Turkish Casino: Gambling of the Future [pdf]

http://jeffmeyerson.com/wp-content/uploads/2015/07/TurkishCasino.pdf

by crablar • 11 years ago

75 31 11 years ago

lnlyplnt• 11 years ago

Something like this could be extremely valuable inside an organization. A lot of very valuable knowledge is locked up by employees with no incentive to reveal it.

hellofunk• 11 years ago

Turning the world's casinos into massively parallel crowd-sourced processing centers is a great idea.

KON_Air• 11 years ago

And even more wasteful then a Turkish HR department!

egeozcan• 11 years ago

I know where the name comes from but still, the word Turk being used as the name of the unit workforce while making automation by using mass manual labor is a bit inconvenient. We already have an ambiguity with the name of our country in the English language, give us a break :)

StavrosK2• 11 years ago

I thought you were just being a stickler referring to Mechanical Turk, but then I saw the PDF, and it didn't actually refer to a casino in Turkey. Wow, this is some next level confusion right there. What were the authors thinking?

crablar4• 11 years ago

I was thinking it is a catchy name.

You are turning a casino into a nexus for mechanical turks. What is complicated about it?

It's also not meant to be offensive, if anything it is a compliment to Turkish people. They are good at solving problems with no definitive solution.

TY• 11 years ago

Something similar but without the gambling aspect has been done before:

https://en.wikipedia.org/wiki/ESP_game

0xdeadbeefbabe• 11 years ago

The hard part is making work into a game. I think Michael and Jane Banks would agree this is the foremost reason they prefer Mary to other possible nannies.

dsjoerg• 11 years ago

I like the general idea — gamblers can do useful work for society — but one area that needs improvement is the kind of things that the participants will be betting on.

"IBM wants to offer the minimum amount that the candidate will accept," however the gamblers are not asked to figure that out, instead the "entire wager pool is awarded to the Turks who had the closest answers [to what IBM actually offered]".

Thus the smart gambler tries to figure out not what the candidate would accept, but what IBM is likely to offer. At this point the prediction market does not predict what the candidate will accept, but what IBM will offer.

One might be able to construct a prediction market where the gamblers incentive is to identify the lowest number the candidate would accept, but this is not it.

_up• 11 years ago

Also if the "third party/asker" decides who wins. The third party could try to colude with one player.

crablar4• 11 years ago

The bets are contrived for sure, but that was for the purpose of simplicity.

I believe the LinkedIn profile betting example is something feasible in the future, that we can't imagine very well by today's mores.

elwell• 11 years ago

These solutions provide no value to anyone.

They provide entertainment value sometimes, and for winners they provide monetary value. This is like saying watching a movie or playing football has no value. The particularities of the game may create the value. E.g., poker is a well balanced game.

crablar4• 11 years ago

Entertainment is not mutually exclusive with the utility.

fastbetshenry3• 11 years ago

If one can make this work this would be huge. One could offer casino style games at a house edge of 0% (house edge = the amount that the house wins on average on every bet). This would not only be a super compelling value proposition to gamblers, it would be a huge step towards eliminating problem gambling: A gambling addict that plays at a 0% house edge would not loose money in the long run.

I run a small bitcoin casino called fastbets.io. We should talk to see if there is a potential for a collaboration (feel free to email henry@fastbets.io).

alexjikim• 11 years ago

A gambling addict that plays at a 0% house edge would not loose money in the long run.

This isnt true because the house edge is the average return. the individual players can have a positive or negative edge depending on their skill. this would be like playing poker without a rake.

bobbygoodlatte• 11 years ago

Yep. Also this: https://en.wikipedia.org/wiki/Gambler%27s_ruin

jabgrabdthrow• 11 years ago

The first organization to figure out how to make a proper free-market prediction market platform will make a killing.

Casinos seem like a perfect way to make it happen as they already have special regulatory exemption for a transaction type BB doesn't like.

cm2012• 11 years ago

Theoretically, is this idea possible?

1) Make a casino with slot machines/etc. Connect slot machines to a stock market day trading system that acts at random on each lever pull (I.E. buying and then selling 10 seconds later, when the price is a bit different)

2) Gamblers will gamble as they usually do, but since the stock market grows on average by 7% a year, they should theoretically come out ahead.

3) Casino can take 3.5% for setting everything up.

What would make this impossible?

ddp26• 11 years ago

The gamblers will not come out ahead. The 7% return you mention requires buying and holding.

lotyrin• 11 years ago

You'd have to run it as a sort of recurring scheme where the current players are earning an expected value based on the investments of the previous years' players - start with a negative expected value until the backing fund achieves a decent annuity, then trend it toward positive.

fastbetshenry3• 11 years ago

A slot machine based on this principle exists and was presented at ICE last year (forgot the name). However keep in mind that the stock market grows at single digit percentage points per year. Slots machines only run for a couple of seconds, so there's no real profit made in that time frame. (As far as I can recall the rational for this system was of legal nature).

justzisguyuknow• 11 years ago

The whole appeal of a slot machine is that you might put in a dollar and get back a thousand. The prospect of a huge payoff is what keep s people pulling the lever.

Since that would essentially never happen on a 10-second buy/sell action on a real stock, I doubt any actual gambler would care to participate.

mason55• 11 years ago

You can make payout tables similar to what they are now but instead of having a 1% house edge there's actually a small player edge. The house then covers the player edge and makes money by the stock investment.

The issues would be

1) If you assume a 7% annual return, the return on the 10 second investment would be extremely small. 7 percent / 250 working days per year / 8 market hours per day / 60 minutes per hour / 6 ten second periods per minute = 9.7e-6 percent return every 10 seconds. On a billion dollars play through in a year you'd be looking at a $97 return.

2) I don't know what commissions are like on the volume of small, short trades you'd be doing but I imagine they would be more than $0.0000001 cents per trade which means you're losing money.

p0ckets• 11 years ago

You can use some form of leverage to allow the possibility of 1000x jackpots by using leverage - options, leveraged ETFs etc.

eru• 11 years ago

Look up how insurance companies (like Warren Buffett's!) use `float'. The mechanics are similar.

herghost• 11 years ago

One potential problem is that instead of assessing problem objectively, since the payout is based on "most popular" answer, people will choose which answer they expect to be most popular, rather than the one they necessarily believe in. This drives down the value of the crowdsourced data - because it's not real "insight", is already tainted by the very gamification you're looking to monetise!

crablar4• 11 years ago

There is certainly a metagame but there is also metagame between spam and spam filters.

It's a dynamic equilibrium that can be accounted for.

m00dy• 11 years ago

BTW, Casinos are forbidden in Turkey.

Turkish Casino: Gambling of the Future [pdf]

Related Stories