Author Topic: Bitshares shorting is flawed and needs to be fixed ASAP  (Read 14769 times)

0 Members and 1 Guest are viewing this topic.

Offline paliboy

Same goes with closing position. It is ridiculous that you need to deposit additional funds on top of collateral to settle your debt on DEX,  or find other workarounds, because the only purpose of collateral is to be used to settle the debt. And again, the amount of collateral which you need to spend depends on the price at which you buy back you debt, cheaper you buy, less you spend. If somebody kindly sends you bitAsset for free, you just keep your collateral.

Let's say that I borrowed 10 bitUSD @ 100 BTS with 2000 BTS collateral and then sold it for 10x100 BTS.

  • I buy 5 bitUSD @ 84 BTS in market, half of the debt could be automatically settled - do we want to do it? How would it adjust the collateral? Proportionately or should it keed some platform-defined/user-defined collateral ratio?
  • I want to settle half of my debt, I click "Update position" in "Margin Positions" in Bitshares wallet, enter 5 into Debt input box and click "Update positon" button. Network buys necessary assets for me at market rate and updates the collateral ratio - the same problem as in point 1.

IMHO it shouldn't be very difficult to implement (semi-)automatic closing of positions. Which one would people prefer? Or do you want something completely different?

Offline yvv

  • Hero Member
  • *****
  • Posts: 1186
    • View Profile
I understand that polo way is basically the same (just in different order), but much easier for use and have less friction.

Yes, this is what I am talking about. Just forget about their higher leverage for a moment (although this is also important factor). The hole process of shorting at polo is one simple action: you place a sell order. To settle your debt, you place a buy order. The final result is the same as in BTS, but everything is seamless, and you don't need to make unnecessarily  high deposits. The mechanics behind this is very simple, and I don't see a reason why it should not work at DEX. You can still use your bitAssets as money, i.e. buy/sell/transfer etc.   

The amount of collateral which you need to put upfront depends on the price at which you short sell your asset. If you short at feed price, you need to put only MCR-1 upfront, because the rest you get when you sell your asset, and before that you don't have no dept. If you sell at premium, you need to put less upfront, if you sell at discount, you need to put more. If you want to short, but don't want to sell, you need to put entire collateral upfront and then do what you want with your asset, same as you do now.

Same goes with closing position. It is ridiculous that you need to deposit additional funds on top of collateral to settle your debt on DEX,  or find other workarounds, because the only purpose of collateral is to be used to settle the debt. And again, the amount of collateral which you need to spend depends on the price at which you buy back you debt, cheaper you buy, less you spend. If somebody kindly sends you bitAsset for free, you just keep your collateral.

« Last Edit: April 21, 2017, 12:05:39 am by yvv »

Offline yvv

  • Hero Member
  • *****
  • Posts: 1186
    • View Profile

Well said. I wonder why @yvv didn't respond to this.

IMHO current design of MPA is not for high-risk seeking traders, but the opposite.

If we want to attract high-risk seeking traders, we need other products (done by 3rd party or not).

What do you expect me to respond? I agree with what the guy said. No, BTS can't offer 50:1 leverage at current market depth. This is not what I raise in OP.

Offline yvv

  • Hero Member
  • *****
  • Posts: 1186
    • View Profile
What if creating a MPA and shorting an asset are two different features?

Maybe we just need to build an additional shorting feature like on Poloniex for only that purpose?

Why don't we fix MPA shorting instead? Would not this be much simpler?

Offline abit

  • Committee member
  • Hero Member
  • *
  • Posts: 4664
    • View Profile
    • Abit's Hive Blog
  • BitShares: abit
  • GitHub: abitmore

The leverage you like so much is merely an illusion of safety made attractive by potential profits. That safety will evaporate faster than a bot can place an order when the inevitable collapse begins.

You didn't even understand what am I talking about.  Yes, I like leverage, because this is what gives motivation to short assets instead of just holding them. And since MPA are created through shorting, no shorting means no liquidity. Let's forget about it for a moment. Suppose we are fine with 1:1 leverage, then BTS shorting is still screwed. You still need to put unnecessary high amount funds when you open and close position. How does this make BTS more safe than polonoiex? I say, it does not. In fact, it hurts BTS a lot.

Yvv is correct that the way the current system is designed, it is cumbersome and confusing to short and trade... pair that with the lack of order types (Take profit, stop loss) and you have something that isn't appealing or similar to what traders currently use.

The question is, as Xeroc eluded too, should BTS change it's risk profile to add leverage and cater to traders? personally I think 3rd parties should be the ones offering leverage and then use the DEX as a way to manage their risk.  The way poloniex and most exchanges operate is that they make you have a minimum margin when using leverage, if the value of your position falls below the minimum margin your position is automatically closed.  In order for this to work, ample liquidity and deep order books are mandatory.  MF-Tzo referenced when the SNB depegged the swiss frank from the euro... Liquidity evaporated and there was a 20% move in the currency.  This caused many FX traders to go under as well as some exchanges (FXCM had to be bought out).  This is not a position that the DEX should ever be put in, and even if it was decided that BTS should be like FXCM or poloniex, I'm going to guess some drastic changes would need to be made to the code. 

Truth be told... most brokers that offer 50:1 or more leverage are gambling houses.  The big bank traders don't come close to that leverage (if they use any at all), and usually their orders are passed directly to the interbank market.  The 50:1 brokers typically use a dealing desk and take the opposite side of their clients trades.  They will sometimes pass the orders to the interbank market to help manage their risks, but this is typically only done for large orders.

Well said. I wonder why @yvv didn't respond to this.

IMHO current design of MPA is not for high-risk seeking traders, but the opposite.

If we want to attract high-risk seeking traders, we need other products (done by 3rd party or not).
BitShares committee member: abit
BitShares witness: in.abit

Offline nmywn

  • Sr. Member
  • ****
  • Posts: 266
    • View Profile
You would need special margin account that limit you for only one use case. So this is not flaw in current implementation, as you stated, but lack of additional feature. I understand that polo way is basically the same (just in different order), but much easier for use and have less friction. Maybe workaround (for now) would be bot that allow you pass all the necessary steps behind the scenes, while creating illusion that you already have full funds in moment of deposit. But, disadvatange/advantage is clear: you can't create fake market pressure this way (sell walls).

@Chris4210 yeah.

Offline Chris4210

  • Sr. Member
  • ****
  • Posts: 431
  • Keep Building!
    • View Profile
    • www.payger.com
  • BitShares: chris4210
What if creating a MPA and shorting an asset are two different features?

Maybe we just need to build an additional shorting feature like on Poloniex for only that purpose?
Vote Chris4210 for Committee Member http://bit.ly/1WKC03B! | www.Payger.com - Payments + Messenger | www.BitShareshub.io - Community based fanpage for the BitShares Blockchain

Offline yvv

  • Hero Member
  • *****
  • Posts: 1186
    • View Profile
I can buy any worthless UIA with borrowed bitUSD, which wouldn't be possible with polo scheme. Why? Because you MUST sell it against BTS to cover collateral. That sucks. Any fiat gateways  using MPA's instead of UIA would be forced to buy bitXXX from the market to build reserves. Now they can simply borrow, not for shorting purposes, but for operating deposit/withdrawals, making markets etc. without need of selling BTS power. It's not shorters heaven, because it wasn't designed for them. It is not dumb, look top volume markets for OPEN.BTC. It's working.

You do need to put the whole collateral upfront if are  not selling bitUSD for BTS (selling it for shit token or giving it away), there is no problem with that. But you don't need to put the whole collateral upfront if you are selling for BTS, and you are required to put it upfront, this is a problem. A better solution exists and it works on other exchanges.

In fact, although the base collateral asset for bitUSD is BTS, you don't actually need to keep BTS to back it. If you don't trust BTS, you could keep bitCNY or other bitAsset instead and that would be as good security as BTS. If this was allowed, BTS shorting would be really powerful. But this is another story which should be discussed somewhere else.

Offline nmywn

  • Sr. Member
  • ****
  • Posts: 266
    • View Profile
I can buy any worthless UIA with borrowed bitUSD, which wouldn't be possible with polo scheme. Why? Because you MUST sell it against BTS to cover collateral. That sucks. Any fiat gateways  using MPA's instead of UIA would be forced to buy bitXXX from the market to build reserves. Now they can simply borrow, not for shorting purposes, but for operating deposit/withdrawals, making markets etc. without need of selling BTS power. It's not shorters heaven, because it wasn't designed for them. It is not dumb, look top volume markets for OPEN.BTC. It's working.

Offline yvv

  • Hero Member
  • *****
  • Posts: 1186
    • View Profile

Not the margin trading but in general, the whole system. Do they show proof of reserves for their system? I don't know.

As a centralized exchange, they surely can do all kind of bad things, no doubt in that. But we are talking about margin trading here. And this is what they do the right way. Not only higher leverage than bitshares, but trading mechanics is different too.

Offline Pheonike


Not the margin trading but in general, the whole system. Do they show proof of reserves for their system? I don't know. 

Offline yvv

  • Hero Member
  • *****
  • Posts: 1186
    • View Profile

You must also remember that when exchanges like POLO are providing leverage, all those transaction are just number in a database.  They are running a fractional reserve . 

When you open a margin position at polo, you are required to back it with minimum 140% collateral, which is allowed go down to 120% later. How does this become a fractional reserve?

Offline Pheonike


You must also remember that when exchanges like POLO are providing leverage, all those transaction are just number in a database.  They are running a fractional reserve .  We bitshares cannot have a fractional reserve. Most of those can't cover all the deposits on hand. Bitshares can not run a fractional reserve. It was deigned to be the direct opposite. The systematic risk is too high. We can't Fudge the numbers like other places do.

Offline lil_jay890

  • Hero Member
  • *****
  • Posts: 1197
    • View Profile

The leverage you like so much is merely an illusion of safety made attractive by potential profits. That safety will evaporate faster than a bot can place an order when the inevitable collapse begins.

You didn't even understand what am I talking about.  Yes, I like leverage, because this is what gives motivation to short assets instead of just holding them. And since MPA are created through shorting, no shorting means no liquidity. Let's forget about it for a moment. Suppose we are fine with 1:1 leverage, then BTS shorting is still screwed. You still need to put unnecessary high amount funds when you open and close position. How does this make BTS more safe than polonoiex? I say, it does not. In fact, it hurts BTS a lot.

Yvv is correct that the way the current system is designed, it is cumbersome and confusing to short and trade... pair that with the lack of order types (Take profit, stop loss) and you have something that isn't appealing or similar to what traders currently use.

The question is, as Xeroc eluded too, should BTS change it's risk profile to add leverage and cater to traders? personally I think 3rd parties should be the ones offering leverage and then use the DEX as a way to manage their risk.  The way poloniex and most exchanges operate is that they make you have a minimum margin when using leverage, if the value of your position falls below the minimum margin your position is automatically closed.  In order for this to work, ample liquidity and deep order books are mandatory.  MF-Tzo referenced when the SNB depegged the swiss frank from the euro... Liquidity evaporated and there was a 20% move in the currency.  This caused many FX traders to go under as well as some exchanges (FXCM had to be bought out).  This is not a position that the DEX should ever be put in, and even if it was decided that BTS should be like FXCM or poloniex, I'm going to guess some drastic changes would need to be made to the code. 

Truth be told... most brokers that offer 50:1 or more leverage are gambling houses.  The big bank traders don't come close to that leverage (if they use any at all), and usually their orders are passed directly to the interbank market.  The 50:1 brokers typically use a dealing desk and take the opposite side of their clients trades.  They will sometimes pass the orders to the interbank market to help manage their risks, but this is typically only done for large orders.

Offline yvv

  • Hero Member
  • *****
  • Posts: 1186
    • View Profile

It seems I'm not the one who lacks understanding here.

This is very simple to understand. Let me show this to you ELI5 style. Suppose, you have 1 USD worth of BTS in your account. With double collateral (no leverage), you should be able to short 1 bitUSD, right? Tell me step by step, how would you do this in bitshares. Just answer my question, and I will repeat myself how I would do the same at poloniex to make the difference clear. @Thom