

TBurette wrote: post #50

A while ago I wrote a similar interactive article to prove that the martingale betting technique at the roulette doesn't work : [1]. The martingale is when you bet 1, double the bet each time you lose, start again at 1 when you win. The aim was to create a single page anybody could link to that proves without a doubt the abysmal odds of making money off the roulette with the martingale technique. Instead of pre-writt…

Theoretically I learned it works, but only if you have unlimited funds.

Thing is, at any limited amount of funds it doesn't work. It's not like it works better as you get more funds. It just gets more volatile.

Re: Monte Carlo methods – Why it's a bad idea to go to the casino

#73

Post by listenallyall » Fri, Jul 05, 2019, 6:44 AM UTC

screaminghawk wrote: post #29

listenallyall wrote: post #22

This article should be down-voted, the only reason it isn't is due to the clickbaity nature of discussing casino games. It's a terrible initial introduction to Monte Carlo simulations, he doesn't even discuss the possibility of seeing the distributions of the results of the games he proposes, which is essentially the most valuable reason to perform a Monte Carlo simulation.

If you read it instead as an advertisement for easylang.online it's a pretty nice introduction for the language

...and if the headline was "Learn to Program in easylang.online" it never would have sniffed the front page of HN

Re: Monte Carlo methods – Why it's a bad idea to go to the casino

#74

Post by Gustomaximus » Fri, Jul 05, 2019, 6:54 AM UTC

spaceflunky wrote: post #60

It's (relatively) easy to show the math on why casino games are impossible to win, but what about sports betting? Does the house always have the advantage? Perhaps it's slightly better for the gamer? FWIW I was looking a betting website recently [0] and they were offering DOUBLE your initial deposit for signing-up, up to $1000. So to reiterate, they are willing to give up to $1000 for FREE just for creating an accoun…

You have to bet the bonus money at least once.

I did this. Bet the bonus on some super low odds trots race to place that was as good odds as anything to get my money back and then bet the other half on a 36:1 long shot that Bernie would win the nomination.

I got my money back with the low odds Trotter, otherwise I would have bet the balance on another low odds event. Obviously didn't win the long shot.

Not sure why I did it as I don't normally gamble but generally curious and though it was a fun way to risk a small amount to potentially win a "free" holiday.

Re: Monte Carlo methods – Why it's a bad idea to go to the casino

#75

Post by denormalfloat » Fri, Jul 05, 2019, 7:04 AM UTC

WalterBright wrote: post #32

Investing: Return on Investment > 0 Gambling: Return on Investment For the casinos, it's an investment, for the customers, it's gambling.

Perhaps we could combine the two? Make a slot machine that pays out in penny-stocks, rather than chips.

Re: Monte Carlo methods – Why it's a bad idea to go to the casino

#76

Post by WalterBright » Fri, Jul 05, 2019, 7:15 AM UTC

denormalfloat wrote: post #75

WalterBright wrote: post #32

Investing: Return on Investment > 0 Gambling: Return on Investment For the casinos, it's an investment, for the customers, it's gambling.

Perhaps we could combine the two? Make a slot machine that pays out in penny-stocks, rather than chips.

We call them "online brokers" :-)

Re: Monte Carlo methods – Why it's a bad idea to go to the casino

#77

Post by chkas » Fri, Jul 05, 2019, 7:18 AM UTC

listenallyall wrote: post #73

screaminghawk wrote: post #29

Earlier quoted context omitted.

If you read it instead as an advertisement for easylang.online it's a pretty nice introduction for the language

...and if the headline was "Learn to Program in easylang.online" it never would have sniffed the front page of HN

https://news.ycombinator.com/item?id=19299760

Re: Monte Carlo methods – Why it's a bad idea to go to the casino

#78

Post by nurettin » Fri, Jul 05, 2019, 7:26 AM UTC

I made an effort to download previous draws form my local lotto website [1] And parsed them and applied a montecarlo method to see:

1. If I could find an optimal way of playing the game to maximize my profits.

2. If I could find a bias towards certain numbers.

The result was:

1. I found a way with 1/3 chance of winning big and paying back all expenses you played for the past 20 years. It is a significant investment and everyone will think you are crazy, but you have a very high chance of early retirement.

2. I found no obvious difference between the frequency of numbers.

The idea is to spread your betting numbers as far apart as possible in order to follow the normal distribution. If you reduce the overall gap between your numbers, your chances of winning is significantly reduced.

[1] http://www.millipiyango.gov.tr/sonuclar/_cs_superloto.php

Re: Monte Carlo methods – Why it's a bad idea to go to the casino

#79

Post by alexgmcm » Fri, Jul 05, 2019, 7:45 AM UTC

roenxi wrote: post #62

petermcneeley wrote: post #10

Earlier quoted context omitted.

Really dependents on your philosophy. Would you rather do error propagation using this method http://lectureonline.cl.msu.edu/~mmp/labs/error/e2.htm or would you rather perturbate your variables and resolve mean and variance. My philosophy drives me towards the computational approach over the analytical mathematical approach.

Are you sure there is even a philosophical difference there? I'm still going to use a program to do the heavy lifting of calculating the errors. I'm happy to argue that there is no philosophical difference; if both approaches are feasible and take comparable amount of time to implement then one gets the exact answer to the question at hand and the other is a poor man's shortcut for people who aren't confident in thei…

I think the value is in checking assumptions - if the analytical approach assumes a normal distribution and then you do MCM and get a vastly different result then it is probably worth checking the assumptions made by the analytical model.

Re: Monte Carlo methods – Why it's a bad idea to go to the casino

#80

Post by tapland » Fri, Jul 05, 2019, 7:52 AM UTC

eric_h wrote: post #57

TBurette wrote: post #50

A while ago I wrote a similar interactive article to prove that the martingale betting technique at the roulette doesn't work : [1]. The martingale is when you bet 1, double the bet each time you lose, start again at 1 when you win. The aim was to create a single page anybody could link to that proves without a doubt the abysmal odds of making money off the roulette with the martingale technique. Instead of pre-writt…

All of that is completely invalidated by that one time I hit 13 on the roulette table at two different casinos and quit gambling up $1500 ;)

Just a few varied anecdotes from working as a dealer and croupier for a few years:

I once saw 13 be hit 4 times in a row (and once more within my 45 min rotation at the nearby table). Don't think I've personally seen any other 4 number streaks.

A lof of people won a lot of money during the two last ones.

I've only played a few slot games online and visited a real casino as a guest once, where I bought chips for $150 and left with $5400. The goal was losing it all and at the end I was putting $1500 on the table at a time, and dipped below $500 just a few minutes before making the $5400 and leaving.

One of the best nights I had while working was a regular from the poker room coming to my Black Jack table trying to win back the $15k he was down. Had the 'chef' (senior position at the casino floor, poker equivalent would be 'brush') there talking to him since he wasn't doing well. He managed to win it back which was a great relief.