We have virtually no risk.
The market doesn't slip every 5 seconds. And during most of the violent moves previously, we received no or only a few orders. The high volume was always generated after the violent moves.
Now thankfully, Zhoutong, you seem to have the risks of slippage covered as explained above this quote, but your reasoning in this quote is flawed. Never _ever_ think something is not going to happen because it hasn't happened before. Always expect the worst and then some. This applies to anything, not just slippage risks. If you haven't made simulations in which there was heavy trading _during_ price movements, you should! Even though I don't have any practical experience to prove this (because it never happened before), I do have theoretical consideration. The main problem of any system with guaranteed liquidity is adverse selection. We are exposed to the systematic risk of customers trading only when they have opportunities to take advantage of price differences. A similar example is exchange quotation. TradeHill gives users 5 seconds to confirm a Instant trade quotation. But since there's no cost to get a quotation, theoretically there's one possibility: a user gets a quote every 5 seconds, and confirm only when there's huge slippage. What I have done here is to delay all order execution by 1-4 seconds. This makes sure that when you click Buy/Sell button exactly during the slippage, the actual execution will happen after the major price move. And this is still considered as "guaranteed liquidity" because we update prices globally, not just for individual pending orders. Mushoz: can u translate this? It means he is protecting himself against slippage due to random moves by using spreads, but the biggest risk is people trying to exploit slippage. Let's say the price is at 4$ on Mtgox and suddenly a spikes causes it to go down to 3.60. Someone wanting to exploit slippage could quickly sell at Bitcoinica, where the prices wouldn't have been updated yet. That's what the 1-4 second order execution delay is for. It allows the price on Bitcoinica to catch up to the market price, so that people aren't able to exploit slippage. what if the spike comes at exactly 4 sec after the customer pushes his button? what if a series of ramps occur? i know, i know; he says he only executes 50 BTC at a time but in his explanation above he says 50, 100, sometimes 150 BTC. which is it? If the spike comes exactly 4 seconds after order placement, we can be sure that the customer was not intending to exploit the spike because no one actually knew. So it can be 0, or 50. 100 and 150 are possibilities when multiple customers place order at the same time, and it's extremely rare. what if i'm the one who pushes the button on Bitcoinica, counts off 4 sec, and pushes the button on mtgox creating the spike? Then you would have lost a lot of money creating a sizable spike, would you not? not if at a given pt in time both gox and bitcoinica have respective asks say @ $4 and $4.10 due to low activity in the market. example: i have accts at both gox and bitcoinica. i push the buy button at bitcoinica to buy say 100 btc, count off 4 sec to secure the buy @ $4.10, then push the mtgox buy button to buy whatever it takes to create a spike to $4.5 before bitcoinica has a chance to hedge @ $4. once bitcoinica adjusts its bid to say $4.4 i then sell those same 100 btc on bitcoinica for $4.4 In that case, with perfect execution you will profit 30 cents per BTC times 100BTC, which will net you 30$. The USD required to create such a spike is _far_ more and it is far too risky to try and profit 30$. It would only be possible if the depth was very shallow, weren't it for the fact in that case the spread is a lot bigger to compensate for the shallowness of the orderbook. |
|
|
|
Just tested it! Working great! Thanks for implementing this  |
|
|
|
I agree. Those long-squeeze statements were out of character and a bad move on his part. Doesn't change anything for me with regard to Bitcoinica. I haven't used it and don't have any intentions to, not because I think it's necessarily a bad service, but because I simply don't want to get involved in that sort of leveraged trading.
Same here, although I thought about buying Bitcoins with all my Bitcoin savings as margin, but ended up leaving it at a test of 2 BTC for 2 BTC, which are up from 3.5. I’m not going to use it because it’s too risky for me, but if you are really certain about a move (or certain about a bottom), it should be a great tool. I wonder if the ability of buying BTC on margin will have the same disastrous effect upwards as shorting had downwards? For every BTC shorted at Bitcoinica, a BTC was needed by Bitcoinica. How did Bitcoinica get those BTC? By buying them. You are correct it can have an upward effect though. (The investors of) Bitcoinica provided USD that can be used to buy more BTC, which in turn will drive up the price. And I must say I agree that the "market advice" Zhoutong provided was unprofessional and out of line for someone in his position. I'm glad he apologized and hopefully such a thing won't happen again |
|
|
|
BID WALL!
It's gone. Its owner is playing the shark-fin strategy. He just wants us to know he's swimming around. OR he took away the wall so he can use those funds to buy more BTC and spark another rally above 5$ :p Nah, that's just wishful thinking ^^ |
|
|
|
|
Yeah can't connect either. I bet their number of connected people would be a lot higher if they can get their act together during big movements.
|
|
|
|
Great work! You managed to stay online even though the trading volume was huge again! Can't say the same about Mtgoxlive.com =) I'd love to see ticker data similar to Mtgoxlive if possible, though |
|
|
|
I already pointed out the problem with this. The price can change, and if you're holding the wrong asset you lose. Period. No USD + downswing = reserves are now fractional.
Wrong. If your holding the wrong asset the customer that took that position loses, and the loss is deducted from the customer's balance, so that Bitcoinica's balance stays equal. Please read the topic that explains how hedging on Bitcoinica works. Bitcoinica has a certain BTC and USD balance, and not accounting for spreads, those balances would stay exactly the same if everyone liquidates. Those imbalances only exist temporarily as long as people have open positions. |
|
|
|
We have virtually no risk.
The market doesn't slip every 5 seconds. And during most of the violent moves previously, we received no or only a few orders. The high volume was always generated after the violent moves.
Now thankfully, Zhoutong, you seem to have the risks of slippage covered as explained above this quote, but your reasoning in this quote is flawed. Never _ever_ think something is not going to happen because it hasn't happened before. Always expect the worst and then some. This applies to anything, not just slippage risks. If you haven't made simulations in which there was heavy trading _during_ price movements, you should! Even though I don't have any practical experience to prove this (because it never happened before), I do have theoretical consideration. The main problem of any system with guaranteed liquidity is adverse selection. We are exposed to the systematic risk of customers trading only when they have opportunities to take advantage of price differences. A similar example is exchange quotation. TradeHill gives users 5 seconds to confirm a Instant trade quotation. But since there's no cost to get a quotation, theoretically there's one possibility: a user gets a quote every 5 seconds, and confirm only when there's huge slippage. What I have done here is to delay all order execution by 1-4 seconds. This makes sure that when you click Buy/Sell button exactly during the slippage, the actual execution will happen after the major price move. And this is still considered as "guaranteed liquidity" because we update prices globally, not just for individual pending orders. Mushoz: can u translate this? It means he is protecting himself against slippage due to random moves by using spreads, but the biggest risk is people trying to exploit slippage. Let's say the price is at 4$ on Mtgox and suddenly a spikes causes it to go down to 3.60. Someone wanting to exploit slippage could quickly sell at Bitcoinica, where the prices wouldn't have been updated yet. That's what the 1-4 second order execution delay is for. It allows the price on Bitcoinica to catch up to the market price, so that people aren't able to exploit slippage. |
|
|
|
That his customers were long was disclosed by Zhoutong last night after bitcoinica ran out of USD. I for one am boycotting Bitcoinica until they move away from fractional reserve and hold assets to back every position. Regarding disclosure, I would prefer the long vs. short balance updated in real time on the site, but I will not boycott for that if they solve the backing issue. With realtime long/short balance users would know if they were taking a position that had a large risk of being squeezed.
Why do you think they do fractional reserve? Zhoutong has denied this many times - is he lying? As to the USD shortage - this is easily explained without any conspiracy - when people want to go long on Bitcoinica, Bitcoinica needs to buy BTC on MtGox with their USD funds and those funds are not unlimited - so when they are drained they refuse to let people go long. This is a result of them not wanting to have uncovered positions - quite the opposite of what you are suggesting. By they way I hate Bitcoinica for the volatility it produces. Obviously the money has been used up to purchase Bitcoins for the longs.
You have money in Bitcoinica account doesn't mean it's 100% in reserves, it can be borrowed by someone else to buy/sell Bitcoins. At any point, Bitcoinica is a:
- Full reserve in BTC and fractional reserve in USD, or
- Full reserve in USD and fractional reserve in BTC.
(Yesterday, we had a full reserve in BTC, and a depleted reserve in USD. Users couldn't withdraw USD until the situation was resolved.)
This is how hedging exactly works to ensure that we ourselves and our customers are always having almost the same profits (or we call it internally, rate of change of asset value with respect to market price).
The problem is there were short positions that were uncovered because their USD was used to buy BTC for the longs. I don't think Zhou used the term "fractional reserve" properly here. He had full reserve of his users accounts, just not in USD. It would be fractional reserve if the value of the bitcoin he held + the value of USD he held were less than the stated amount of the sum of his users accounts. Exactly! Nothing to add to this post. This is exactly what happened. |
|
|
|
I approve of your post and thinking  |
|
|
|
I think credit instruments in Bitcoin should be avoided like the plague. Getting away from the bankster-issued, debt-backed funnymoney is imho the *prime* benefit of Bitcoin. For the same reason, am I very apprehensive about leveraged trading, since this is conventionally done with "margin"="credit"=debt/paper money. Clearly, you can't (and thus don't) sell "paper" Bitcoins on mtGox - yet, so for now it might be ok, but what if credit (promise-to-pay) Bitcoin starts appearing for real within Bitcoin?? I think mtGox credits are a step in that dangerous direction. Please people - never forget the difference between real Bitcoin and promisary notes! That leads to the dark side  -practical question: Does Bitcoinica follow the uptic-rule? If not, why not? What do you think about Bitcoin redeemable MtGox codes? those seem strikingly similar to "paper" bitcoins. Those are 100% backed by Bitcoins though. Or in the case of USD redeemable codes 100% backed by USD. didn't most paper money start out backed by one thing or another? It just seems to me that if you don't like "paper" money then you should't like the precursors to it either. Kinda the same way that if you want to quit smoking, it's a good idea to avoid places where you liked to smoke. There's plenty of examples to think of where the precursor is a good thing. As long as it stays at the precursor (in this case, as long as it stays 100% backed), it's fine with me. It's a very convenient way to move funds instantly. I've used it a few times to move funds between Bitcoinica and Mtgox. |
|
|
|
I think credit instruments in Bitcoin should be avoided like the plague. Getting away from the bankster-issued, debt-backed funnymoney is imho the *prime* benefit of Bitcoin. For the same reason, am I very apprehensive about leveraged trading, since this is conventionally done with "margin"="credit"=debt/paper money. Clearly, you can't (and thus don't) sell "paper" Bitcoins on mtGox - yet, so for now it might be ok, but what if credit (promise-to-pay) Bitcoin starts appearing for real within Bitcoin?? I think mtGox credits are a step in that dangerous direction. Please people - never forget the difference between real Bitcoin and promisary notes! That leads to the dark side -practical question: Does Bitcoinica follow the uptic-rule? If not, why not? What do you think about Bitcoin redeemable MtGox codes? those seem strikingly similar to "paper" bitcoins. Those are 100% backed by Bitcoins though. Or in the case of USD redeemable codes 100% backed by USD. |
|
|
|
My apologies. I guess I was wrong on the "hedge before trade" part. Looking back, it does make sense now, as Bitcoinica is guaranteeing liquidity up to 50BTC, which would be impossible if they indeed used a "Hedge before trade" system. I must have misinterpreted Zhoutong's post on the first page. This indeed makes cypherdoc's concerns on slippage valid, but thankfully Zhoutong seems to have solutions to greatly reduce those kind of risks. The fist and best solution is what they have already implemented, which is to work with X BTC blocks (In their case 50BTC) max. Even with huge slippage their losses shouldn't be hard to cover because it's only going to be over one or a couple of 50BTC blocks on huge movements, instead of over potentially huge trades. Their second solution is risk analysis using extreme simulations, to see at what kind of risks those extreme situation brings them. If they can cope even with the greatest movements, as they seem to have analyzed looking at Zhoutong's post (a spike to $9648.84556 BTC/USD prices), combined with 50BTC blocks max, we can be certain they can deal with the risks they are exposing themselves to. You have to remember it's not bad if Bitcoinica turns in a loss over a small time period due to the risks, as long as they turn in profits over the long run. This leaves the slippage problem on forced liquidations, which could indeed be disastrous if all goes wrong. But thankfully this has already been discussed in this topic, and thankfully Zhoutong listened to my advice in the topic start. He is going to create a new algorithm which gives liquidating short positions priority over going long by preventing more longs _before_ they run out of USD reserves, so that they always have reserves left to allow people to liquidate. Of course the same is going to be true in reverse for liquidating long positions and going short. Prevent more shorts _before_ they run out of BTC, so that people can still liquidate their longs. We have virtually no risk.
The market doesn't slip every 5 seconds. And during most of the violent moves previously, we received no or only a few orders. The high volume was always generated after the violent moves.
Now thankfully, Zhoutong, you seem to have the risks of slippage covered as explained above this quote, but your reasoning in this quote is flawed. Never _ever_ think something is not going to happen because it hasn't happened before. Always expect the worst and then some. This applies to anything, not just slippage risks. If you haven't made simulations in which there was heavy trading _during_ price movements, you should! |
|
|
|
And that's why the trade on Bitcoinica doesn't go through before they hedged the position on Mtgox. The situation you're describing simply isn't possible. They try to hedge for 4$. When that isn't possible due to whatever situation you can come up with, and the price falls to 3,80, that's the price the costumer is going to get. Unfair? That's the risk of market orders. If the customer used a limit order instead, which he should if he cares about the risk, and places the limit order at 4$, the trade simply wouldn't go through, because Bitcoinica is unable to hedge at 4$ as the price already fell to 3,80$.
this part can't be correct. the way i understand it is that the customer gets his order executed before Zhou has a chance to hedge on mtgox. this is why the spreads are higher on Bitcoinica to acct for slippage or his increased risk of doing it this way. example: if mtgox bid:ask is $4:$4.10 then Bitcoinica might be $3.80:$4.30 thus you, as a customer of Bitcoinica, can short at the ask of $3.80 with a limit order and you will get it filled immediately. the $0.20 difference is what Zhou's algorithm has figured to be a safe cushion to give him time to hedge on mtgox and sell his btc at or somewhere btwn $3.80 and $4 so he doesn't lose money. in your example, you state that even if you short at the ask price of $3.80 it won't be filled until and if Zhou is able to sell a corresponding amt of btc on mtgox at $3.80. if that was the case then his spreads should never be higher than mtgox's since he doesn't assume any risk. correct me if i'm wrong. They have those spreads to protect against slippage when people get forced liquidated. When there's a sudden movement, a lot of positions might have to be liquidated. If they can't hedge against those liquidations they _still_ have to liquidate those positions, because they are forced after all. That's what the spread is for. (And for profit of course) |
|
|
|
the other scenario Mushoz's simplistic argument ignores is what i've been calling slippage. My post wasn't an argument. I wanted to explain how Bitcoinica works and how they hedge. Clearly in the other thread you said it was weird Bitcoinica said they had loads of BTC, but no USD, and that people couldn't buy BTC to go long or buy BTC to liquidate their shorts. All I wanted to do with my post is explain why liquidating shorts or going long requires Bitcoinica to have USD, and _not_ BTC. This "simplistic scenario" was clearly needed, because without it you failed to understand how hedging works, yet you were arguing about it like you did know how it works. I agree 100% with you that people getting stuck with shorts and being unable to liquidate them is unacceptable. And that's what my suggestion to Zhoutong was for. He already said he is going to implement such a function, so that's good to prevent a similar situation in the future. In the mean time he already said he was going to pay out of his own pocket for the risk of others. I simply can't think of a way he could have handled this situation better. the grossest example of this is when Zhou lost his Yubikey and thus his connection to gox. for example: Of course they are using the API, and not a Yubikey... let say you sell or short 5 BTC at Bitcoinica for $4 in a downtrending market like we had a coupla months back. he loses his connection to gox for whatever reason. in the meantime the price decreases further to $3.80. he then sells 5 BTC @ $3.80 only recovering $19 instead of $20. this begins to erode his USD balance or reserves. now we begin to understand where all the USD went. And that's why the trade on Bitcoinica doesn't go through before they hedged the position on Mtgox. The situation you're describing simply isn't possible. They try to hedge for 4$. When that isn't possible due to whatever situation you can come up with, and the price falls to 3,80, that's the price the costumer is going to get. Unfair? That's the risk of market orders. If the customer used a limit order instead, which he should if he cares about the risk, and places the limit order at 4$, the trade simply wouldn't go through, because Bitcoinica is unable to hedge at 4$ as the price already fell to 3,80$. even assuming he has his maximum connectivity in place, the fact that he's one step removed from gox still results in slippage as he can never execute (hedge) at the same price as his customers did on his site. multiply that by thousands of orders each week. Again, read the response to the quote above. now you will argue his algorithm takes this into acct through the spread. i argue its impossible to predict the violent swings and ignores the fact that there is constant complaining from customers to reduce that spread and thus his protection. it appears that he may in fact not have been charging enough via the spread given what has happened. It's impossible to know exactly how much the risk is going to be. That's why they overestimate the risk in their spreads. Unfair? Not at all. As you understand they need to protect themselves against the risk, and of course they need to profit as well. That's why the spread is bigger than required. Both as an extra buffer against the risk and for profit. Ferroh: assuming that Bitcoinica does indeed acct for 1/3-1/2 of gox's trading, what do you think about leveraged shorting and its effect on the USD/BTC price? it seems to me that it has caused an asymmetric artificial overshoot to the downside but will not contribute to an overshoot to the upside due to his lack of USD reserve to continuously feed an upward spiral. Do you know what's required on Bitcoinica to short? BTC. Do you know how they get BTC? That's correct. They buy BTC. So for every BTC that was shorted, a BTC was bought. Actually, there's _more_ BTC bought by Bitcoinica than there was sold through shorting. Do you know why? Because they need reserves. on the way down, when there were few to no buyers, he was able to easily step in as a market maker with his USD reserves to buy up the selling/shorting pressure b/c the price was so low. this may not work to the same degree with higher prices to the upside though as we're witnessing. How do you know how Bitcoinica's balance is doing? AFAIK they have some pretty big investors, and turning in some nice profits. Why do you have to come up with stories how they might be running out of USD because of losses, when you have no idea how they are doing? Could it maybe be, that there are actually too many people long for their reserves to handle? I mean, look at the last month. It's been an extremely Bullish market. It isn't weird that a lot of people are long currently. in this sense Bitcoinica has been destructive to the price and perhaps arguably to the economy. am i missing something?
Yes, a lot. Honestly no offense, it's good to have skeptics as well. But please don't try to present things as facts and use them as arguments, when so many things are clearly wrong in your reasoning. If you think something might be bad for the market, Bitcoinica or anything else, but you are not sure how it works, just ask others, instead of stating it as facts. It would prevent a lot of FUD spreading through these forums. Thanks for reading! |
|
|
|
Yes, thank you, that's what I was imagining that would happen, as soon as the losses "touch" the total balance (including fees and protections etc), they would force the liquidation of your position, even if it's not expired yet.
The thing is, if right after the liquidation, but still before the expiration of your order, the prices swing on the other direction, you're still screwed in spite of having been "right on your bet"...
Is that how leverage is done on conventional financial markets as well? You always have to provide some guarantees before hand? For example, if I write a call option to someone, must I always prove before hand that I actually have what I would need to sell in case that option is exercised? I imagined such requirements were laxer on "non-anonymous scenarios", and that eventually people who lose too much would go bankrupt, with police going after them to seize their belongings and all that thing. Anyway, I thought you could go really bankrupt and remain owing people money by playing wrongly in these markets, instead of "at most losing your reserves".
That is correct. But then again, your bet was wrong on how much leverage to use. It's a risk vs reward calculation you have to do for yourself. Taking higher risks, and you are able to profit more, but there's less room for error. No idea how it works on regular markets. I *think* you are allowed to have negative balances on some of them, but it would require all kinds of verification procedures, so that they can be relatively sure you will fill up negative balances to zero or positive later. But then again, I have no experience at all with any financial markets. This is the first time I've spent time learning how these kinds of systems work. By studying Mtgox, Bitcoinica and financial articles. |
|
|
|
Interested newbie here...
Thank you for this topic, the "reserve problem" one was blowing my mind. I mean, I know what a call option or a put option is, but all that vocabulary of "covering his shorted coins slipping under his long leveraged position blablabla" is Greek to me.
If I understood correctly, the costumer always has to have a balance capable of paying off the risks he's running, is that it?
So, in the end, actual leverage (investing more money than you own) isn't yet possible in the bitcoin world, as I suspected?
And what changes if that 1:1 constant in your example isn't 1:1 anymore?
Thank you for the explanations.
Yes, you can actually invest more money than you own. Let's say I deposit 100$ into my bitcoinica account and the price of BTC/USD is 2$. If I set my maximum leverage to say 10:1, I can actually buy 10 times as many BTC as I would be able to buy on Mtgox. So in this case instead of 50BTC, I'm allowed to buy 500 (!) BTC with just 100$. The thing is, now that I have so many BTC, if the price drops by just 20 cents, I'm losing 500 * 0.20 = 100 dollars on that price movement, so I no longer have any cash left in Bitcoinica, so my position is forced liquidated, leaving me with neither Dollars nor BTC. So in a formula: When going long, using X:1 leverage, the bitcoin prices are allowed to drop by 100 / X % before I'm liquidated. Now here's the kicker: You actually have to subtract another 4% from that number, because that's how Bitcoinica protects themselves against sudden price swings. Because if they liquidate your position too late because the swing was too sudden, you would end up with a negative balance. So for the different leverages, this is how much the BTC/USD is allowed to drop before you are forced liquidated: 1.0 : 1 100/1 - 4 = 96% 2.5 : 1 100/2.5 - 4 = 36% 5.0 : 1 100/5 - 4 = 16% 10.0 : 1 100/10 - 4 = 6% Hope this clears it up for you! |
|
|
|
|
Do you think it would be a good idea to sticky this thread?
|
|
|
|
|
Do you know if the wallet.dat is usable on Windows clients as well? I could help out in that case, I'm familiar with switching out wallets and using the rescan option to retrieve transactions from the blockchain.
|
|
|
|
Op great explanation. However: Advice to Zhoutong
Please give people that have open short positions higher priority on buying BTC. It is simply unacceptable people were unable to liquidate their positions for several hours because of USD liquidity problems. Stop allowing more long positions _before_ you run out of USD, so that there's enough USD left for people with shorts to liquidate them.
Is a load of.... unless you're advocating the same in a reverse situation. Otherwise this is just fucking up a market that already has issues. Yes, that is correct. People with long positions should also get priority on selling. Thanks for the advice, I'll edit the post in a moment. Edit: Done. Thanks again for the tip! |
|
|
|
|
Show Posts
