Topic: [OFFICIAL]Bitfinex.com first Bitcoin P2P lending platform for leverage trading - page 125. (Read 723903 times)

soon (adverb, sooner, soonest)
A term invented, patented, copyrighted and trademarked by ButterflyLabs Inc., extending the word's original meaning (see ancient dictionaries for reference) by several months, sometimes years.

Just saying

So, I keep returning to the FRR problem in my head. Seems pretty clear that "abolish it" isn't really on the table, so... next best thing, make it better. To that end, I'd like to offer up for criticism a simple(ish) model for a self-adjusting rate, not based on an slow-moving average.

Principles/assumptions:

• (1) Ignore fixed-rate swaps, they're set in relation to the FRR, so basing the rate on them is just going to produce incestuously self-referential results
• (2) Also don't try to anticipate demand based on the price of bitcoin, that sounds like a hideously complicated pattern-matching game and is liable to be horrendously glitchy
• (3) Instead try to use supply/demand directly - if the available FRR funds are being taken then the rate should increase, if they aren't then the rate should decrease
• (4) Most of the time only small-scale adjustments are needed to fine-tune the rate to recent activity, but there are occasional major movements that will require major adjustments
• (5) Setting a floating rate without reference to the fixed-rate activity is bound to end up using arbitrary magic numbers, which can be used as parameters to tweak and tune the model - fiddle with the dials and experiment to see what happens and make it work better. Those will be marked out with

Actual mechanics:

• For small adjustments: , if the sum of funds available at the FRR has increased (more new offers than swaps taken/offers cancelled), decrease the rate by <1%>
• Conversely if the wall has shrunk (more funds taken than the supply of new offers), increase the rate by <1%>
• For large upward adjustments: if the sum of available funds at FRR drops then increase the rate by - limit this to happening at most
• For large downward adjustments: if the volume of swaps taken at FRR is less than <5%> of the overall swap market over the last <24 hours> then decrease the rate by <25%> - limit this to happening at most

The small adjustments should keep the rate orbiting around the point where supply equals demand, or if there's a clear and sustained direction (wall continually draining away slowly or a continually creeping increase) then these changes can add up to produce a moderately sized change per day and a fairly large change across several days.

Then for those times when the small adjustments prove to be insufficient to balance supply/demand, we look for clear signals that the rate needs to be moved significantly.

For the upward direction, I'm looking to eliminate the ever-frustrating situation where the wall is finally gone, stupidly high fixed-rate swaps have been taken... and then funds pop back up at the still-low FRR and stop any continued rise in rates. Exactly how big an increase is needed can be experimented with and possibly ought to depend on where the rate started - doubling from 0.1% seems reasonable in a way that doubling from 0.8% would not... but then, the FRR isn't often at 0.8% and if all the funds were taken despite that high rate than some sort of upward jump would still be justifiable.

For the downward direction, I'm now looking to mitigate the situation where the FRR gets stuck too high and doesn't get taken at all. The "5% of total daily volume" figure is pulled from nowhere and should probably be a different number, but the principle is "If no-one's taking FRR swaps at all, the rate is far too high". Also not sure of the proper size of that adjustment - the obvious mirror image of cutting the rate in half feels like too much, so I went for a change of half that magnitude.

Thoughts on fine tuning the parameters:

• The large-scale adjustments could perhaps be made smoother by reducing the size of the change but allowing them to happen more often. Although with a strong signal that the rate is in the wrong place, you do want the adjustment to be significant - not just a few percent.
• The FRR wall being completely gone might be too high a bar; maybe instead use the size of the wall dropping below some low value (so that one guy with $50 in funds can't stall a much-needed rate increase... not that he'd be able to do so for long if the rate really was too low)
• The size of the small adjustments don't necessarily need to be the same amount for upward and downward movements - seems fair to keep it symmetric but if one turns out to be much more common than the other (maybe the wall tends to slowly shrink and quickly grow or vice versa) then the size of the change should be adjusted to aim for a steady rate. Or possibly the size of the increase/decrease should be linked to how large the change in the size of the wall was.
  

So, that would be a margin sell stop with the trigger price set to 340 and an equally sized margin buy stop set at 350. But I'm not sure whether it's possible to set up the orders in advance such that one of them being executed triggers the other one being placed (as you would want here - no sense having the 'stop loss' order in place before the first order has executed).

There's "one cancels the other" orders, but I don't remember there being "one places the other". I think you would have to set up the first stop that enters the short then wait for it to execute before you set up the stop loss. But you can at least add an email notification for the execution of the first order so that you're reminded.

Hi, I'm new on the site and I don't know how to use it.

I would like to know if it is possible to create a short order (using margin trade) that would be triggered only if the BTC price fall below 340 , with a stop loss at 350.


Thanks for your help

you mean that there is no reserve to borrow at that time, so I can't complete the order?

Probably not.
If you're away from the computer, and want to be guaranteed your order will go through, you can reserve margin ahead of time here:  https://www.bitfinex.com/credit/index/btc.

Temporarily running out of reserves can happen on any exchange during high demand times, though it is pretty rare, as I'm sure others here will tell you.  Just reserve, and you're guaranteed a rate and execution.

Will it happen next time?

The demand was high there for a while.  It's back to normal and you can short again.
You can even get a higher short now.

It shows the order fully completed but It happened nothing at last, and it shows the order completed on bitcoinwisdom also, what's the matter?

And the status of order history is "canceled no reserve"

http://i58.tinypic.com/24q1gg7.jpg

Plenty of coins available again to short with below FRR.

I know it isn't BFX's model, but it makes it a very lopsided exchange when the liquidity to short BTC dries up.  Appears BTC were reserved to cause a run up, but over 1K have since been released again (without closing a position from what I see).  Nothing new.

No we can't, both are obviously responsible for the market. And the swap fee also plays a part in it.
On a side note, stop your immature troll-style postings please, I doubt your rage crap will have positive effect on convincing the Finex guys of anything.

I'm sorry, I had to have 2FA disabled on Bittrex as well.. I got the emails confused and yes you guys did contact me that day. Sorry about the inconvenience..
Regards,
Brian

Yeah, right... Despite what you wrote after this post I'm going to stand by Qwerty's (and probably others) side and say: Removing the FRR from patient swap-market (aka leverage-*FUND*ing) is removing cancer from the patient, I don't see how that would be removing the patient.

Cancer blocks communication between cells - those too unfrequented soon die off, they leave the market of available cells in that body, so to say. Not enough cells left and the patient dies.

Applied to the FRR: it blocks the communication between swap-providers - which of course translates to communcation between cells on a larger scale.

Now I don't know the numbers, I'm throwing darts which may be off by a mile but could you maybe share these stats with us? How much of (percentage-)growth has BFX seen during the past 12 months:
  1) in number of users
  2) in invested capital
  3) and in amount of swap provided

Have these numbers correlated over time or is one of them diverging to the south?

Without trying to be rude, to me the real patient here is BFX itself. And I'm not saying you're not listening to us, the opposite has often been the case, but we're just trying to be the hobby-doctors you're asking us to be so will you please let us do our job and let us discuss freely and quit telling us to use a band-aid when the patient has cancer?

Edit: Typo

Also, if you're referring to my proposals as removing the FRR entirely, nope: converting the swap-market into a "fund" is turning the FRR (and only the FRR) into what is currently the entire swap-market, only the numbers would be based on other factors (of which I talked at length). The noise around the FRR may be the real cancer here after all. Oh the pointlessness

Are you fucking serious? Why are you playing word games with us? If you're insisting on framing the debate as "how can we improve the FRR, regardless of whether it's good for the platform?" instead of "how can we improve the platform?", then YOU ARE ASKING A WORTHLESS QUESTION. Don't fault us for being generous in our interpretations rather than outright assuming that you're an idiot.

I like how you try to paint QwertyCore as the only one suggesting the removal of the FRR when MANY OF US HAVE AGREED WITH HIM.

Hmm, nice hyperbole...Ebola or Mass Murderer.

Either way, what you meant to say was YOU do not like the FRR, and you don't have to offer funds at the FRR. You can offer swaps at whatever rate you like. It isn't "dangerous" and currently, there is quite a spread between rates people are willing to take, and rates people are offering. I can find two or three comments just in the last two pages of people who LIKE the FRR. So, not EVERYONE is against it. Either way, if you don't like it, don't use it. If rates are too low, you don't have to offer swaps. That, again, is the point of a market, which is what we have, a market for swaps. It isn't a fund, it doesn't exist to bring the returns you may want, but it is an option for those who do want to use it.

The main issue, in my opinion, of those who dislike the FRR, is that they want something different from what reality currently offers. I for example, want an iPhone that has a one week battery life. Unfortunately, that is not an option. Markets don't give you what you want, they allow you to choose the best option available. So, the fact that the FRR is too popular (if no one used it, would we even have this discussion), means that a lot of people like that option. We are tweaking it, to make a popular tool more useful. Again, if you don't like it, build a bot and manage your swaps manually, I believe you will probably get a better return over time.

Ok so to continue on the analogy, you have a mass murderer who committed a genocide and plans to commit more, people tell you to not save him, and you decide to make every effort to save him anyway, although the rest of the world would rather not have him on earth ?

We do not want you to improve FRR, we want to get rid of FRR.



*EDIT : I read after your other reply to someone else, and it is clear, the FRR is Ebola, everyone acknowledges it, you could eradicate it, but instead your are trying to get us to help you to make it less dangerous. Well, no thanks.

Wow!  Just absolutely, unbelievably, freaking WOW!

You win, I give up.

Again, you are mixing levels. You are seeing the patient as the swap market, you are answering the question "How can I heal the patient?". We are not seeing the patient as the swap market, we see the patient as the FRR. We are asking, "How can I heal the patient?".

The fact that you are having a conversation, that is in no way the same as the conversation that we are having, shows why there is so much confusion.

I understand that that is the question you want to answer, but it is not the question we asked. This is what I have tried to explain, over and over, that you are answering a question, and providing a solution to a question that we are not asking. We asked how we could improve the FRR, not how we can improve the swaps market. I understand that you, and others, may have suggestions on a wide range of topics, and I'm sure they are well-thought out and probably very useful, but in this specific case, we are looking to answer the question "How can we improve the calculation of the FRR?".

So, for example, getting rid of the FRR is not a good answer to the question that we are actually asking. Saying something like, "Weighting it to take into account more recent activity more than past activity." IS a good suggestion, or at the very least, answering the question we asked, and participating in the conversation that we are involved in.

Hey Brian,

I found the email, it looks like it was responded to. I'm not sure if there was an issue after that. Specifically, John asks you to send the request from the email associated with the account, this is one of security measures in place to ensure that only the account holder can make a change to 2FA settings. I am looking into this further, but it appears, since John asked for an email from the address registered with the account, that the email you used was not it. Could you try to send an email to support from the other email address you mentioned in your email?

I will check the support emails, and make sure to hurry this along, but it appears that we are just waiting on an email from the email address associated with the account.

-Josh

Wow!  Do you really believe this?

Killing the patient is analogous to closing the swap market and no one has suggested that. Removing the FRR is not killing the patient, it is removing the disease.

The patient hosts the disease.  Removing the disease does not is any way remove the host.