Author

Topic: [OFFICIAL]Bitfinex.com first Bitcoin P2P lending platform for leverage trading - page 340. (Read 723990 times)

legendary
Activity: 1870
Merit: 1023
Two things to track to help decide what kind of leverage ratio is sustainable:
1) percent of time Bitstamp is down  (and does this happen more or less often during major price swings)
2) how often BFX runs out of its balance on Bitstamp.

Connecting to another exchange is of course top priority as it would fix this.

That said, it could be that the internal BFX exchange is developed enough to handle closing positions on its own?  I know it at least more capable of doing this than in April.
sr. member
Activity: 391
Merit: 250
I give you guys at Bitfinex A+ for the effort. If I were to describe - in very simplistic terms - your attitude towards us, the users, I'd use one word: CARE
legendary
Activity: 1470
Merit: 1007
Margin requirement = STDDEV(trades of last x days/hours) * scaling_factor for example... Smiley

Then they could allow nearly any leverage, as the limiting factor would be the margin required for new positions (10% margin means in the end you can only go to 10:1 in practice, as a 100:1 leverage has to be bolstered by 10 units, making it 100:10 in the end). In calmer phases that means people get larger leverage (meaning hey still increase their risk to loose all) while if the price is more volatile, it already IS volatile, so there is no need to enforce it via leverage.

very good suggestion.
Not only this one, but also the previous one where lenders can choose at which level of leverage they want to borrow.
I'll talk to Raphael about it and we will see how to implement it.

Have a good day

Giancarlo
Bitfinex Team


Both are good suggestions, IMO. Thanks for monitoring this thread and picking them up, it's what makes bitfinex outstanding I believe.

Personally, I like the 'lender choses max leverage' a bit better... in the end, I'm all for the choice of market participants -- if both sides are comfortable with a higher risk, let them do the deal.

probably would be wise to add a warning though, along the lines of "allowing traders to take a leverage higher than x:1 brings along substantial risks to your capital, consider yourself warned."
hero member
Activity: 763
Merit: 500
Margin requirement = STDDEV(trades of last x days/hours) * scaling_factor for example... Smiley

Then they could allow nearly any leverage, as the limiting factor would be the margin required for new positions (10% margin means in the end you can only go to 10:1 in practice, as a 100:1 leverage has to be bolstered by 10 units, making it 100:10 in the end). In calmer phases that means people get larger leverage (meaning hey still increase their risk to loose all) while if the price is more volatile, it already IS volatile, so there is no need to enforce it via leverage.

very good suggestion.
Not only this one, but also the previous one where lenders can choose at which level of leverage they want to borrow.
I'll talk to Raphael about it and we will see how to implement it.

Have a good day

Giancarlo
Bitfinex Team


Thank-you for being more active in the thread as of late.
sr. member
Activity: 446
Merit: 250
CAT.EX Exchange
Margin requirement = STDDEV(trades of last x days/hours) * scaling_factor for example... Smiley

Then they could allow nearly any leverage, as the limiting factor would be the margin required for new positions (10% margin means in the end you can only go to 10:1 in practice, as a 100:1 leverage has to be bolstered by 10 units, making it 100:10 in the end). In calmer phases that means people get larger leverage (meaning hey still increase their risk to loose all) while if the price is more volatile, it already IS volatile, so there is no need to enforce it via leverage.

very good suggestion.
Not only this one, but also the previous one where lenders can choose at which level of leverage they want to borrow.
I'll talk to Raphael about it and we will see how to implement it.

Have a good day

Giancarlo
Bitfinex Team
legendary
Activity: 1008
Merit: 1000
Margin requirement = STDDEV(trades of last x days/hours) * scaling_factor for example... Smiley

Then they could allow nearly any leverage, as the limiting factor would be the margin required for new positions (10% margin means in the end you can only go to 10:1 in practice, as a 100:1 leverage has to be bolstered by 10 units, making it 100:10 in the end). In calmer phases that means people get larger leverage (meaning hey still increase their risk to loose all) while if the price is more volatile, it already IS volatile, so there is no need to enforce it via leverage.

This seems like a much better way to address the risk.

Any reason this doesnt work?
legendary
Activity: 2618
Merit: 1007
Margin requirement = STDDEV(trades of last x days/hours) * scaling_factor for example... Smiley

Then they could allow nearly any leverage, as the limiting factor would be the margin required for new positions (10% margin means in the end you can only go to 10:1 in practice, as a 100:1 leverage has to be bolstered by 10 units, making it 100:10 in the end). In calmer phases that means people get larger leverage (meaning hey still increase their risk to loose all) while if the price is more volatile, it already IS volatile, so there is no need to enforce it via leverage.
legendary
Activity: 1870
Merit: 1023
About a week or two ago I was thinking that Bitcoin was becoming more stable and they could probably increase the leverage ratio.  It's got to eventually become more stable, right?

On the other hand, demand and supply for lending USD on BFX is already growing rapidly - so I'm guessing this isn't going to put much of a damper on growth.

I wonder if they could develop an algorithm for changing the leverage ratio based on volatility?  Past volatility is a decent predictor of future volatility.
legendary
Activity: 1638
Merit: 1001
₪``Campaign Manager´´₪
I don't think bitfinex is "making decisions for the traders".  (Ok, it restricts options for the traders, but on the other hand the lenders are more protected). They are just looking at the volatility, which has increased lately, thereby increasing the chance for a big liquidation.  It makes sense to me to lower leverage in that scenario.  

And yes, there is the insurance pool, but it is always occupied.  The rate at which it grows due to interest is way less than the increase in lending amounts that has gone on in the past months.
hero member
Activity: 763
Merit: 500
legendary
Activity: 2618
Merit: 1007
I also would expect Bitfinex to push more BTC to Bitstamp in case they think a correction is near rather than narrowing the margin (is 2:1 even worth the hassle to select where to lend from etc. compared to just do a 1:1 trade and be done with it?) to be able to sell into the correction as quickly as possible once positions get liquidated. Also they might increase margin requirements (e.g. 15% instead of 10%) if they are afraid that they are not able to sell fast enough.

I would be happy to lend my money to someone trading 10:1 with 10% margin, if I am assured that Bitfinex has a proper connection and a dedicated server monitoring prices and trading on Bitstamp (ideally on a VPN, whitelisted by any firewall/load balancer from Bitstamp side).
I might charge high(er) interest, like about 1-2% per day for the higher risk though.

It sucks that there seems to be no way for me (aside from private deals) to allow people who want to, to take this lending offer. Maybe we could also specify not only the lending duration (a not so good concept anyways, in my opinion...) and rate but also the max. leverage we allow our borrowers to take, maybe even up to more ridiculous ones like 20:1 - there the current margin requirement alone though would anyways not allow this...

If you get nervous, just up the margin requirement (in the current case it would be 50%) for new positions and deal with the protests from traders and lenders. Removing options however is not so cool in my opinion.

Also, why is the margin on BTC also reduced? It would make more sense in my opinion to even increase it there, as this would counterbalance your potential losses if you can't sell BTC fast enough. It seems you are more afraid of people dumping BTCs than someone suddenly buying up 10 million USD of coins with a market order...

About trading on margin elsewhere:
Kraken wants to allow trading on margin once they have more volume - unfortunately they seem to have weird developers which don't like bots so the API sucks as it is binning data unnecessarily. It might be enough for some more general arbitrage scripts but I doubt that anyone will write liquidity providing bots for that unprecise mess soon until they get their act together there.
There is always the chance to do the lending manually, but then again there is a VERY high need for trust - I wouldn't do that without a written contract and meeting anyone that wants to do that in person before transferring any funds in USD or BTC.

On a different note:
Please publish audits, if (when!) you do them, it is quite vital for people to see that you are actually able to cover the funds deposited and a positive audit statement from an independent 3rd party is definitely helping there. If you don't have any to publish, please consider doing one as soon as possible.
legendary
Activity: 2413
Merit: 1003
Are there other sites where you can get greater than 2:1 leverage?

BTC-e has 3:1 leverage on MT4 platform.

Plus500 has CFD with 7:1. Only fiat. And I don't think they accept US citizens.

BTC-E on MT4 is also only Fiat. So you wont have more leverage than trading 2:1 with BTC at Finex (at least on a long position)
sr. member
Activity: 391
Merit: 250
I am ok with the change, as long as you don't keep it more than necesarry. You're doing a good job so far. Also, Raphy, is that No Script (google issue) something to worry about?
In summary: if you're as concerned about maximizing security as I am, get NoScript, and *don't* allow googleusercontent. The site will look a bit worse, but trading still works. If you didn't use NoScript anyway, nothing changes Smiley

Thanks, I'll take a look into it.
legendary
Activity: 1470
Merit: 1007
I am ok with the change, as long as you don't keep it more than necesarry. You're doing a good job so far. Also, Raphy, is that No Script (google issue) something to worry about?

since I brought it up (in my question), hope it's okay if I (try to) answer it:

No, absolutely nothing to worry about. It's just about maximizing the security on *your own* side. I trade only in a Linux environment, since it's safer than Windows. And for that reason, I use the NoScript addon.

But in order to see the bitfinex webpage in all of its beauty, I would have to allow NoScript to run content from googleuserscontent. In particular, it seems some page fonts are stored there. Nothing essential.

In summary: if you're as concerned about maximizing security as I am, get NoScript, and *don't* allow googleusercontent. The site will look a bit worse, but trading still works. If you didn't use NoScript anyway, nothing changes Smiley
sr. member
Activity: 391
Merit: 250
I am ok with the change, as long as you don't keep it more than necesarry. You're doing a good job so far. Also, Raphy, is that No Script (google issue) something to worry about?
hero member
Activity: 763
Merit: 500
Doesn't look like the change is a popular one?  Any chance of reverting this back?
member
Activity: 111
Merit: 10
Bitcoin isn't volatile at all!!!  its steady at 300 for hours...
hero member
Activity: 492
Merit: 500
Are there other sites where you can get greater than 2:1 leverage?

BTC-e has 3:1 leverage on MT4 platform.

Plus500 has CFD with 7:1. Only fiat. And I don't think they accept US citizens.
legendary
Activity: 1008
Merit: 1000
I mean, I do agree with the better safe than sorry approach, especially with everything that has gone on in the last month.  Maybe they could play with the leverage and margin requirements and come up with a better happy medium.

The obvious risk is a flash crash forcing margins.  Is there enough liquidity with the internal exchange and bitstamp.  I don't leverage trade very often and for the most part I am a lender.  I don't like the sounds of lenders not wanting to use the site anymore, but I don't fully understand the logic, while it may limit potential profit it's still better than trading 1:1 direct on the exchanges.

I would much rather trade 1:1 than 2:1.  

Trade fees and loan interest take up a significant portion of your gains at 2:1 but your losses are still accelerated.

I feel like the advantages of leverage are lost at 2:1 but the risks are still there.

That doesn't make sense, the fees and borrowing interest are the same on the leverage portion of the money be it 2:1 or 10:1 as they are percentage based.

I can see the risk reward argument I suppose though, because in a flash crash if you lose its not all your money, but if it gains it is all yours.

Good point about the fees and interest.  Turns out Im not the brightest.

I still dont like it.

I had a 4:1 long last night.  If I knew about this I wouldnt have realized the profits.  Now I could only repoen half that position.  Shitty.
hero member
Activity: 763
Merit: 500
I mean, I do agree with the better safe than sorry approach, especially with everything that has gone on in the last month.  Maybe they could play with the leverage and margin requirements and come up with a better happy medium.

The obvious risk is a flash crash forcing margins.  Is there enough liquidity with the internal exchange and bitstamp.  I don't leverage trade very often and for the most part I am a lender.  I don't like the sounds of lenders not wanting to use the site anymore, but I don't fully understand the logic, while it may limit potential profit it's still better than trading 1:1 direct on the exchanges.

I would much rather trade 1:1 than 2:1. 

Trade fees and loan interest take up a significant portion of your gains at 2:1 but your losses are still accelerated.

I feel like the advantages of leverage are lost at 2:1 but the risks are still there.

That doesn't make sense, the fees and borrowing interest are the same on the leverage portion of the money be it 2:1 or 10:1 as they are percentage based.

I can see the risk reward argument I suppose though, because in a flash crash if you lose its not all your money, but if it gains it is all yours.
Jump to: