Topic: [OFFICIAL]Bitfinex.com first Bitcoin P2P lending platform for leverage trading - page 163. (Read 723861 times)

Is there anyway to do a "stop-loss" when I sold my BTC on the "EXCHANGE" not "MARGIN TRADE".

Basically when it goes above a certain price, I want to do a market buy.

I did "EXCHANGE" because I don't need the extra leverage.

I am not RDMINER.  He must have copied my post - or tried to quote me and messed up.

There are some big whales going longer and longer if my analysis is right. They took huge amounts of loans.

It is pretty puzzling that the USD swap rate is falling so slow.  With the current bid/asks - I'd expect it to be around 0.08%/day, and to fall by at least 0.01%/day.

hy!
this does only show recent trades, why??

orderspayloadobject = {
         'request':'/v1/trades/btcusd',
         'nonce':str(long(time.time() * 100000)),         
         'timestamp':'1374943149',                     #Only show trades at or after this timestamp.
         'limit_trades':int('50')               #Limit the number of trades returned.
         }

thx

I understand. Got it. It works.

Thanks for your help!

Sorry for quoting my own question (it is getting buried), I just wanted to make sure Bitfinex team sees it.

Those "missing" tid are transaactions done in other currency pairs: ltcusd drkbtc etc.

hy, when using
https://api.bitfinex.com/v1/trades/btcusd
there are missing tids. are those cancelled orders?

Don't forget trust is an important factor when it comes to exchange.

Icbit.se comparisons are important, but I'd like to know a lot more about the platform and the people behind it before I'd use it.

He forgot that he was logged into his sock puppet

Are you kidding me?

Ente

No, the equivalent is 0.1% per day

Months ago, Giancarlo wrote about this when the behavior changed:

Lets assume I took five $10 swaps some time ago. You have a $100 offer up, which is the best offer, and with a better rate than my initial swaps. I close mine, and my position is refilled with $50 I borrow from you.
So far so good.
However, the engine makes me take five $10 swaps from you. Even though they are all from you, and opened at the same time.
Fast-forward a few months, where everyone has some position open, everyone has auto-lend active, and people replace old or expensive swaps with better ones: voila, the number of swaps grows infinite.

It is almost impossible to do anything against it, like "defragmenting", with the current engine behavior. Even when you wait until all your funds are returned, and then make one big offer again, people will come and take minuscule amounts because they replace minuscule amounts.

And with every "step" it all fragments more and more.

This behavior takes up a lot of my spare/bitcoin time, when I have to mark dozens, hundreds of tiny swaps to replace them. I do this for months now, and don't even want to know how many swaps I marked, replaced and fragmented since then.

And, that's why I asked for the feature of "combining swaps into one line with a +/- button" a few posts above.

Ultimately, I guess this exponential fragmentation will kill all servers where the engine works on, and maybe they change that behavior then ;-)

Ente

Thank you. Keep up the good work! I really hope BFX supports you as best as they can as you are providing the absolute best analyzing tool for any of the exchanges to the point where it becomes a real reason to stay and interact with BFX.

No all calculations in my first post were for april 1 2013 until today 12.00 UTC, typo

Here are the calculations for 01-01-2014 until today 12.00 UTC (free of charge  ):

USD
0.20398005772999340188 % per day
74,4527 % annually
2998,32641861725027 (average amount per swap)

BTC
0.06278128456304585053 % per day
22,9152 % annually
5,69183286033181 (average amount per swap)

LTC
0.043692172678537965439 % per day
15,9476 % annually
67,59164635514019 (average amount per swap)

Calculation: Sum (Transaction Amount * Transaction Rate ) / Sum ( Transaction Amount ) for the selected period. For me it is a few lines in a php script and run it against the db, however it is really hard on the db so I can't (for now at least) just put a page together that runs this query when a visitor specifies a period. It would demand WAY too much resources on my server and kill the rest of the site if a few people at  the same time would run it.  And I would have to write some HTML/CSS and I don't feel like doing that today  . I might do something like a weekly / monthly / quarterly / yearly / all time average soon. Quite a nice suggestion you did actually...  and put it in some tables and charts. for now I have to do it manually  

Oh my, you are fast! Nicely done.

You wrote:

"Between the 01-04-2013 and 26-07-2014 the average daily swap rate is: 0.24363676973540734048 % per day (I used Volume Weighted Average Rate) for the period 01-01-2014 till today 12.00 UTC"
So which is what? :-) ^^
For between the 01-04-2013 and 26-07-2014 the average daily swap rate is: 0.24363676973540734048 % per day
and for 01-01-2014 till today 12.00 UTC ? I started in January so that is what would come in handy for me.
I'm assuming the latter has to be far lower as the 1 % swaps from 2013 stopped nearly entirely and we had a long period (feb till May or something) for abysmal rates.

How did you calculate it? Is there an easy way to calculate that for myself so I don't have to bother you in the future? Unfortunately BFX doesn't offer this data themselves. It's probably the most important metric for new "investors" to estimate whether lending on bfx is worth it or not.

Between the 01-04-2013 and 26-07-2014 the average daily swap rate is: 0.24363676973540734048 % per day (I used Volume Weighted Average Rate) for the period 01-01-2014 EDIT: 01-04-2013 (dd-mm-yyy) till today 12.00 UTC. So un-compounded that would come to 88,9274 % annually. Do the compounding yourself  

ow, while I'm at it:

USD:
0.24363676973540734048 % (per day)
88,9274 % (annually)
$2282.455 USD average amount per swap

BTC:
0.057396292048379445646 % (per day)
20.9496 % (annually)
6,6623753816802 BTC (average amount per swap)

LTC:
0.039356676351540241087 % (per day)
14.365187 % (annually)
69,01023669836132 LTC (average amount per swap)

Question to Bitfinex technical team about "hidden orders":
It used to be that hidden orders would be executed with the price not originally set, but with the one offered by counter-party. What was happening is that if let's say a hidden order was a sell for $600, but there is a bid for $600.5, seller (owner of hidden order) would get offered $600.5. In this case the "hidden order" indeed remained hidden. To compensate for such advantage, hidden orders were also having lower than regular priority in terms of their execution (meaning that hidden orders are executed after the regular ones, or per some fractional algorithm, you guys came up with).

Right now, hidden orders can be VERY easily discovered. Thus, if you don't see the last execution price matching current bid/ask, you found the "hidden order", so it instantly becomes no longer "hidden". Execution priority right now is also the same as for regular orders, so you know that until hidden order for that particular price "runs out", price is not going to move. The only thing which remains "hidden" right now is the size of such order. For this reason, people may even not want to place their orders, but wait until situation becomes more clear. Moreover, since the use of hidden orders is a rather heavily used feature at Bitfinex (I know this for a fact from personal experience), traders may be discouraged from more active trading and this in turn may be one of the reasons which makes our current trading so incredibly "thin". We have some $30 mln in active swaps, yet at times just a few BTCs can move the price several points.

I really liked the old model a lot better, mainly because such orders (even if they are big) did not interfere with natural price movements that much. I also have a strong feeling that the old model made trade patterns look a lot "smoother" (so to speak). Besides, and as I am sure you know, there are also a lot of concerns out there about order book being very "thin", so the observers out there are comparing our (comparatively speaking) "thin" order book with that huge borrowed swaps volume and naturally concluding that the flash-crash is imminent. If only our hidden orders would be truly hidden as in earlier days, I am pretty sure that there will be a lot more of them on the order book on both sides and we will have curves instead of current rectangles in the trading patterns. Right now even the 10 - 15 BTC orders look (and often behave) like "walls".

Were there any significant reasons for the change in structure of hidden orders?

Thanks.