Bitcoin Forum
December 28, 2024, 01:13:45 PM *
News: Latest Bitcoin Core release: 28.0 [Torrent]
 
   Home   Help Search Login Register More  
Pages: « 1 ... 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 [244] 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 ... 361 »
  Print  
Author Topic: [OFFICIAL]Bitfinex.com first Bitcoin P2P lending platform for leverage trading  (Read 723971 times)
This is a self-moderated topic. If you do not want to be moderated by the person who started this topic, create a new topic.
gog1
Hero Member
*****
Offline Offline

Activity: 756
Merit: 500


View Profile
November 15, 2014, 07:41:58 PM
 #4861

The biggest problem with FRR is the distortion and the huge wall it puts up.  I think the FRR right now is just weight average rate of all fixed rate swap.  The easiest way to solve it with minimal disruption to existing system is simply to put a markup of min (2.5%, 0.1 * FRR); the 2.5% is the rate / year which is a small markup and let's call this effective FRR.  This way, people can put a fixed rate lend offer slight below the effective FRR - ie. the fake wall, and if taken, it'll still push up the FRR.

Lastly, make the lending fee for FRR 20% and make it 12.5% for fixed rate lender to discourage FRR offers.
DoubleSwapper
Full Member
***
Offline Offline

Activity: 172
Merit: 100


View Profile
November 15, 2014, 08:04:15 PM
 #4862

I've written tons of posts regarding the FRR. I feel though they are largely ignored by the people who have to decide on this topic. Instead most here stick to their positions by ignoring opposing arguments and repeating their own.
So I will just comment on your post as detailed as I can for the last time and then go quiet and just take the necessary actions if the problems remain:
So, there is a "wall" of FRR offers, 1.5 million. There are people who are willing to trade a higher possible return, for a return right now, which is why there is half a million before you even reach the FRR. If you took away the FRR, it would not change the fact that rates will always be as low as the one person who wants the least return.
This is highly debatable and very possibly false. Correct is that the rate could be as low or even lower in times of high supply but that doesn't mean that the FRR doesn't artificially keeps the rate low. Realistically the orderbook starts at 1% (cap for autoborrow). From there on it's a race down to the bottom. If what you say is correct we should see a roughly equal distribution from 1% to the lowest offer. this is evidently not the case though.

What we see most of the time is a deeply stacked orderbook often to the lowest pip (0.0001%) then a massive FRR wall and then some offers with quite a bit of space in between and then empty...a large gap until 1% because nobody really expects his offer to get taken any time soon if it is behind a +2m wall that is constantly replenished.

This alone shows that the FRR is not indifferent to the price discovery but in fact has a huge influence on the effectively reached rate.

Quote
There is a variable but finite amount of demand for swaps, call it x, and there is the response (supply) call it y. If y is greater than x, someone is not getting their funds used, and getting 0 as a return. So, it will always be a race to the lowest rate, with or without an FRR. My personal bot, simply looks to see what is the lowest rate and jumps in front of it.
So then why doesn't his race start at 1% but at the FRR?
Do you remember when a month ago Bitstamp had the 30k bitcoin limit sell order on their book? Was this order indifferent to the market? Of course not this wall had a huge influence on the market and so does the FRR wall.

Quote
The vast majority of the traders, IMO, don't check or even watch rates. So with whatever the current demand to open positions in, the person who is willing to settle for the least will be the first one to get a return. In other words, I feel like the group of people who wish rates were higher want to act as like a cartel, basically saying "if we all only offer at high rates, we will all get high rates", but in practice, it is impossible to prevent people undercutting.
When the offerbook ran out of autoborrowable funds yesterday, it absolutely had an influence on the market. If people want to undercut, let them but why give them a tool that through your method of calculation lets tons of dumb people create an artificial wall that is not even that effective at lending out their money. Often enough money sits for several hours or days at the FRR before finally taken. If these people didn't have the FRR they would have their money sitting at higher rates creating depth for the offerbook or not having it on the book at all because they forgot to even put in a new offer. But now these "dumb money" guys have a way to ride down with a huge wave.
Quote
Luckily, our volume is growing rapidly, and I think that as it continues to grow (we have been about half of the volume in USD/BTC lately), this will necessitate more swaps, and it stands to reason, rates will rise.
I remember that not too long ago you basically said the opposite and expected rates to get much, much lower in the longer term (which is a natural assumption as the rate of BFX swaps is much higher compared to bond markets etc. creating a natural arbitrage opportunity adjusted by the default risk of BFX.

Quote
I guess, the way I see it, is that people who currently use the FRR are basically saying "I will take whatever rate, I don't want to manage it, and I don't want to deal with it", and if you got rid of the FRR, they would just put offers a little lower than whatever the current lowest is, and in fact, could drive rates even lower (assuming that they want to invest the roughly 60-120 minutes to write the bot).

Maybe these people would be too dumb to even have their money lent out continuously? Why give them a super cheap way to constantly drive down the rates instead of having limit orders creating depth to the book? Markets this size (relatively small) are not nearly as efficient as you might think. Somebody who doesn't even care about at which rate his money will be taken will absolutely not permanently readjust the rate.
Also the FRR is not even that good at getting your money lent out:

As BFXDATA shows 733,000$ of swaps came from FRR offers while over 5 million came from fixed offers and manually filled demands. The current size of the FRR is about 2 million and was like this in that time frame. This means your money won't necessarily be even lent out that quickly at FRR and it also means that the vast majority of "deals" were done below the FRR. Why does this artifical wall has to depress the whole market?
Quote
Like, I have seen other people complain that as rates go down, people close their swaps and get a new one for lower. That is going to happen regardless, there is no way to avoid competition, and if anything, I think that people who use the FRR are the ones who care the least about the rate rising (I could be wrong). One other thing, that I just was thinking of, was this...the way the FRR is calculated currently is public knowledge, so what if, someone wrote a bot to always offer at the currently calculated FRR? If enough people used that bot, the situation would be roughly identical to how it is now, and there would be nothing we could do to "remove the wall", because people can put their rates in as whatever they like.
Haha, that's a clever thought. But why would you assume that users would use a third party provider as frequently as an in-built feature of your site? And why would you assume that a third party provider would use the same faulty algorithm? Also having a third party provider would lead to competition, people making other, better bots, advertised with "adjusts more quickly to a rising market" etc.

But the thought is still clever because you should ask yourself, why would you provide a such a faulty tool yourself?
Quote
I guess what I am trying to say is this, I know some people want higher rates, but the rates will always be just high enough to get you to offer the swap and no higher. That is not our policy, it is simply the way markets work. Markets try to find the most efficient price, and that is the price where the offerers are offering at, by definition, the lowest price. So, I personally think that a lot of the rage against the FRR is really a rage against competition in a market, and although we are obviously working to make a better tool, I think that no matter what we do, basic market competition is something no one can escape.
For me it's not necessarily about higher rates (I commented muliple times that I consider 1% as absolutely ridiculous unless there is a bubble or something happening) it's about more consistent rates and more efficient price discovery, because as it stands right now the FRR disrupts this discovery.

It's nice you are there to talk about such features. Other companies would not even consider talking to the community about that. But if you really want to defend the FRR (in it's current form) please answer this:

In the last two days (during this short lived upswing) the entire FRR wall was eaten (+3m) at a rate of 0.07%, after this the complete offerbook was cleared out up to 1% (some people's offers were rejected because of no reserves). Interestingly though the increase of total swaps was just about 3.4m at max. And in the time when the offers were at 0.7 to 0.9 % even then FRR came in and offered money for scraps (0.09%)
Do you consider this to be efficient price discovery? Why does the rate adjust so slowly upwards and misprices in obvious situations (offer for 0.07 when the lowester offer is 0.7) but gets drawn down so quickly?

In a thick orderbook that incentivizes people to search for the best rate they can get and not fight some artificial whale wall of dumb money the orderbook would have never been cleared out and rates would not haven risen above 0.5% but they would also have been at 0.2% to 0.3 for longer before and after the rally instead of being surpressed like this. It was like this at the beginning of the year when the FRR would rarely accumulate more than a few hundred thousand dollars and then the dumb money came in.
gog1
Hero Member
*****
Offline Offline

Activity: 756
Merit: 500


View Profile
November 15, 2014, 08:32:12 PM
 #4863

very well said by DoubleSwapper!

I really hope they seriously consider my simple adjustment
DoubleSwapper
Full Member
***
Offline Offline

Activity: 172
Merit: 100


View Profile
November 15, 2014, 09:06:36 PM
 #4864

Woah, what happened there?

BTC Swaps just halved from under 10k to 5. Was there any corresponding price action?
noggin-scratcher
Full Member
***
Offline Offline

Activity: 136
Merit: 100


View Profile
November 15, 2014, 09:09:03 PM
 #4865

I have another FRR suggestion to throw onto the heap: instead of setting the rate algorithmically, hire a human being to do it (possibly informed by algorithmic means, but capable of over-riding the algorithmic suggestion whenever it makes no goddamn sense), and charge FRR swap providers a management fee for the privilege.

Bitfinex referral code: uOaxAuXdVX
nrd525
Legendary
*
Offline Offline

Activity: 1870
Merit: 1023


View Profile
November 15, 2014, 09:56:29 PM
 #4866

When is BFX introducing 20x leverage like OkCoin?

 Grin

Digital Gold for Gamblers and True Believers
PRab
Member
**
Offline Offline

Activity: 98
Merit: 10


View Profile
November 15, 2014, 10:38:43 PM
 #4867

I don't read this thread too often, but I just saw a lot of posts complaining about the FRR wall. I don't mind if the exact details of FRR get modified, but I only have a small amount of money getting lent, so it is not worth my time/effort to log into Bitfinex even once a day. I autolend at FRR because it let me put my account on "autopilot" and get a few cents a day as interest. If there wasn't a FRR auto lend option, I would probably just withdraw my money or set it to lend at a lower than market rate so that I wouldn't have to worry about it sitting there not earning any interest.
2586
Member
**
Offline Offline

Activity: 77
Merit: 13


View Profile
November 16, 2014, 12:02:09 AM
Last edit: November 16, 2014, 02:41:08 AM by 2586
 #4868

Here's an idea for replacing the FRR while still catering to lazy lenders.

When choosing variable rate autolending, lenders will have two parameters to set:

  • Allow no more than $x of competing offers to be priced lower than or equal to mine
  • Do not reduce my offer below y%

Do not provide default values.

The system should place the lenders offer as high as it can without violating the specified parameters. If two competing offers would end up in a race to the bottom (for example, x=0, y=0), just match them with the highest available swap demands.

This would break up the massive FRR wall while still allowing autolenders to benefit from increases in swap demand.

An alternative that might be simpler to implement:

  • Place new swap offers at x%
  • Reduce swap offer rate by y% for each hour(or minute?) that it remains unfilled
  • Do not reduce swap offer rate below z%

This one also has the advantage that default values can be provided, since it will still create a range of offers even if everyone uses the same parameters.
DoubleSwapper
Full Member
***
Offline Offline

Activity: 172
Merit: 100


View Profile
November 16, 2014, 12:43:13 AM
 #4869

When is BFX introducing 20x leverage like OkCoin?

 Grin
Hpefully never. Futures trading has significant counter party risk. Look at 796 and how often they have to socialize losses. 20x on margin....wow I don't know about that.

I don't read this thread too often, but I just saw a lot of posts complaining about the FRR wall. I don't mind if the exact details of FRR get modified, but I only have a small amount of money getting lent, so it is not worth my time/effort to log into Bitfinex even once a day. I autolend at FRR because it let me put my account on "autopilot" and get a few cents a day as interest. If there wasn't a FRR auto lend option, I would probably just withdraw my money or set it to lend at a lower than market rate so that I wouldn't have to worry about it sitting there not earning any interest.
Sorry if this comes of as blunt but if you can't even be bothered to check your money once a day (technically you don't even have to, you can just offer 30 days swaps and enable "notice me when this swap gets returned") you should really not profit especially not at the expense of big lenders for which the FRR makes a difference of getting $300 or $1000 at the end of the day.
DoubleSwapper
Full Member
***
Offline Offline

Activity: 172
Merit: 100


View Profile
November 16, 2014, 08:39:31 AM
 #4870

When is BFX introducing 20x leverage like OkCoin?

 Grin
Hpefully never. Futures trading has significant counter party risk. Look at 796 and how often they have to socialize losses. 20x on margin....wow I don't know about that.

I don't read this thread too often, but I just saw a lot of posts complaining about the FRR wall. I don't mind if the exact details of FRR get modified, but I only have a small amount of money getting lent, so it is not worth my time/effort to log into Bitfinex even once a day. I autolend at FRR because it let me put my account on "autopilot" and get a few cents a day as interest. If there wasn't a FRR auto lend option, I would probably just withdraw my money or set it to lend at a lower than market rate so that I wouldn't have to worry about it sitting there not earning any interest.
Sorry if this comes of as blunt but if you can't even be bothered to check your money once a day (technically you don't even have to, you can just offer 30 days swaps and enable "notice me when this swap gets returned") you should really not profit especially not at the expense of big lenders for which the FRR makes a difference of getting $300 or $1000 at the end of the day.
Just to add to my former post:

http://www.reddit.com/r/Bitcoin/comments/2mdtxl/huobi_takes_1271000_usd_from_users_profits_during/

QED. Don't add futures trading, Bitfinex, seriously, just don't.

EDIT: Also read the top post of Huobi themselves. Socializing losses seems to be a common tactic for futures trading platforms in the bitcoin environment. As said 796 does it all the time.
noobtrader
Legendary
*
Offline Offline

Activity: 1456
Merit: 1000



View Profile
November 16, 2014, 01:50:59 PM
 #4871

if i offer swap of btc for 2 days and some ppl take it only for 3 hours, do i receive interest of 1 day or  just 3 hours Huh


thanks

"...I suspect we need a better incentive for users to run nodes instead of relying solely on altruism...",  satoshi@vistomail.com
medicine
Hero Member
*****
Offline Offline

Activity: 697
Merit: 501



View Profile WWW
November 16, 2014, 02:08:46 PM
 #4872

if i offer swap of btc for 2 days and some ppl take it only for 3 hours, do i receive interest of 1 day or  just 3 hours Huh


thanks
Pretty sure interested is compounded hourly so if someone returns a swap after 3 hours you only receive 3 hours of interest, not 1 whole day. 
Mythoughts
Member
**
Offline Offline

Activity: 63
Merit: 14


View Profile
November 16, 2014, 03:13:01 PM
 #4873

Wouldn't the swap/FRR issues solve themselves if it was possibly to take (cheap) swap offers and lend them out at higher rates?

Nearly everyone wins here:
- the cheap lenders can set everything on auto-pilot
- those who care to micro-manage their swaps can use market inefficiencies to earn some more
- more commission for BFX
russianptr
Full Member
***
Offline Offline

Activity: 154
Merit: 100


View Profile
November 16, 2014, 04:02:55 PM
 #4874

Wouldn't the swap/FRR issues solve themselves if it was possibly to take (cheap) swap offers and lend them out at higher rates?

Nearly everyone wins here:
- the cheap lenders can set everything on auto-pilot
- those who care to micro-manage their swaps can use market inefficiencies to earn some more
- more commission for BFX

would be too easy to manipulate the market:
buy up everyones cheap swap offers and offer all the funds at 1 % a day... many traders don´t care about rates or would be forced
be forced to pay your usurious interest rates  Wink

Mythoughts
Member
**
Offline Offline

Activity: 63
Merit: 14


View Profile
November 16, 2014, 04:13:47 PM
Last edit: November 16, 2014, 06:46:05 PM by Mythoughts
 #4875

Only if there's
- a steady stream of really cheap swap offers (they'll adjust their cheap offers upwards if people use them instantly to re-offer them) and
- the guys taking the cheap offers can lend most of them out. If they take offers cheap but cant find anyone who want them, they risk losing money

It would adjust in a way more natural way than it does now.

Could try this slowly by requiring a big margin for re-lending.
gog1
Hero Member
*****
Offline Offline

Activity: 756
Merit: 500


View Profile
November 16, 2014, 05:11:33 PM
 #4876

Here's an idea for replacing the FRR while still catering to lazy lenders.

When choosing variable rate autolending, lenders will have two parameters to set:

  • Allow no more than $x of competing offers to be priced lower than or equal to mine
  • Do not reduce my offer below y%

Do not provide default values.

The system should place the lenders offer as high as it can without violating the specified parameters. If two competing offers would end up in a race to the bottom (for example, x=0, y=0), just match them with the highest available swap demands.

This would break up the massive FRR wall while still allowing autolenders to benefit from increases in swap demand.

An alternative that might be simpler to implement:

  • Place new swap offers at x%
  • Reduce swap offer rate by y% for each hour(or minute?) that it remains unfilled
  • Do not reduce swap offer rate below z%

This one also has the advantage that default values can be provided, since it will still create a range of offers even if everyone uses the same parameters.

2586, much appreciate people giving thoughts to the issue - but I see this as a "big" change.  The (x=0, y=0) => lend at highest available swap demands will kill all the swap demand; I wouldn't be shocked if a lot of people leave it or make it x=0 , y =0.  I believe my idea of creating an 'effective FRR' by charging a markup over FRR (do not change the way FRR is calculated) involves less disruption to the way people react to it right now, it somewhat solves the wall issue as lending offers that are taken slightly below FRR will still result in FRR increasing.
oyvinds
Newbie
*
Offline Offline

Activity: 48
Merit: 0



View Profile WWW
November 16, 2014, 06:58:58 PM
 #4877

Woah, what happened there?

BTC Swaps just halved from under 10k to 5. Was there any corresponding price action?

Someone bought back a 5k short and then reserved 5k BTC. If you plan to add a big short then you reserve first and wait (days if need be) for the right price. The obvious reason is that people look at bfxdata and notice big shorts. If you reserve 200 BTC or 5000 BTC at one point in time and do the trade at another then nobody knows when you actually did the trade.
gog1
Hero Member
*****
Offline Offline

Activity: 756
Merit: 500


View Profile
November 16, 2014, 07:20:15 PM
 #4878

People are talking about futures trading.  I really hate the existing futures at 796 / OKCoin / etc., not because of the counter-party risk, but that it is settled in BTC - if it's like real commodities futures, where it's settled by 'physical delivery', or paying the difference in fiat - I'm all for it.  Imagine you went short BTC, BTC fell which is great, but then you get paid by having more BTC in your account - o crap, now you have more of the thing that you wanted to short in the first place!
2586
Member
**
Offline Offline

Activity: 77
Merit: 13


View Profile
November 16, 2014, 08:33:06 PM
 #4879

2586, much appreciate people giving thoughts to the issue - but I see this as a "big" change.  The (x=0, y=0) => lend at highest available swap demands will kill all the swap demand; I wouldn't be shocked if a lot of people leave it or make it x=0 , y =0.

That would be the same as offering your swaps at a fixed rate of 0%. It seems like very few people if any would choose to do that. When was the last time that the USD swap book had no demands on it?

After thinking about it some more, I like my second idea better, though:

  • Place new swap offers at x%
  • Reduce swap offer rate by y% for each hour(or minute?) that it remains unfilled
  • Do not reduce swap offer rate below z%

This one also has the advantage that default values can be provided, since it will still create a range of offers even if everyone uses the same parameters.

Offers slowly cascade down until they reach the current supply/demand equilibrium point. As demand for swaps rises, it pushes that equilibrium rate up. As swap supply rises, it gradually pushes the equilibrium rate back down. You could set everyone at a default of x=1%, y=0.1% (per hour, or 0.00167% per minute), z=0.05%, and let people fiddle around with it from there. With those settings, your offers usually won't sit idle for more than 10 hours, which is probably better than the FRR on average. When demand goes crazy and the offer book gets cleared, your new offers start getting filled near %1, which they'd never do if set to the FRR.

Quote
I believe my idea of creating an 'effective FRR' by charging a markup over FRR (do not change the way FRR is calculated) involves less disruption to the way people react to it right now, it somewhat solves the wall issue as lending offers that are taken slightly below FRR will still result in FRR increasing.

This would cause FRR offers to sit idle for even longer than they already do, and doesn't remove the wall. It's less bad than the current FRR setup, but still has the same primary weakness: spikes in demand aren't apparent until the offer book has been entirely cleared out. The wall is allowed to float up slowly, but the price (interest rate) signalling mechanism still gets suppressed. The main benefit of your proposal would be that it'd make the FRR less attractive to lenders. May as well just remove the FRR and its distorting effect entirely.
russianptr
Full Member
***
Offline Offline

Activity: 154
Merit: 100


View Profile
November 16, 2014, 08:56:14 PM
 #4880

People are talking about futures trading.  I really hate the existing futures at 796 / OKCoin / etc., not because of the counter-party risk, but that it is settled in BTC - if it's like real commodities futures, where it's settled by 'physical delivery', or paying the difference in fiat - I'm all for it.  Imagine you went short BTC, BTC fell which is great, but then you get paid by having more BTC in your account - o crap, now you have more of the thing that you wanted to short in the first place!

that´s exactly my concern with futures trading on the other BTC sites. Good point!

@Mythoughts
after reading your reply I really like your proposal.

@Bitfinex
make it happen!  Grin

Pages: « 1 ... 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 [244] 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 ... 361 »
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.19 | SMF © 2006-2009, Simple Machines Valid XHTML 1.0! Valid CSS!