Author Topic: Proposal - Significant Enhancement to Market Engine  (Read 24868 times)

0 Members and 1 Guest are viewing this topic.

Offline bytemaster

Think of the shorts as usd issuers who cannot run off with your money and are competing to put the most on the line. 

Right now few shorts want to cover because their losses are big and they don't think btsx can go much lower. 

I hope they are right. 

A market maker in my case is someone who doesn't mind a little short exposure because they are btsx bullish.   BitUSD will never surge to more than $1 while there are so many bulls like myself looking for leverage. 

The key is to allow shorts to cover with no fee as soon as possible to provide liquidity for usd holders. 

If people have high confidence that they can sell their usd near a dollar then they will buy.  The problem is that there is not any confidence in that and thus little demand. 

Monday will change all of that.  We will create a trading bot that will buy all the usd it can short at 99.5% of market price.

There will be huge buy and sell walls right near the peg.   

Once those walls are proven then usd holders will consider the yield and utility and demand will come.
For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

Offline GaltReport

While not understanding this stuff, it seems the last solution price fixed collateral (Sub-optimal)  but this one price fixes fee/interest. (Sub-optimal)

This solution makes it better for market makers but decreases genuine BitAsset demand as it offers interest rates that are 300%+ lower probably??

Isn't the best system one that encourages market makers and genuine BitAsset holders, Optimal interest + optimal collateral?

I don't know how to implement this & haven't shorted before, but thoughts..

1. Simplicity

Regardless of shorting system,  perhaps the main BitUSD market page should maybe just show & allow buying and selling to keep it very simple for users.

Then have a separate tab - 'Short BitUSD (Advanced)'

2. For the long term shorts it should be made simple. They just choose short and compete on fee/interest. (The system automatically places/keeps their order at 1-1 whatever that rate happens to be.)

For the market makers they enter shorts as per this proposal.

50% of new BitUSD is awarded to highest collateral and 50% to highest fee/interest when demand is at or above feed.

Then we have a system with more collateral and attractive interest that encourages market makers and genuine BitAsset demand (who will also appreciate more collateral & tighter range)

 +5% for #1

Offline Markus

  • Sr. Member
  • ****
  • Posts: 366
    • View Profile
I think that now that the price of BTC, it will be a good idea to implement this
https://bitsharestalk.org/index.php?topic=9075.0
as discussed by emski and drltc.

If you can buy directly bitUSD with BTC it will allow day traders to close their positions at a minimal risk / expense.
Not trustless, not liquid, but trading:
https://bter.com/trade/BTC_BITUSD

Offline betax

  • Hero Member
  • *****
  • Posts: 808
    • View Profile
I think that now that the price of BTC, it will be a good idea to implement this
https://bitsharestalk.org/index.php?topic=9075.0
as discussed by emski and drltc.

If you can buy directly bitUSD with BTC it will allow day traders to close their positions at a minimal risk / expense.
https://metaexchange.info | Bitcoin<->Altcoin exchange | Instant | Safe | Low spreads

Offline CLains

  • Hero Member
  • *****
  • Posts: 2606
    • View Profile
  • BitShares: clains
My two cents,

The people who provide liquidity are not gambling on a fair and balanced game, they are flossing sheep who graze on bitUSD. No sheep grazing on bitUSD, no liquidity.

Offline Empirical1.1

  • Hero Member
  • *****
  • Posts: 886
    • View Profile
While not understanding this stuff, it seems the last solution price fixed collateral (Sub-optimal)  but this one price fixes fee/interest. (Sub-optimal)

This solution makes it better for market makers but decreases genuine BitAsset demand as it offers interest rates that are 300%+ lower probably??

Isn't the best system one that encourages market makers and genuine BitAsset holders, Optimal interest + optimal collateral?

I don't know how to implement this & haven't shorted before, but thoughts..

1. Simplicity

Regardless of shorting system,  perhaps the main BitUSD market page should maybe just show & allow buying and selling to keep it very simple for users.

Then have a separate tab - 'Short BitUSD (Advanced)'

2. For the long term shorts it should be made simple. They just choose short and compete on fee/interest. (The system automatically places/keeps their order at 1-1 whatever that rate happens to be.)

For the market makers they enter shorts as per this proposal.

50% of new BitUSD is awarded to highest collateral and 50% to highest fee/interest when demand is at or above feed.

Then we have a system with more collateral and attractive interest that encourages market makers and genuine BitAsset demand (who will also appreciate more collateral & tighter range)

Ggozzo

  • Guest
What if the market shoots strait up to 105btsx/1$.  I am having trouble seeing where the market maker has very little to no risk.

This is an important objection to consider.  I think there are two good responses:

- If the price is steady at 100 BTSX / $1 for 8 hours, and each hour the market maker successfully shorts at 100 BTSX and buys at 99 BTSX, then they made 8 BTSX.  If during the 9th hour the price then goes up to 105 BTSX in a sudden move after the market maker has shorted but before they can cover, then taking the 5 BTSX loss still leaves 3 BTSX profit.

- Sudden moves happen in both directions.  They lose 5 BTSX when they're short and BitUSD rises to 105 BTSX, but they gain 5 BTSX when they're short and BitUSD falls to 95 BTSX.  If you're bullish about BTSX in the long term, then you expect the downward BitUSD movements to outnumber upward BitUSD movements.  bytemaster noted that this improvement is specifically targeted to making things more convenient for BTSX bulls.

I think we are throwing "market maker" around to loosely. Rarely will a true market maker be one directional. A MM won't put trades on without hedging risk, or having a spread advantage before the trade is initiated. This is where options and "beta weighting" play big parts in trading. A good MM will stay near delta neutral. This means they will offset risk by selling premium with low probabilities or buy premium with high probabilities.  We do not have that option right now. When the markets become more liquid we will. Rules will not bring liquidity quicker. Traders/people will. There is a reason the SPY and the QQQ are highly traded with a PENNY SPREAD!

Offline theoretical

What if the market shoots strait up to 105btsx/1$.  I am having trouble seeing where the market maker has very little to no risk.

This is an important objection to consider.  I think there are two good responses:

- If the price is steady at 100 BTSX / $1 for 8 hours, and each hour the market maker successfully shorts at 100 BTSX and buys at 99 BTSX, then they made 8 BTSX.  If during the 9th hour the price then goes up to 105 BTSX in a sudden move after the market maker has shorted but before they can cover, then taking the 5 BTSX loss still leaves 3 BTSX profit.

- Sudden moves happen in both directions.  They lose 5 BTSX when they're short and BitUSD rises to 105 BTSX, but they gain 5 BTSX when they're short and BitUSD falls to 95 BTSX.  If you're bullish about BTSX in the long term, then you expect the downward BitUSD movements to outnumber upward BitUSD movements.  bytemaster noted that this improvement is specifically targeted to making things more convenient for BTSX bulls.
BTS- theoretical / PTS- PZxpdC8RqWsdU3pVJeobZY7JFKVPfNpy5z / BTC- 1NfGejohzoVGffAD1CnCRgo9vApjCU2viY / the delegate formerly known as drltc / Nothing said on these forums is intended to be legally binding / All opinions are my own unless otherwise noted / Take action due to my posts at your own risk

Offline Helikopterben

  • Sr. Member
  • ****
  • Posts: 202
    • View Profile
I am still not sure about this.  Say the price feed is 100btsx/1$ and the spread is 1btsx.  If my short at 100btsx/$1 takes priority because it is highly collateralized and it gets filled, where is the guarantee that if I place a bid to cover at 99btsx/$1 that it will get filled?  I presume that if my short gets filled then all asks up to 100btsx/1$ have been filled and new asks would have to come in at 99btsx/$1 for me to cover.  What if the market shoots strait up to 105btsx/1$.  I am having trouble seeing where the market maker has very little to no risk.

Ggozzo

  • Guest
Quote
It sounds horrible.  You'd be giving an immense amount of control to utterly un-accountable third-party exchanges.

Where do you think the published price feeds come from now?

Offline starspirit

  • Hero Member
  • *****
  • Posts: 948
  • Financial markets pro over 20 years
    • View Profile
  • BitShares: starspirit
Earlier I said that market makers do not help the market to move toward the peg. They are merely providers of liquidity around the market's equilibrium price, and only narrow the spread around that price wherever that is. They can only help to maintain the peg if that is indeed where the market wants to be. Nobody in this thread has noted or rebutted this comment. Is this important or not?

Offline arhag

  • Hero Member
  • *****
  • Posts: 1214
    • View Profile
    • My posts on Steem
  • BitShares: arhag
  • GitHub: arhag
For that reason, I see value in providing moderate yields (if possible) that out-compete what people can get by holding the real world assets. So a 5% annual yield on BitCurrencies kicks the ass of any other bank out there. But why do we higher yields than that and why do we need to spend that money on yields for BitAssets where people could not get that kind of interest by holding the underlying, such as gold, oil, or BTC?

Because we want to kick ass in as many markets as possible.  You could invest in a gold ETF and get gold exposure, but the vault and management fees come out of investor funds.  Or you could invest in BitGold, get the same financial exposure and earn interest.  From BTSX investors' standpoint, the most likely path to wild success and great riches appears to be mass market adoption of BitUSD and/or BitCNY as a payment and deposit account product for ordinary consumers.  But that doesn't mean we shouldn't court commodity and currency exchange investors and traders as well.  Demand for BTSX as collateral, and transaction fees, is good for BTSX investors regardless of the source.  So stimulating that demand with yield makes sense -- especially for investments that don't normally have yields.

I understand that, but these are strategic decisions that I think ultimately the shareholders should have control over. If we can reduce the yield on BitGold for example, but now we have more money to spend on marketing or development, we might be able to grow BTSX value faster.

As for why we would want higher yields -- in theory, we would always want our yield to be the smallest needed to out-compete our competitors, convert the rest to BTSX and burn the BTSX.  In practice, hard-coding a yield now will make us brittle and inflexible if the real world sees a higher interest rate environment in the future.  And letting humans control yields makes the system opaque and political.  It would be an interesting experiment, but I think there are excellent marketing reasons to keep the more controversial aspects of the real-world financial system far from any blockchain we're going to advertise as part of the BitShares brand.

That being said, I would support selling some fixed fraction of BitAsset-denominated fees for BTSX which would be burned / given to delegates in the same way as other BTSX-denominated fees.  (We don't want to give the BitAssets directly to delegates in order to ensure BTSX investors to get a return.)

In my view we should let the shareholders vote on proposals adjusting the yield rate caps and the prioritization of BitAsset yields (for example, if we have limited funds available lets make sure we give BitUSD and BitCNY their full 5% before we use excess to provide yield to BitGold for example). I understand this makes the whole system more political and how that could be made to look bad by our competitors, but the alternative is to let the chaos of bid-ask overlap determine these (in my opinion) incredibly strategic business decisions that have an impact on the future success of this DAC. I think it is even more important now that the new market rules means the market fees from bid-ask overlap will be much lower (for good reason of course, although I still haven't been convinced it is better than the capital gains tax method of prioritizing shorts). It will also be especially important if any clones pop up that decide to do smarter distributions of the DAC's revenue.

In my view it would be smarter to take the value that would have been spent on those yields and instead use them to either boost the yields of the BitCurrencies if necessary, or give them to the delegates so they can invest in growing the BitShares ecosystem faster. And yes, I know this complicates the code, but I think it is worth it to eventually build into the system.

That gets into a debate on the proper role and place of delegates.  I tend to view delegates as something that was a technical implementation decision undertaken mainly to have reliable ten-second block times, which (it may be argued) is a necessary precondition of having a liquid and dynamic on-chain market, which itself is a precondition to the whole BitAsset short/long system.  So I would support less active delegates with lower payouts, especially if we're giving them income as well.

I agree with that view actually. I just said "give them to the delegates" for simplicity. In reality, I would like to see a proposal system used to allow the shareholders to decide how the DAC's revenue is distributed.
« Last Edit: September 19, 2014, 01:50:00 am by arhag »

Offline theoretical

Why not just make the feed the end all? Don't let delegates publish feeds, let them decide which exchanges should be used to discover the median price. Then the median price feed from all third party exchanges along with their prices in real time should be displayed in the GUI. It should update as close to real time as possible. Then you just don't allow any order away from the median feed at that given moment to execute.  It will always be true to the outside value of BTSX. Not sure if it's possible but sounds good.

It sounds horrible.  You'd be giving an immense amount of control to utterly un-accountable third-party exchanges.

Currently there's a delegate in the loop to decide which exchanges are reliable, what data to use, how to pull or combine data from multiple exchanges, etc.  That delegate is accountable to BTSX shareholders because they can always be removed.

And this proposal is also basically technically impossible.  Because everybody has to agree on what the exchange says -- which we can't do unless the exchange publishes signed timestamped price notifications.  Of course we could have a "gateway" that takes raw data from the exchange and publishes a signed feed, but you're adding another un-accountable middleman with that approach -- we can only check what gateway said the price feed was.  But, you say, we could have multiple gateways, all publishing their own feeds and cross-checking each other.  But then you have a bunch of message spam for tiny price movements, so you have to impose transaction costs on the gateways so they don't publish trivial updates too often.  And it would be nice to have a method to remove gateways if they're misbehaving.

So if you take your idea to its logical conclusion, taking into account real-world engineering constraints, what you end up building is...exactly the system that's already in place, where the "gateways" in the above description are the delegates.

Oh, and btw, having everyone's client continuously contact a single third-party site is problematic because it imposes a lot of bandwidth costs on that site without compensation, inviting them to ban us or even driving them out of business.  Plus there's privacy implication -- if the site notices transactions to/from a certain address tend to appear on the blockchain when a BitShares client from a certain IP address is active, they've effectively de-anonymized someone.

« Last Edit: September 19, 2014, 12:53:10 am by drltc »
BTS- theoretical / PTS- PZxpdC8RqWsdU3pVJeobZY7JFKVPfNpy5z / BTC- 1NfGejohzoVGffAD1CnCRgo9vApjCU2viY / the delegate formerly known as drltc / Nothing said on these forums is intended to be legally binding / All opinions are my own unless otherwise noted / Take action due to my posts at your own risk

Offline theoretical

For that reason, I see value in providing moderate yields (if possible) that out-compete what people can get by holding the real world assets. So a 5% annual yield on BitCurrencies kicks the ass of any other bank out there. But why do we higher yields than that and why do we need to spend that money on yields for BitAssets where people could not get that kind of interest by holding the underlying, such as gold, oil, or BTC?

Because we want to kick ass in as many markets as possible.  You could invest in a gold ETF and get gold exposure, but the vault and management fees come out of investor funds.  Or you could invest in BitGold, get the same financial exposure and earn interest.  From BTSX investors' standpoint, the most likely path to wild success and great riches appears to be mass market adoption of BitUSD and/or BitCNY as a payment and deposit account product for ordinary consumers.  But that doesn't mean we shouldn't court commodity and currency exchange investors and traders as well.  Demand for BTSX as collateral, and transaction fees, is good for BTSX investors regardless of the source.  So stimulating that demand with yield makes sense -- especially for investments that don't normally have yields.

As for why we would want higher yields -- in theory, we would always want our yield to be the smallest needed to out-compete our competitors, convert the rest to BTSX and burn the BTSX.  In practice, hard-coding a yield now will make us brittle and inflexible if the real world sees a higher interest rate environment in the future.  And letting humans control yields makes the system opaque and political.  It would be an interesting experiment, but I think there are excellent marketing reasons to keep the more controversial aspects of the real-world financial system far from any blockchain we're going to advertise as part of the BitShares brand.

That being said, I would support selling some fixed fraction of BitAsset-denominated fees for BTSX which would be burned / given to delegates in the same way as other BTSX-denominated fees.  (We don't want to give the BitAssets directly to delegates in order to ensure BTSX investors to get a return.)

In my view it would be smarter to take the value that would have been spent on those yields and instead use them to either boost the yields of the BitCurrencies if necessary, or give them to the delegates so they can invest in growing the BitShares ecosystem faster. And yes, I know this complicates the code, but I think it is worth it to eventually build into the system.

That gets into a debate on the proper role and place of delegates.  I tend to view delegates as something that was a technical implementation decision undertaken mainly to have reliable ten-second block times, which (it may be argued) is a necessary precondition of having a liquid and dynamic on-chain market, which itself is a precondition to the whole BitAsset short/long system.  So I would support less active delegates with lower payouts, especially if we're giving them income as well.
BTS- theoretical / PTS- PZxpdC8RqWsdU3pVJeobZY7JFKVPfNpy5z / BTC- 1NfGejohzoVGffAD1CnCRgo9vApjCU2viY / the delegate formerly known as drltc / Nothing said on these forums is intended to be legally binding / All opinions are my own unless otherwise noted / Take action due to my posts at your own risk

Ggozzo

  • Guest
Why not just make the feed the end all? Don't let delegates publish feeds, let them decide which exchanges should be used to discover the median price. Then the median price feed from all third party exchanges along with their prices in real time should be displayed in the GUI. It should update as close to real time as possible. Then you just don't allow any order away from the median feed at that given moment to execute.  It will always be true to the outside value of BTSX. Not sure if it's possible but sounds good. This of course should only be done with the currencies.