ICBIT Derivatives Market (USD/BTC futures trading) - LIVE

[Stephen Gornick]

Clearly you can't determine the settlement price when there are 4 hours left in the settlement day.

A (24 hour) volume weighted adjusted price is just that -- the VWAP for the trailing (24-hour) period.  It can be any 24 hour period of time.  The contract doesn't say 00:00 to 23:59 UTC of settlement day, it says the "volume weighted average price", and the settlement occurs at 20:00 UTC.   So the 24 hour WAP goes from 20:00 on the day prior settlement day through 19:59 of settlement day.   [Update: Since the contract doesn't actually say "24 hour", it could be ambiguous as to which VWAP was intended ...  24 hour, monthly, or what?  Previous BTC/USD contracts were using the 24 hour VWAP.]

Now I suppose a script can be run to check the numbers, but settlement happens in real-time.  There's no ability to go pull the raw trade data to come up with the WAP in real-time.   If that computation is done later and comes up with a different number, then it needs to be analyzed why that is.

I can see the potential for a problem though.  If Mt. Gox exchange were to misreport the weighted average, possibly from a staff member trying to manipulate the ICBIT settlement price, then it would be hard to prove what value was showing.   I can see the need to have a separate computation performed later against the raw trade data for verification purposes.

[1714009299]

1714009299

[1714009299]

1714009299

[zebedee]

The contract was settled at 20:01 UTC, just like the contract says.

Link?

[Stephen Gornick]

The contract was settled at 20:01 UTC, just like the contract says.

Link?

I didn't have any BUH3 myself so I don't have a log entry.   At a minimum it settled before 20:29 UTC when the Tweet was sent.  And before 20:39 UTC which is the timestamp when Fireball's forum was posted here.

I did personally see open buys and sells for it right up before 20:00.

So I don't know how externally one could verify the exact millisecond the settlement numbers were reported.  The settlement price though is as-of exactly 20:00 UTC.   There is evidence that this settlement price was obtained between 20:00 UTC (when trading for it was no longer possible) and 20:29 UTC (the time of the tweet), and other traders have reported this timestamp in their logs:

I did and here's the log entry:

2013-03-15 20:00:01   332   Variation margin, last = 47.0764

as well as mutliple traders confirming that at the time of settlement (or minutes after) the 24 hour VWAP price from Mt. Gox was at that 47.0764 level, including:

I can also confirm that I looked at MtGox at the time of settlement, and the price was correct at least to four digits (two decimals after the point).  

[Edit:  OK, maybe I'm overlooking something basic and am missing your point entirely.  Are you instead asking for the link that shows when settlement happens?

Trading sessions
Clearing happens at 20:00 GMT, clearing duration is flexible.

- https://icbit.se/futures  ]

[zebedee]

The contract was settled at 20:01 UTC, just like the contract says.

Link?

I didn't have any BUH3 myself so I don't have a log entry.   At a minimum it settled before 20:29 UTC when the Tweet was sent.  And before 20:39 UTC which is the timestamp when Fireball's forum was posted here.

I did personally see open buys and sells for it right up before 20:00.

Right, that's my point - there isn't any.

Please stop referring to clearing times, there is no relationship at all as there needn't be any.

e.g. LIFFE Libor futures clear in the late afternoon / evening, but settlement fixing determination happens at 11am.  Clearing doesn't happen at 11am.  There is no link between clearing and settlement price determination.

Hence there is no reason, given the existing contract wording, that settlement price should be determined at 8pm and not midnight, given the prevalence of UTC on the website and the fact it refers to the weighted average price on the settlement day.

All they need to do is add "at 8pm" to the contract wording, and calculate the price themselves with evidence, for me (and anyone of reasonable mind) to be happy.

[killerstorm]

Suppose I have negative amount of money on account on ICBIT.se...

Should I feel morally obligated to settle this debt, or can I just walk away from it?

What are possible consequences?

Can exchange operators go after people who owe to ICBIT.se and label them as scammers on forum, for example?

What's more worrying is that my position is still open while I have negative amount... So I'm getting more and more debt as Bitcoin goes up. But I can also get back into profit if exchange rate plummets =)

[Fireball]

Suppose I have negative amount of money on account on ICBIT.se...

Should I feel morally obligated to settle this debt, or can I just walk away from it?

You can walk away but that does not mean you're forgiven ;-)

[picobit]

Suppose I have negative amount of money on account on ICBIT.se...

Should I feel morally obligated to settle this debt, or can I just walk away from it?

What are possible consequences?

Can exchange operators go after people who owe to ICBIT.se and label them as scammers on forum, for example?

What's more worrying is that my position is still open while I have negative amount... So I'm getting more and more debt as Bitcoin goes up. But I can also get back into profit if exchange rate plummets =)

This was essentially what I asked about a week ago or so.  I am worried about getting in the same situation if the market crashes, and was assured that the "margin call" mechanism should prevent my account from going negative by closing the positions.  I really hope this is true, because on one hand I do not want to run from my debt, on the other hand I have already placed the amount I am willing to risk on the site.

By the way, I guess that calling it a "margin call" is not really correct, it is more like a forced closure.  A margin call would imply contacting the customer and asking for more money to be deposited before the account goes negative.

[Stephen Gornick]

All they need to do is add "at 8pm" to the contract wording,

To me it was implied that settlement would occur at the end of the trading day (which ends at 20:00 UTC) at the VWAP at the time of settlement.  But I suppose that could be ambiguous so your point is valid.   It doesn't mention what duration of VWAP is used (24 hour) nor at what time it is taken (end of trading day) so including both of those would reduce any confusion.

[killerstorm]

and was assured that the "margin call" mechanism should prevent my account from going negative by closing the positions.

It doesn't. I currently have -0.05 BTC on account, and it was negative for a week or so.

As I have negative initial margin it doesn't allow me to send orders which would close my position.

So to close it I'll have to deposit some money.

[Stephen Gornick]

As I have negative initial margin it doesn't allow me to send orders which would close my position.

If that is correct, that's new-ish.

[picobit]

and was assured that the "margin call" mechanism should prevent my account from going negative by closing the positions.

It doesn't. I currently have -0.05 BTC on account, and it was negative for a week or so.

Ok, negative amounts of fractions of a BTC (or even a few BTC) do not worry me

As I have negative initial margin it doesn't allow me to send orders which would close my position.

That, on the other hand, is bad.  And a bit strange, it could be because you have other open orders "in the same direction", possibly outside the current trading range.

[Fireball]

It doesn't. I currently have -0.05 BTC on account, and it was negative for a week or so.

As I have negative initial margin it doesn't allow me to send orders which would close my position.

So to close it I'll have to deposit some money.

What's the rejection status stated? It should allow such orders, of course, I tested this scenario numerous of times.

[killerstorm]

What's the rejection status stated? It should allow such orders, of course, I tested this scenario numerous of times.

It works now. Perhaps I just forgot about orders I already had...

[Fireball]

Trading range enhanced, and some shorts had to be forcefully closed at ~ $53-$54 range (the level above which one person who shorted becomes bankrupt). It'll be shown on the bar chart.

[random_cat]

Trading range enhanced, and some shorts had to be forcefully closed at ~ $53-$54 range (the level above which one person who shorted becomes bankrupt). It'll be shown on the bar chart.

The forcible close may be a problem of your own creation.

The page https://icbit.se/futures clearly states "Clearing happens at 20:00 GMT."  By clearing early you may force positions closed while additional deposit of funds to avoid margin call is still being confirmed.  Do traders not have a reasonable expectation that the exchange follow the letter of its contracts?

[random_cat]

Also since info@icbit.se appears not to respond to email, I will note here (as Stephen Gornick mentioned in passing in the icbit.se, the bucket shop thread https://bitcointalk.org/index.php?topic=125376.40) that there is certainly a rounding error issue in variation margin calculation.  I have tried to account for it
in a few ways, but none of my techniques match the errors I see.  At this point I speculate errors or insufficient precision in floating point operations though it is possible I did not find the icbit algorithm that amplifies a 0.5-1 satoshi error.

[Fireball]

Also since info@icbit.se appears not to respond to email, I will note here (as Stephen Gornick mentioned in passing in the icbit.se, the bucket shop thread https://bitcointalk.org/index.php?topic=125376.40) that there is certainly a rounding error issue in variation margin calculation.  I have tried to account for it
in a few ways, but none of my techniques match the errors I see.  At this point I speculate errors or insufficient precision in floating point operations though it is possible I did not find the icbit algorithm that amplifies a 0.5-1 satoshi error.

Thanks for your email, of course I got it. This issue was queud to be thoroughly examined, but I highly doubt there is an error because I personally thoroughly tested this, calculating everything manually and comparing results.

[Fireball]

By popular demand, the real-time chat was added to the web trading. Web caching is too aggressive in caching .css styles, so if the layout appears broken - clear browsers cache.

At least, this way we can also talk live.
However, please note, that I appear only by the username "admin" there. I will add some color distinction, but don't forget that phishing may very well be possible (someone could register admln or something).
Also, the chat is public, so never give up any personal details. Logs are not published, though.

If your username appears as "usrXXXXX" where XXXX is some number, then please click Logout and then login again. This will update your username to the correct one.

[davidoski]

Why there's no sell orders at present?

[Fireball]

Why there's no sell orders at present?

That happens when everyone is too optimistic about rate moving upwards

Seriously, it means that traders don't find it economically viable to sell inside the given price range. The solution usually is to make an additional clearing to enhance trading range, which is going to be done soon.