Introduction to BitShares - Video

[clout]

The price within a market represents the current public consensus on the value of a given asset. Does the notion of termination enter into the stock market or for that matter the cryptocurrency markets. The price of these assets are reflective of both fundamental value as well as the prediction of increases or decreases in value. Its the same idea within bitshares. As long as any of the assets from which bitAssets are derived remain in existence there will not be a termination of that market for a given bitAsset, as participants in the market will always be altering and adjusting their positions based on newly formed predictions for the market given changing information pertaining to the market. The notion that markets need to have a termination date is inherently illogical.

You say that "the crucial difference with Bitshares" and prediction markets "is the lack of a termination or settlement date when the winners and losers are know with certainty." The assumption that winners and losers are know with certainty is probably where you are confused. In a fluid market such as the stock market there are not certain winners and losers, as the price on a given day can change direction on following days. The losers on one day can become the winners on another.

[Yui Xie]

Prediction markets have been around for 50 years and there are plenty of examples that show how effective they can be.  The crucial difference with Bitshares is the lack of a termination or settlement date when the winners and losers are known with certainty.  I (and Winslow Strong earlier in this thread) asked for an example of a successful non-terminating prediction market that had been implemented in the past.  Since there was no satisfactory response (opening and closing positions in a terminating futures market is not valid, in my opinion) I started researching it.  The only thing I could come up with was the simExchange:

http://en.wikipedia.org/wiki/The_simExchange

that used prediction markets for video game sales.  Players used virtual money and the exchange was composed of both terminating (futures) and non-terminating (stocks) trading entities.  Futures trading was shut down in 2010 leaving only the non-terminating component.  The site then fell into disuse (see Current Status in the above link).   

[clout]

If you simply do research on prediction markets you will see why this works. All markets are driven by psychology more than underlying "value" and fundamentals, particularly in the short term. Prediction markets don't require and underlying asset, but their efficacy is not experimental, it is academically proven, which is why you have professors at Princeton attempting to create prediction market using bitcoin. Bitshares X essentially does this on the network and through the block-chain in order to cut out any potential intermediaries. 

[Yui Xie]

Agreed -- the overwhelming number of futures traders do not take delivery.  But somebody does.  There is a clearly defined settlement date and an actual product changes hands.  If not, there is absolutely no point in having a futures market.  For this reason, Bitshares is NOT the same thing.  It is an extrapolation of the futures market that lets time -> infinity, with no deliverable and no price pegs.  Relaxing these parameters is not trivial.  Since this has never been implemented before on either a de-centralized or centralized exchange, you are making some crucial assumptions about market behavior.  As such, it should be viewed as a risky and still unproven experiment.

[bytemaster]

I can't get my mind around the notion that a prediction market will converge to a correct value if there is no causal link to the delivery of a real commodity at some point in the equation.  Without it, it just looks like a collective herd mentality.  This sure seems like a system that could be manipulated, as indicated in earlier posts by Winslow Strong and by others in different threads.

I'm not saying it won't work, I'm just skeptical.  My skepticism would be greatly diminished if someone could point me (us) to an example of where this has worked successfully in the past.  It would be easy enough to implement in a centralized exchange, there's nothing special about it being decentralized in the BitShares model.  And no, I will not accept that successively opening and closing positions in the futures market is the same thing.  I need an example of a predictive market that is non-terminating with no deliverables.  If you insist that this is just an extrapolation of the futures market, we'll just have to agree to disagree.

This price of PTS and AGS is reflective of the markets current assessment of the likely success.  We have no need to convince you, the market will prove it one way or another soon enough.

[Yui Xie]

I can't get my mind around the notion that a prediction market will converge to a correct value if there is no causal link to the delivery of a real commodity at some point in the equation.  Without it, it just looks like a collective herd mentality.  This sure seems like a system that could be manipulated, as indicated in earlier posts by Winslow Strong and by others in different threads.

I'm not saying it won't work, I'm just skeptical.  My skepticism would be greatly diminished if someone could point me (us) to an example of where this has worked successfully in the past.  It would be easy enough to implement in a centralized exchange, there's nothing special about it being decentralized in the BitShares model.  And no, I will not accept that successively opening and closing positions in the futures market is the same thing.  I need an example of a predictive market that is non-terminating with no deliverables.  If you insist that this is just an extrapolation of the futures market, we'll just have to agree to disagree.

[santaclause102]

Why would it be an advantage to have terminal dates?

Because at that point some party has to deliver a real asset.  The market must converge to a true price since a physical commodity is changing hands.

Are you sure you are getting oil or gold when you are betting on it with a centralized system?

I think changing physical things is not necessary and not done in existing betting markets. But what they have is an index that represents the market where the real physical assets are exchanged. 

Correct me if I am wrong: In a system without an index given by a central authority you rely on everyone's assumption that everyone will follow the price of the underlying asset (must be precisely defined) so that it makes sense to bet on where the asset's price will be in the near future. Near future here is defined for short positions as up to the point where their asset looses 50% and for long positions there is no termination (I dont see a problem in the latter).
There are two assumptions/implications for this to work:
(1) The asset is exactly defined (for example 1 barrel of WTI brand; or whatever is traded physically).
(2) It is habitual behavior where the shared understanding to follow the price of the underlying asset is reinforced the more people do it. And you win when you predict the future price that deviates from the current price right as long as no short squeeze takes place (when you are short and your asset looses 50%) -> What is therefore needed or at least helpful is reputation which creates mass / herd behavior. If a big bank puts out such a system the mass would easily rely on that everyone is following the underlying asset price

*I used the word "underlying asset price" as the price that is actually paid for a good when it is physically exchanged.

[Yui Xie]

Why would it be an advantage to have terminal dates?

Because at that point some party has to deliver a real asset.  The market must converge to a true price since a physical commodity is changing hands.

[santaclause102]

I'd also like to pose a question:  Have you stopped to consider why financial markets don't offer products like BitUSD?  I.e. a derivative without a terminal horizon, designed to track an underlying via a yield mechanism as the tracking incentivization?

Isn't a short position limited to 50% of its own value and a long position actually infinite (meaning when the respective bitasset get less and less valuable against bts its value falls logarithmically)?

i) But all futures/options markets have a terminal date.

Why would it be an advantage to have terminal dates? As I understand it you could always give up you short/long position and make a new one. I think of bitshares as a decentralized exchange for everything.
Would they be hard to implement terminal dates in bitshares if necessary?

[Yui Xie]

i) But all futures/options markets have a terminal date. 

ii) Even though the overwhelming majority of derivative traders don't take physical possession of the commodity, there's still the fact that some sort of tangible commodity/cash is delivered to somebody at the time of termination. 

BitShares don't have these two points and they may indeed be critical.  Not saying the idea won't work and you've already conceded it's an experiment.  We are simply asking if it (a prediction market with infinite time and no deliverable) has been tried on a centralized exchange.

[bytemaster]

I'd also like to pose a question:  Have you stopped to consider why financial markets don't offer products like BitUSD?  I.e. a derivative without a terminal horizon, designed to track an underlying via a yield mechanism as the tracking incentivization?

I asked for an example of a financial derivative traded on a market that tracks an underlying merely via a mechanism of yield-adjustment, without lump-cash settlements that compensate directly and linearly for price changes of the underlying.  I've never heard of such a product, and it's premise seems to me to be ill-founded.

Winslow Strong raises a compelling question here.  The concept of a predictive market has been around for over half a century.  BitShares, however, adds key new components: temporally unlimited with no commodity delivery required.  But this is not a difficult extrapolation.  Surely financial entrepreneurs would have considered such a construction at some point in the past.  I can't imagine why it wouldn't have been implemented on a conventional exchange.  Is there any example of this being done before?  If not, it suggests that the idea has not held up to market scrutiny.  And I don't see why having it de-centralized is the crucial missing piece.

It has held up to market scrutiny if you look at all of the 'futures markets' that trade in debt contracts backed by fiat.  What didn't exist before was a decentralized store of value (Bitcoin).    Even our so-called futures markets in Gold and Silver *settle* in fiat (sometimes forced even when taking physical delivery is requested).

[Yui Xie]

I'd also like to pose a question:  Have you stopped to consider why financial markets don't offer products like BitUSD?  I.e. a derivative without a terminal horizon, designed to track an underlying via a yield mechanism as the tracking incentivization?

I asked for an example of a financial derivative traded on a market that tracks an underlying merely via a mechanism of yield-adjustment, without lump-cash settlements that compensate directly and linearly for price changes of the underlying.  I've never heard of such a product, and it's premise seems to me to be ill-founded.

Winslow Strong raises a compelling question here.  The concept of a predictive market has been around for over half a century.  BitShares, however, adds key new components: temporally unlimited with no commodity delivery required.  But this is not a difficult extrapolation.  Surely financial entrepreneurs would have considered such a construction at some point in the past.  I can't imagine why it wouldn't have been implemented on a conventional exchange.  Is there any example of this being done before?  If not, it suggests that the idea has not held up to market scrutiny.  And I don't see why having it de-centralized is the crucial missing piece.

[santaclause102]

Just found this http://www.youtube.com/watch?v=-8ZJ3xTDwbI
I think it is more credible when the person that knows the system best speaks of it. Better than a random women that is not bad at presenting it but doesn't shine either....

[MrJeans]

I don't understand the mechanism by which bitUSD would track the exchange rate of paperUSD/BTS.

The market will price the asset at what they believe it is worth and want to buy slightly below this value and/or sell slightly above.
In order to understand innovations such as Bitcoin and BitShares one needs to let go of allot of the old ways of thinking we have learnt growing up.
For example, Bitcoin has no assets backing it, so when I speak about it to conservative bankers they say no intrinsic value or assets backing it means no value (some just scream ponzi scheme). But Bitcoin has value because of the service it provides, a concept that takes a while to conceptualize the first time.

Take a look at Litecoin. It is said to be the silver to bitcoin (being gold). And as a result the relationship between bitcoin and litecoin is akin to that of gold and silver.

http://thegenesisblock.com/understanding-the-gold-silver-ratio-and-how-it-may-apply-to-bitcoin-and-litecoin/

http://thegenesisblock.com/bitcoin-litecoin-ratio-returns-historic-norm-peercoin-climbs-200/

Having said that I believe litcoin is massively undervalued as a silver to bitcoin because of its faster transaction times. That being said it still tracks Bitcoin in a silver to gold fashion simply because its developers suggested that its silver to bitcoins gold. Cant even find that suggestion on the litcoin site anymore.

In a system such as BitShares, where a bitasset is brought into existence with the tangible asset price in mind, the tracking of the tangible asset price will be much more precise.

[bytemaster]

If I have MtGoxUSD then MtGox will allow me to withdraw it into my bank account into BankUSD.  MtGox basically guarantees this and this is why MtGoxUSD is "valuable".

If I have BitUSD who will allow me to withdraw BitUSD into BankUSD?  Is this a service that needs to be built?  Or is BitUSD only backed by BitShares?   It seems like if I own BitUSD no one will convert it directly to BankUSD without first converting it to BitShares.

Since I can convert MtGoxUSD into BankUSD easier than converting BitUSD to BankUSD, it seems like MtGoxUSD is more valuable than BitUSD.  Your thoughts?

BankUSD is a promise to pay a USD worth of value in BTS at the time of withdraw.   So you convert BitUSD to USD like so....   BitUSD -> BTS -> BTC -> BitStamp -> BankUSD.   Now that process has many stages and fees, but you could just as easily sell anyone BitUSD for USD on the street corner.   As holding BitUSD pays you interest, this interest can cover most of your transaction fees provided you hold it for a while.

MtGoxUSD has counter-party risk, the funds could be stolen, and you still have wire-transfer costs *if* you are able to withdraw it in a timely manner.   There is also KYC and loss of privacy.   GoxUSD doesn't pay you interest and you have withdraw limits.  If Gox is hit with a DOS you lose access to your funds.... etc etc...