Bitcoinica: the ponzi nobody saw?

[thezerg]

The one theory that seems to strangely ring true to me is the theory that bitcoinica became an accidental ponzi.  I mean nobody else could/can offer this kind of leverage.  Maybe because they actually put some math behind their web site . 

Becoming a ponzi can be a slow process of steadily increasing minor indiscretions until you realize what it all adds up to.  Let me propose the following scenario:

Bitcoinica is making mad cash during the bear, its easy to bet against the typical bitcoin fanatic and win in a bear market.  But during the current long slow bull, bitcoinica goes underwater, betting against the highly leveraged bets from its own users.  Sure I read somewhere that bitcoinica "hedges" on Mt. Gox... but how does that work if you are offering 10-1 leverage and Gox is not? 

So what is Zhou going to do?  Just quit, when another bear market would equal more massive earnings?  Exploration of the possibilities yields help from some millionaire with a dubious reputation and a strategy. 

First, fake volatility in the market (its YOUR market after all) to get some of your depositor's money.  Look back at the "Zhou Thonged" thread https://bitcointalk.org/index.php?topic=49445.0 and you'll see some people complaining that their position got force liquidated due to some minor flash-crash that other markets did not even see...

When that's not enough we get a hack.  Then a sale to shift responsibility and finally of course full looting commences.

Put yourself in the head of a 17yo, in a culture where hackers are somewhat celebrated and tolerated (for one: http://www.thomascrampton.com/china/james-areddy-china-hackers/).  You're famous but you've got a tiger by the tail, no experience/wisdom, and now what to do?

ZT may be being emotionally honest in saying he wasn't involved (read: didn't want to be involved)  -- but he accidentally let the professional criminals in and at this point through a various forms of pressure they may have been running the show.

I joined bitcoin too late to be part of this mess.  Does this theory match your experience?  Discuss!

[humanitee]

Innocent until proven guilty. Rampant speculation like this gets us nowhere.

When the courts decide, then we talk.

[damnek]

Cypherdoc and others have repeatedly warned for Bitcoinica. See e.g. https://bitcointalk.org/index.php?topic=56350.0

[ChrisKoss]

Sure I read somewhere that bitcoinica "hedges" on Mt. Gox... but how does that work if you are offering 10-1 leverage and Gox is not?  

Nope.  There is no evidence that Bitcoinica was a ponzi scheme.  You lack of understanding of margin trading hedges does not mean they were committing a crime.

Imagine there are 100 users at bitcoinica, each has deposited 10BTC:

- 50 of them have long 100 BTC
- 50 of them are short 100 BTC

That means the total of all the long positions is 5000 BTC, but bitcoinica only has 1000BTC on deposit.  How is this possible?  Internal order matching.

Because the NET bitcoinica position is 0, when one person loses money, it goes directly into another customer's pocket.  Their opposing trades effectively cancel each other out.

What if its not balanced? Bitcoinica hedged the net position of its customers.  So if there was a total of 1000BTC long positions and 500 BTC short positions, Bitcoinica would use their USD reserves to purchase 500 BTC on MTGOX.   If they ran out of USD or BTC (turned it all into the other currency to hedge), they would put up the * (lovingly called the starfish), on the side of the trade that they did not have the funds to hedge.

Bitcoinica's had a lot of problems, but there is no reason to suspect them of being a ponzi.

Innocent until proven guilty. Rampant speculation like this gets us nowhere.

When the courts decide, then we talk.

+1

[thezerg]

Innocent until proven guilty. Rampant speculation like this gets us nowhere.

When the courts decide, then we talk.

Hello?
If you don't want to speculate then don't read the speculation forum!!!

[humanitee]

I forgot to laugh. It's one thing to speculate on the price of Bitcoin, but it's an entirely different matter to speculate on the outcome of judicial affairs.

This is way, way more serious than people here are acting.

Regardless, this subforum's description reads: "Speculation about the Bitcoin price"

[thezerg]

Sure I read somewhere that bitcoinica "hedges" on Mt. Gox... but how does that work if you are offering 10-1 leverage and Gox is not?  

Nope.  There is no evidence that Bitcoinica was a ponzi scheme.  You lack of understanding of margin trading hedges does not mean they were committing a crime.

Imagine there are 100 users at bitcoinica, each has deposited 10BTC:

- 50 of them have long 100 BTC
- 50 of them are short 100 BTC

That means the total of all the long positions is 5000 BTC, but bitcoinica only has 1000BTC on deposit.  How is this possible?  Internal order matching.

Because the NET bitcoinica position is 0, when one person loses money, it goes directly into another customer's pocket.  Their opposing trades effectively cancel each other out.

What if its not balanced? Bitcoinica hedged the net position of its customers.  So if there was a total of 1000BTC long positions and 500 BTC short positions, Bitcoinica would use their USD reserves to purchase 500 BTC on MTGOX.   If they ran out of USD or BTC (turned it all into the other currency to hedge), they would put up the * (lovingly called the starfish), on the side of the trade that they did not have the funds to hedge.

Bitcoinica's had a lot of problems, but there is no reason to suspect them of being a ponzi.

I certainly understand all this elementary information.  Of course when you offer 10 to 1 leverage your reserves better be pretty huge.  And your "hedge" explanation is buying BTC to cover the bet, but it does not change that those 500BTC are a bet by bitcoinica against it user.  Buying options would probably be a better choice but there was no options market...  If you don't do that even once (even by accident, say the market moved before you could put the order in) and lose then suddenly you could be underwater.  If that happened, would you just close up shop?  No I think most people would keep going and try to make it back next time or maybe in fees.  So losing .5% becomes 1% become 5% and suddenly your bucket ship is in ponzi territory.

I proposed this because it may actually be a simpler explanation; 3 completely independent hacks is a little hard to swallow.  A single underlying motive makes more sense.



 

[byronbb]

Why they offered such large leverage is a mystery to me.

[davout]

Why they offered such large leverage is a mystery to me.

Why people ask such questions is beyond me. Because there is demand maybe ? 10x leverage is nothing compared to what you can find on lots of Forex brokerage platforms.

[davout]

The one theory that seems to strangely ring true to me is the theory that bitcoinica became an accidental ponzi.  I mean nobody else could/can offer this kind of leverage.  Maybe because they actually put some math behind their web site . 

[... theories, speculation, wild guesses ...]

I joined bitcoin too late to be part of this mess.  Does this theory match your experience?  Discuss!

The source got leaked buddy, go see for yourself.
And I'm not just saying that, I put my time where my mouth is.

See http://pastie.org/4257541 for a starters, that's my interpretation of the matching job.

[ChrisKoss]

And your "hedge" explanation is buying BTC to cover the bet, but it does not change that those 500BTC are a bet by bitcoinica against it user.  Buying options would probably be a better choice but there was no options market...  If you don't do that even once (even by accident, say the market moved before you could put the order in) and lose then suddenly you could be underwater.  If that happened, would you just close up shop?  No I think most people would keep going and try to make it back next time or maybe in fees.  So losing .5% becomes 1% become 5% and suddenly your bucket ship is in ponzi territory.
 

As soon as the 500 BTC purchase goes through, Bitcoinica no longer has any currency risk.  They don't make any money when BTC goes up or lose when BTC goes down - the risk of those coins are owned by Bitcoinica customers.

You are correct though that problems with hedging could cause slight losses.  However, the more trades are hedged on MtGox, the more trade fees they would pay to them. Bitcoinica said they had a hedge-buffer, where they would only hedge after the whole order book moved a certain amount, so they would take something like 50-100BTC of currency risk so they don't have deal with hedging every 0.1BTC trade someone makes on their system.

As far as we know, I don't think Bitcoinica ever had a serious hedging problem.  I would guess their hedging strategy generally made them money (by saving them fees). It sounds like the problem was that after the funds were stolen, even though Tihan personally put in most of what was necessary to make everyone whole, they expected to make up the rest in operating profit.  The lack of volatility from Feb-April meant there was less money to be made trading, because you only make (or lose) money when the price changes. Then they got hacked again, and the rest is history...

[byronbb]

Why they offered such large leverage is a mystery to me.

Why people ask such questions is beyond me. Because there is demand maybe ? 10x leverage is nothing compared to what you can find on lots of Forex brokerage platforms.

10x leverage was too much for the market imo.  Forex is a 4 trillion a day volume. BTC is a bit smaller.

[davout]

Why they offered such large leverage is a mystery to me.

Why people ask such questions is beyond me. Because there is demand maybe ? 10x leverage is nothing compared to what you can find on lots of Forex brokerage platforms.

10x leverage was too much for the market imo.  Forex is a 4 trillion a day volume. BTC is a bit smaller.

Guess I see your point here, and I tend to agree. On the other hand, if you examine the code you'll find this limiting mechanism.

Code:

class RestrictingLeverageJob
  
  @queue = :normal
  
  def self.perform
    User.all.each do |user|
      if user.margin_balance > 20000
        user.leverage = 2.5 if user.leverage > 2.5
      elsif user.margin_balance > 2000
        user.leverage = 5.0 if user.leverage > 5.0
      end

      user.save
    end
  end
end

[thezerg]

Brilliant... the leverage restrictions are hard-coded into the code, have absolutely no relation to cash-on-hand or any other market activity, and use some if-then logic rather then actually calculating the maximum exposure (i.e. = account size * max leverage). 

And this may be obvious to all coders but still must be mentioned; this leaked code may not be exactly what was running on the live site.  It would be easy for anyone with access to patch in chunks of code that essentially do whatever the author wants to be done without checking in to github.

[nimda]

Brilliant... the leverage restrictions are hard-coded into the code, have absolutely no relation to cash-on-hand or any other market activity, and use some if-then logic rather then actually calculating the maximum exposure (i.e. = account size * max leverage). 

And this may be obvious to all coders but still must be mentioned; this leaked code may not be exactly what was running on the live site.  It would be easy for anyone with access to patch in chunks of code that essentially do whatever the author wants to be done without checking in to github.

lol I thought the same thing. If 20000, max is 2.5, if 2000, max is 5. Genious. Never mind the fact that 10x leverage on 1999 is bigger than 5x leverage on 2000

It's not like it's written in brainfuck; a log scale would be easy enough.

[Fjordbit]

Brilliant... the leverage restrictions are hard-coded into the code, have absolutely no relation to cash-on-hand or any other market activity, and use some if-then logic rather then actually calculating the maximum exposure (i.e. = account size * max leverage). 

And this may be obvious to all coders but still must be mentioned; this leaked code may not be exactly what was running on the live site.  It would be easy for anyone with access to patch in chunks of code that essentially do whatever the author wants to be done without checking in to github.

lol I thought the same thing. If 20000, max is 2.5, if 2000, max is 5. Genious. Never mind the fact that 10x leverage on 1999 is bigger than 5x leverage on 2000

It's not like it's written in brainfuck; a log scale would be easy enough.

I await the rollout of your leveraged trading platform with baited breath.

[thezerg]

A 17yo in a mostly unregulated market playing with a toy currency can crank this out without much thought.

Fast forward a year or so where BTC has a LOT more legitimacy (and value) and an experienced developer sees a lot of risk:

regulatory laws
endless KYC/AML homework
personal liability against hackers
personal liability operating as a bucket shop (betting against clients to promote liquidity)

And not much profit if you play strictly by the rules.

On top of that its very likely that an experienced developer has a lot more to lose (like a house). 

This is why the exchange grew where it did...

In fact, this becomes obvious looking at the next generation of companies -- they are mostly created by groups of professionals from day 1.  Bitinstant for example has a law firm on retainer.

We'll remember this summer as the period when the amateurs washed out and the pros stepped in...

[paulie_w]

A 17yo in a mostly unregulated market playing with a toy currency can crank this out without much thought.

Fast forward a year or so where BTC has a LOT more legitimacy (and value) and an experienced developer sees a lot of risk:

regulatory laws
endless KYC/AML homework
personal liability against hackers
personal liability operating as a bucket shop (betting against clients to promote liquidity)

And not much profit if you play strictly by the rules.

On top of that its very likely that an experienced developer has a lot more to lose (like a house). 

This is why the exchange grew where it did...

In fact, this becomes obvious looking at the next generation of companies -- they are mostly created by groups of professionals from day 1.  Bitinstant for example has a law firm on retainer.

We'll remember this summer as the period when the amateurs washed out and the pros stepped in...

i don't think we're ever going to see this kind of thing happen as a matter of requirement.

bitcoin is an open-source, decentralized currency.

[thezerg]

I'm not saying its a matter of requirement.  Let me spell it out:

I'm saying that everybody who has honesty, integrity and experience (that is a life and net worth separate from bitcoin) is going to ask themselves:

Do I want to hack up a site to hold in trust the money of a bunch of anarchists, gun-toting libertarians, drug dealers (and oh yea some regular people) to provide a service that is very likely to ultimately be considered in violation of all kinds of securities regulation?

There are 2 possible outcomes: 1. site hacked and chased for the rest of your life by the above crew.  2. site incredibly successful, welcome to the government "farm".  Sorry the only currency we use in here are cigarettes.