This is a mirror of the original BitMEX trollbox archive that used to be online here.
BitMEX disabled their archive after the DDOS attack so I have decided to make my mirror publicly available.

Well now, a few days after I made this mirror available to the public, the original archive from BitMEX is online again.
But since it is still limited (neutered to the last few hundred messages) I will keep my FULL mirror alive.

You have access to more than 50 million trollbox posts. This is more than 7 gigabyte of data hosted on a $5 server so please be patient or consider a donation

<<  >>  English  中文  Русский  한국어  日本語  Español  Français 
[2016-02-13 12:17:35] habibi : nah they are offline, probally partying in striptease club today
[2016-02-13 12:17:41] Tulip_stefan : I don't really case what index they use, my algo can handle it :P
[2016-02-13 12:17:52] rapidtrades : noone axed u
[2016-02-13 12:18:10] habibi : Tulip_stefan: as long as its not settlement time
[2016-02-13 12:19:28] rapidtrades : cmon bitcoin do smth...i have errands to do
[2016-02-13 12:21:37] Tulip_stefan : i'm really wondering, are you watching every settlement? I usually check what my bots do every 3 months or so... This really seems like a waste of time.
[2016-02-13 12:22:44] habibi : i am not as tallented as u to make that winning only algo so i am making money on scalping mostly
[2016-02-13 12:22:52] habibi : so minutes and hours are important for me
[2016-02-13 12:24:12] Tulip_stefan : I'm not that talented. It took me 3 months of research to create a decent algorithm. Then the next 5 months of research resulted in nothing.
[2016-02-13 12:24:15] rapidtrades : yeah I try to watch every settlement cos ppl used to do dumb shit
[2016-02-13 12:24:32] Tulip_stefan : and with 'months of research', i really mean a month where you're working on it like 20 hours every weekend.
[2016-02-13 12:24:55] rapidtrades : so 20 days work tops
[2016-02-13 12:25:46] Tulip_stefan : sounds about right, if you exclude the time implementing API's
[2016-02-13 12:25:52] rapidtrades : also if I have positions on the daily I need to leg in/out you have to be here for settlement
[2016-02-13 12:33:14] Tulip_stefan : why is XBT so overvalued? The index is 389.9 and the XBT24 is 392.3?
[2016-02-13 12:33:50] rapidtrades : ppl are buying
[2016-02-13 12:38:49] habibi : thats so called premium
[2016-02-13 12:45:49] jung1 : where we go with eth?
[2016-02-13 12:45:56] habibi : to 100 usd
[2016-02-13 12:45:57] habibi : :D
[2016-02-13 12:46:09] tradedesk : wishfull thinking
[2016-02-13 12:46:11] mrp1nk : 100 satoshi
[2016-02-13 12:46:31] mrp1nk : :p
[2016-02-13 12:46:47] habibi : i am joking lel. its jung1 thoughs
[2016-02-13 12:46:54] habibi : just making fun of this :)
[2016-02-13 12:47:05] jung1 : lets meet here in 2 years
[2016-02-13 12:47:13] habibi : u said 1 year last time
[2016-02-13 12:47:15] habibi : :(
[2016-02-13 12:47:25] jung1 : lets meet here in 1 year
[2016-02-13 12:47:39] habibi : ok, 100 usd on single eth by then right?
[2016-02-13 12:48:11] jung1 : could be
[2016-02-13 12:48:21] habibi : could be 1 usd as well...
[2016-02-13 12:48:25] jung1 : could be
[2016-02-13 12:48:31] habibi : lel. go away man :D
[2016-02-13 12:48:40] habibi : too much weed
[2016-02-13 12:49:05] jung1 : but during next bitcoin pump (around june/july eth could go as ltc in '13
[2016-02-13 12:49:35] jung1 : dude. this is how markets works. everyone have different opinions
[2016-02-13 12:49:44] jung1 : will see who was right
[2016-02-13 12:50:01] habibi : its not market, its just dreaming
[2016-02-13 12:50:19] jung1 : people said same about bitcoin in '12
[2016-02-13 12:50:33] habibi : and then came willy
[2016-02-13 12:50:36] habibi : :D
[2016-02-13 12:50:57] jung1 : so its not impossible
[2016-02-13 12:51:00] jung1 : as you can see
[2016-02-13 12:51:17] tradedesk : can't see chart is frozen
[2016-02-13 12:51:29] habibi : u need to remember about scale anyway. its much much easier to move market cap from 10 milions to 1 bilion then from 500 milions to i dont know 50 bilions?
[2016-02-13 12:52:10] jung1 : yes you are right
[2016-02-13 12:52:35] jung1 : but if vitalic dont screw up and eth will be used in IoT and all that microsoft stuff
[2016-02-13 12:52:41] habibi : but simple question, why would big players like banks/microsoft or any should buy eth if they just can fork it on their own terms of use?
[2016-02-13 12:52:58] habibi : for their own terms of use*
[2016-02-13 12:53:38] jung1 : why reinwent wheel once again?
[2016-02-13 12:53:50] tradedesk : TCO, usability, user adoption, value
[2016-02-13 12:54:26] jung1 : sceurity over strong, distributed blockchain
[2016-02-13 12:54:28] tscha : dump <3
[2016-02-13 12:54:30] habibi : TCO value is actualyl against buying it
[2016-02-13 12:55:01] tradedesk : not at this price, i agree
[2016-02-13 12:55:03] habibi : usability will be the same
[2016-02-13 12:55:28] tradedesk : compatibility*
[2016-02-13 12:55:47] tradedesk : as in, no vendor lock-in
[2016-02-13 12:56:14] habibi : eth is not made for being currency itself u know Vitalik workds right?
[2016-02-13 12:59:34] habibi : jung1: security when it will become POS? questionable
[2016-02-13 12:59:45] habibi : same for new fork as for old
[2016-02-13 13:12:34] jung1 : Blessed are the people who have not seen and yet have believed."
[2016-02-13 13:16:47] habibi : blessed who made money while others were waiting for miracle
[2016-02-13 13:20:05] jung1 : queen is blessed
[2016-02-13 13:45:27] REKT : Liquidated long on `ETH7D`: Sell 301 @ 0.01224
[2016-02-13 13:45:27] REKT : Liquidated long on `ETH7D`: Sell 600 @ 0.01225
[2016-02-13 13:53:15] habibi : if we fall there will some nice dpe on eth
[2016-02-13 13:53:23] habibi : so i would suggest force liq a bit
[2016-02-13 13:54:27] REKT : Liquidated long on `ETH7D`: Sell 100 @ 0.01211
[2016-02-13 13:54:35] j8 : yeah, basis needs to be adjusted
[2016-02-13 13:55:31] habibi : yep
[2016-02-13 13:57:15] djuan : hey guys ?
[2016-02-13 13:57:28] jung1 : look at monero
[2016-02-13 13:57:34] jung1 : already hit 10mln market cap
[2016-02-13 13:57:35] djuan : has bitcoin started pumping yet?
[2016-02-13 13:58:49] rapidtrades : u know how there's a neutron bomb that kills all living things but leaves everything else intact?
[2016-02-13 13:59:37] djuan : nope. sounds like science fiction
[2016-02-13 14:01:00] rapidtrades : https://en.wikipedia.org/wiki/Neutron_bomb
[2016-02-13 14:02:07] rapidtrades : anyway I want that but for plebs
[2016-02-13 14:02:13] rapidtrades : The Pleb Bomb
[2016-02-13 14:02:45] habibi : then who would u trade against?
[2016-02-13 14:03:19] rapidtrades : good point hajibi
[2016-02-13 14:03:44] habibi : Bitmex stuff u there?
[2016-02-13 14:03:57] jung1 : yes
[2016-02-13 14:07:29] Mangalica : someone just took out a 220 BTC bid wall of ETH on polo
[2016-02-13 14:07:45] jung1 : saw it
[2016-02-13 14:08:04] habibi : coz 220 btc is not a real bid for eth
[2016-02-13 14:08:10] habibi : too much eth both for 2k per btc
[2016-02-13 14:08:17] jung1 : im short since 125. lets see
[2016-02-13 14:08:22] Mangalica : yeah
[2016-02-13 14:08:25] jung1 : polo is now biggest altcoin exchange
[2016-02-13 14:08:26] habibi : bought*
[2016-02-13 14:08:31] jung1 : too bad is so fucking slow
[2016-02-13 14:08:45] habibi : lets see some dpe soon here
[2016-02-13 14:09:01] jung1 : xmr good for short also
[2016-02-13 14:10:57] jung1 : eth 0.01 today?
[2016-02-13 14:11:57] habibi : i dont know, but 100 dollars in year for sure
[2016-02-13 14:13:48] Mangalica : a margin avalanche is in the books, judging from polo BTC lend rates being this high for so long
[2016-02-13 14:14:09] Mangalica : would do much good for my short, but then i've been lending BTC for a while and i also want to get those back
[2016-02-13 14:14:20] Mangalica : :/
[2016-02-13 14:16:05] Mangalica : so i guess i'll just hope for this slow bleed to go on
[2016-02-13 14:16:15] habibi : i wouldnt say its slow
[2016-02-13 14:16:21] habibi : pretty impresive imho
[2016-02-13 14:16:49] Mangalica : slower than 20% in a 5 min candle, which isn't impossible
[2016-02-13 14:16:50] Mangalica : :)
[2016-02-13 14:17:37] jung1 : Mangalica: where you rebuy?
[2016-02-13 14:17:38] Mangalica : alright, given the bids around 0.01 it's not likely, but still. Just wanna see people liquidated in an orderly manner.
[2016-02-13 14:18:02] Mangalica : a Stalinesque sell-off
[2016-02-13 14:18:10] Mangalica : no idea jung1
[2016-02-13 14:18:28] REKT : Liquidated long on `ETH7D`: Sell 5 @ 0.01156
[2016-02-13 14:18:35] jung1 : in last pump I sold fo 0.01 and place a order at 0.085, the it hit 0.00850001
[2016-02-13 14:18:37] jung1 : hehe
[2016-02-13 14:19:29] jung1 : altoicns: loosing money was naver so fun
[2016-02-13 14:19:36] Mangalica : ETH accelerating..
[2016-02-13 14:19:38] REKT : Liquidated long on `ETH7D`: Sell 800 @ 0.01120
[2016-02-13 14:19:38] REKT : Liquidated long on `ETH7D`: Sell 50 @ 0.01144
[2016-02-13 14:19:45] habibi : huge dpe if we wont bounce
[2016-02-13 14:19:47] Tulip_stefan : wow...
[2016-02-13 14:20:53] rapidtrades : no dpe?
[2016-02-13 14:21:05] habibi : strange
[2016-02-13 14:21:28] jung1 : 0.01 is really possible
[2016-02-13 14:21:36] habibi : yea yea
[2016-02-13 14:21:57] Mangalica : 0.011 getting nuked
[2016-02-13 14:27:48] jung1 : I will wait with my bids. lets see what will happen around 0.01
[2016-02-13 14:28:05] jung1 : maybe we will buy some at 0.007
[2016-02-13 14:28:15] habibi : or just wait till 100$
[2016-02-13 14:29:47] jung1 : dont worry
[2016-02-13 14:30:40] Tulip_stefan : i was hoping it would crash...
[2016-02-13 14:31:44] Mangalica : closed half my shorts down there but the other half is not gonna end up green at this rate
[2016-02-13 14:32:06] rapidtrades : dam that reversal
[2016-02-13 14:37:39] jung1 : maybe its not the end of dump yet
[2016-02-13 14:40:22] habibi : yep more dump and then boom-> 100$ <3
[2016-02-13 14:41:04] Mangalica : ok closed the rest of my short, we can crash 30% now
[2016-02-13 14:42:01] Tulip_stefan : that's a hell of a volatile stock...
[2016-02-13 14:42:21] habibi : u said i familiar with altcoin world
[2016-02-13 14:42:24] habibi : u*
[2016-02-13 14:42:25] habibi : :p
[2016-02-13 14:43:39] Tulip_stefan : i actually haven't looked at charts in a really long while.
[2016-02-13 14:44:14] Tulip_stefan : i don't believe in TA. Doesn't make much sense to look at charts then.
[2016-02-13 14:44:44] habibi : me either but support/ressistance lvls/break outs are kinda working for me
[2016-02-13 14:45:26] habibi : and fractals sometimes
[2016-02-13 14:45:27] Tulip_stefan : That's not TA?
[2016-02-13 14:45:37] habibi : not? ah sorry then :D
[2016-02-13 14:46:14] Tulip_stefan : i tried a lot of TA with machine learning and the results where... non existent.
[2016-02-13 14:47:52] rapidtrades : Tulip_stefan: of course it's ta...
[2016-02-13 14:48:30] Tulip_stefan : i know, the tone of my statement was lost somewhere along the internet.
[2016-02-13 14:48:49] rapidtrades : ok then
[2016-02-13 14:48:51] habibi : *confused
[2016-02-13 14:49:08] rapidtrades : he's saying that's TA too and u shuldn't make money
[2016-02-13 14:49:20] rapidtrades : better quit now habiib
[2016-02-13 14:49:38] habibi : god damm it, two options. i am lier he lies
[2016-02-13 14:51:05] Tulip_stefan : i'll go for option 3, the random gods are in your favor.
[2016-02-13 14:51:41] rapidtrades : well that's it habibi...we better pack up our bags and go find another job
[2016-02-13 14:51:57] rapidtrades : TA doesn't work
[2016-02-13 14:52:24] habibi : last time i tried to find a job i spent 20k$ on gpu mining rigs
[2016-02-13 14:53:05] Tulip_stefan : i didn't spent 20k but i also brought a mining rig :P
[2016-02-13 14:53:24] Tulip_stefan : i sold the GPU and re-purposed it as a server about 3 weeks later..
[2016-02-13 14:53:35] Tulip_stefan : still using it to run my trading bot.
[2016-02-13 14:53:42] habibi : i used to be really proud of my rig, it was so beautiful
[2016-02-13 14:54:59] habibi : ah here it is, take a look http://imgur.com/EIF769N
[2016-02-13 14:56:31] Mangalica : new local lows on ETH meanwhile
[2016-02-13 14:57:06] jung1 : its yours huobi ?
[2016-02-13 14:57:20] habibi : used to be, sold it time ago
[2016-02-13 14:58:44] rapidtrades : for how mach
[2016-02-13 14:58:57] habibi : 10k usd
[2016-02-13 14:59:12] rapidtrades : dam 10k loss...did u mine smth
[2016-02-13 14:59:21] Tulip_stefan : Not bad.
[2016-02-13 14:59:55] habibi : yep, profits were made nicely
[2016-02-13 15:00:14] rapidtrades : nice nice....
[2016-02-13 15:00:35] rapidtrades : I got a rig with ATI card to have the option to mine...then BTC crashed from 30>2
[2016-02-13 15:01:24] rapidtrades : I gave up that idea quickly...main purpose wasn't mining tho so not hurt much :)
[2016-02-13 15:02:10] rapidtrades : it retrospect I could make $500 per month but who knew btc would spike after that crash
[2016-02-13 15:02:40] habibi : true, i am getting sad if i think that somehow i didnt knew bitcoin back before 2013
[2016-02-13 15:02:57] Mangalica : i remember looking at the Avalon ASICS in 2012
[2016-02-13 15:03:25] rapidtrades : I knew about bitcoin when it was basically free....but didn't think it was worth it to download the client and cpu mine :)
[2016-02-13 15:03:42] Tulip_stefan : i stepped in right after the $1000 bubble :(
[2016-02-13 15:03:51] Mangalica : had almost the money to buy but decided against
[2016-02-13 15:04:13] Mangalica : those made like 400-500 USD a DAY next spring
[2016-02-13 15:05:01] rapidtrades : But if you look at a lot of the early adapters, not many are millionaires now
[2016-02-13 15:05:18] Mangalica : panic mined some shitcoins with my GPU that spring/summer and that established my trading bankroll
[2016-02-13 15:05:28] rapidtrades : they either got out too early or lost it later on
[2016-02-13 15:05:33] jung1 : back in '11 and early '12 there was faucents that give 5btc avery 15min
[2016-02-13 15:06:21] Mangalica : i keep telling myself i only got to know about BTC in the second half of 2012
[2016-02-13 15:06:46] Tulip_stefan : just wait 3-4 years. Either we'll be a millionaires or btc goes to zero. 50% chance :P
[2016-02-13 15:06:56] Mangalica : but i have a vague memory reading something about dollar parity.. perhaps it's a suppressed memory to keep my sanity :)
[2016-02-13 15:07:58] jung1 : eth back in the game?
[2016-02-13 15:08:16] Mangalica : depends on how far this little bounce goes
[2016-02-13 15:08:30] Mangalica : we could bounce here making lower highs and then test 0.01 for all i know
[2016-02-13 15:08:57] Mangalica : a bearflag if you like
[2016-02-13 15:09:22] jung1 : lokks like small double floor on 5min
[2016-02-13 15:09:25] Mangalica : yeah
[2016-02-13 15:09:32] Mangalica : for now
[2016-02-13 15:10:00] Tulip_stefan : What exactly is the deal with the `margin` websocket call? It will only send the fields it thinks are changed?
[2016-02-13 15:10:15] Mangalica : opened a very speculative long when it bounced the second time but i'm not too confident
[2016-02-13 15:12:06] BitMEX_Sam : Tulip_stefan: All websocket subscriptions send deltas only
[2016-02-13 15:12:25] Tulip_stefan : Ahh i didn't realize that.
[2016-02-13 15:12:27] BitMEX_Sam : It's expected that you get the initial data via an HTTP call or via `pushAccount`.
[2016-02-13 15:12:39] BitMEX_Sam : In the (very) near future we'll be pushing the initial data when you first subscribe to make it easier.
[2016-02-13 15:13:14] habibi : BitMEX_Sam: hey Sam, how it was possible there was no dpe on eth even on the dip?
[2016-02-13 15:13:23] Tulip_stefan : yeah i read that. For the time being i'll just call `pushAccount` on first connect.
[2016-02-13 15:13:43] habibi : ah maybe that yesterday 36k liquidation on 24h cover a bit of 7eth now?
[2016-02-13 15:13:44] BitMEX_Sam : Tulip_stefan: Sounds good - we plan to launch the new API features by Monday so you'll be able to rip that out very shortly
[2016-02-13 15:13:48] BitMEX_Sam : It's already in Testnet
[2016-02-13 15:13:56] BitMEX_Sam : So if you develop with that as a target you'll get the new features.
[2016-02-13 15:14:13] BitMEX_Sam : habibi: I haven't been watching that market closely but I assume the retraces allowed the engine to close liqs out in the black
[2016-02-13 15:14:51] habibi : shieet, cant find "out in the black" meaning
[2016-02-13 15:15:47] jung1 : poza w czarnym
[2016-02-13 15:15:48] BitMEX_Sam : As in, in profit
[2016-02-13 15:15:50] Mangalica : no collateral damage / need for profit adjustment
[2016-02-13 15:16:16] habibi : i was watching market at the time of all three liqs happen, didnt show them closed into market but maybe i am wrong
[2016-02-13 15:16:19] BitMEX_Sam : If the liquidation engine manages to close out a position with some extra money left, that's used to offset any positions that are closed out with a deficit
[2016-02-13 15:16:45] habibi : so there are no liq to get closed right now there?
[2016-02-13 15:17:49] j8 : also the DPE estimate is relative to the mark price, which was about 5% above the market at the time
[2016-02-13 15:18:03] BitMEX_Sam : habibi: Nope, no liquidations: https://www.bitmex.com/api/v1/order/liquidations
[2016-02-13 15:18:32] BitMEX_Sam : But this is a good show of why we chose the mark pricing system, instead of last price
[2016-02-13 15:18:36] habibi : ah that api shows all open liqs?
[2016-02-13 15:18:42] BitMEX_Sam : 15x leverage on this market is quite a lot and yet it can be managed
[2016-02-13 15:18:49] BitMEX_Sam : habibi: Yep
[2016-02-13 15:19:01] habibi : good to know
[2016-02-13 15:19:14] habibi : though it only shows last time for specific ammount of time
[2016-02-13 15:19:48] BitMEX_Sam : It only shows them while they're open
[2016-02-13 15:19:54] BitMEX_Sam : We don't yet have an endpoint for historical data
[2016-02-13 15:20:41] jung1 : keep an eye on monero also
[2016-02-13 15:20:42] jung1 : http://i.imgur.com/kzMREt7.jpg
[2016-02-13 15:22:45] Mangalica : yeah if nothing else, this ETH pump has drawn in a lot of coins to polo
[2016-02-13 15:23:15] Mangalica : which leads one to think the alt volatility is not over
[2016-02-13 15:28:13] tscha : BitMEX_Sam: but you are just adding new api features right? no changes to existing api commands? so everything is backwards compatible?
[2016-02-13 15:28:22] BitMEX_Sam : tscha: Yes
[2016-02-13 15:28:46] tscha : cool :) then I can chill
[2016-02-13 15:35:45] jung1 : looks like end of dump
[2016-02-13 15:35:58] Mangalica : yeah
[2016-02-13 15:36:18] habibi : yep looks like i am the only one playing that waves here
[2016-02-13 15:36:19] Mangalica : we need some follow through
[2016-02-13 15:40:13] jung1 : or not
[2016-02-13 16:10:48] Tulip_stefan : the conclusion of the win-an-ipad contest at cex.
[2016-02-13 16:10:58] Tulip_stefan : and the winner is.. someone who paid 1.5 ipads in fees last month.
[2016-02-13 16:11:19] Mangalica : i'd win a car on Finex
[2016-02-13 16:11:30] Mangalica : or okc
[2016-02-13 16:15:52] Mangalica : i just checked their page and the first sentence reads:
[2016-02-13 16:16:07] Mangalica : "The year 2016 has been truly tremendous for Bitcoin and blockchain. "
[2016-02-13 16:16:46] Mangalica : on the blog. And they wrote this in 2015. Illuminati confirmed.
[2016-02-13 16:26:58] sleger : the car was on 0kc, mercedes c220
[2016-02-13 16:27:32] sleger : I was supposed to win it, and then they created some inside accounts with 0 fees that created fake volumes to push me off to 3rd place...
[2016-02-13 16:28:39] sleger : Which was a macbook pro and an iphone I believe, then they said they could only ship the gifts to china so they offered to pay me btc instead, then they used some screwed up btcusd exchange rate to pay me only half of what it was really worth.
[2016-02-13 16:29:42] habibi : sleger: was it on okc or finex?
[2016-02-13 16:30:25] sleger : habibi: can you read : "sleger: the car was on 0kc"
[2016-02-13 16:31:09] habibi : yes i can! good for u didnt win, u could die while riding.
[2016-02-13 16:31:31] Mangalica : ouch
[2016-02-13 16:31:33] Tulip_stefan : lucky.
[2016-02-13 16:32:09] sleger : what does luck has to do with it?
[2016-02-13 16:33:32] Tulip_stefan : woosh.
[2016-02-13 16:34:24] arbitrage001 : so
[2016-02-13 16:34:42] arbitrage001 : should trader trust their money or btc with okc?
[2016-02-13 16:35:05] sleger : i would say yes, I still do.
[2016-02-13 16:35:25] arbitrage001 : any good reason?
[2016-02-13 16:35:28] arbitrage001 : cuz i am now worry
[2016-02-13 16:35:49] sleger : They did a shady thing to save their business 40k when it wasnt as big as now, but I doubt they would disappear with all that money
[2016-02-13 16:36:17] sleger : that was in nov-dec 2014
[2016-02-13 16:36:39] arbitrage001 : they did pretty shady thing last may also
[2016-02-13 16:36:47] arbitrage001 : with roger ver
[2016-02-13 16:36:58] arbitrage001 : and many of their employees resigned
[2016-02-13 16:36:59] sleger : "BTC 1) Mercedes C200 2) Macbook + iPhone6 3) Canon EOS 700D 4-10) iPhone 6 11-20) Xiaomi 4 LTC 1) Macbook 2) iPhone 6 3) Canon EOS 600D 4-10) Xiaomi TV 11-20) Hongmi Note"
[2016-02-13 16:37:12] sleger : I got one macbook and 2 iphone6
[2016-02-13 16:37:36] arbitrage001 : the competition is on most volume?
[2016-02-13 16:37:44] arbitrage001 : or profit
[2016-02-13 16:38:32] sleger : volume
[2016-02-13 16:38:49] Mangalica : yeah i think i remember that
[2016-02-13 16:39:23] Mangalica : as for trusting OKC, i only keep a set amount of coins there and transfer out profits
[2016-02-13 16:39:31] tscha : we should have prizes her too @BitMEX_Sam
[2016-02-13 16:39:34] tscha : here*
[2016-02-13 16:39:58] Mangalica : that's the sort of trust i have towards them. Would sting if they ran, but it has paid for itself many many times over.
[2016-02-13 16:40:04] BitMEX_Sam : We've thought about doing more social features, not so sure about corny things like giving away Apple products
[2016-02-13 16:40:07] sleger : I sent them this : "With all due respect this is not fair : the prizes have a well known USD value, so the credit should be in USD. If you want to credit me BTC you have to do it at a rate where I can convert back to an equivalent amount of USD, meaning the current rate, at time of credit. Here you forced me retrospectively to take a BTCUSD risk which I did not know about from Jan 1st, and this after the fact and the large price decline. Alternatively, you can also ship me the products as initially agreed. I was ok for the credit in BTC as a favor because I was told it was easier for you, but this can not be costing me so much."
[2016-02-13 16:40:38] sleger : answer : "I can fully understand your frustration, but unfortunately I have my hands tied by our internal terms for the competition, which mean we can only pay out according to the BTC exchange rate on the last day of the promotion. I'm very sorry but I was not aware of this when I made the call to you last week"
[2016-02-13 16:41:31] messiaen8844 : BitMEX_Sam: focus on producs and instructional, especially instruction. Like "Trade 101" video seminars.
[2016-02-13 16:41:37] arbitrage001 : should have threaten to badmouth them on bitcointalk and reddit
[2016-02-13 16:42:04] Mangalica : i really don't think they would care much about that
[2016-02-13 16:42:09] messiaen8844 : the arbitrage ones were nice
[2016-02-13 16:42:20] arbitrage001 : will really affect their reputation
[2016-02-13 16:42:23] arbitrage001 : and business
[2016-02-13 16:42:46] arbitrage001 : since trust is the primary issue for traders
[2016-02-13 16:42:59] j8 : they already have an awful reputation and people still trade there
[2016-02-13 16:43:14] habibi : true
[2016-02-13 16:43:15] tscha : BitMEX_Sam: well, apple sucks, but small bitcoin-related prizes (hardware-wallets for 50 bucks etc) would be cool
[2016-02-13 16:43:28] Mangalica : i'm just thinking after the Roger Ver thing and the allegation that a big cold wallet is at the owner's mother's house they wouldn't care that much
[2016-02-13 16:44:21] sleger : apple is awesome, usually people who badmouth apple is because they can barely afford it
[2016-02-13 16:44:38] arbitrage001 : after using xiami and asus
[2016-02-13 16:44:42] arbitrage001 : then apple
[2016-02-13 16:44:59] arbitrage001 : have to say apple quality worth its price
[2016-02-13 16:47:05] tscha : Their macbooks are nice, always have been
[2016-02-13 16:47:46] arbitrage001 : a friend of mine recommend surface pro 4 from microsoft
[2016-02-13 16:48:03] tscha : I can't bear their software, plus their peripherals are garbage
[2016-02-13 16:48:07] arbitrage001 : full wwindows os
[2016-02-13 16:48:20] BitMEX_Sam : messiaen8844: Agreed - plenty of more constructive things for us to do
[2016-02-13 16:48:22] arbitrage001 : said quality from ms is getting better
[2016-02-13 16:48:43] sleger : yes ms getting better, but windows... :(
[2016-02-13 16:49:04] tscha : typing on an apple keyboard gives me cancer
[2016-02-13 16:49:25] BitMEX_Sam : If you're a dev, unless you buy into the MS toolchain (which is getting better), Windows is not really an option
[2016-02-13 16:49:39] tscha : but their macbooks have nice keyboards compared to other notebooks
[2016-02-13 16:50:05] BitMEX_Sam : When you work on a laptop all day it's worth 2x the price to have to screw around with it less
[2016-02-13 16:50:07] BitMEX_Sam : Time is money
[2016-02-13 16:51:57] arbitrage001 : what budget laptop you recommend?
[2016-02-13 16:52:06] arbitrage001 : thought about chrome book
[2016-02-13 16:52:33] sleger : macbook pro 15 full options
[2016-02-13 16:52:36] BitMEX_Sam : I'm a big fan of chromebooks if you needs something easy/cheap
[2016-02-13 16:52:41] BitMEX_Sam : It's a good laptop for older people as well
[2016-02-13 16:52:46] Strobelov : Microsoft as a dev platform has become untenable. I spend half my day cursing at Visual Studio. The other half is spent in IntelliJ ;)
[2016-02-13 16:53:02] BitMEX_Sam : I was very impressed with the Samsung models, they have really superior build quality for any laptop < $800, and they're all of $200-250
[2016-02-13 16:53:27] BitMEX_Sam : Of course it's a very limited OS - that's both a pitfall and a strength
[2016-02-13 16:53:47] tscha : I disliked Apple from the day they introduced the iPod. Though businesswise, I have huge respect for their ecosystem in which they lock their customers, it's pure genious.
[2016-02-13 16:57:18] sleger : i have a macbook pro, macbook air, ipad air, ipad mini, iphone, but i dont feel trapped i stay by choice
[2016-02-13 16:57:42] tscha : You could really tell that Jobs was thinking far far ahead when he introduced that ecosystem with iTunes
[2016-02-13 16:59:08] Strobelov : There's something to be said for having access to my music and TV/movie libraries on my computer, phone, and TV. But admittedly, if I ever lost access to my iTunes/iCloud account, I'd be out a good bit of money.
[2016-02-13 17:00:14] Strobelov : Love my MBP, though. Until very recently, Apple had the only high-DPI support that didn't feel half-complete.
[2016-02-13 17:04:29] BitMEX_Sam : I like mine a lot, but it has the horrible habit of shutting down when it gets cold
[2016-02-13 17:04:46] Strobelov : Really? That's a new one.
[2016-02-13 17:05:06] BitMEX_Sam : Hah yeah. Kernel panic. I mitigated it somewhat by making it never go to sleep while plugged in, so it stays warm
[2016-02-13 17:05:09] Strobelov : Do you use it outdoors a lot?
[2016-02-13 17:05:15] BitMEX_Sam : But if it's in a bag for a few hours, it'll be off when I open it up
[2016-02-13 17:05:25] BitMEX_Sam : Nah not really. Cold is only about 20 C
[2016-02-13 17:05:34] Strobelov : oh wow, that's bizarre
[2016-02-13 17:05:38] BitMEX_Sam : Aka much warmer than it should have any reason to shut off on
[2016-02-13 17:05:48] BitMEX_Sam : Almost certainly a GPU issue. The dedicated GPU was a mistake
[2016-02-13 17:06:13] Strobelov : You mean having two GPUs?
[2016-02-13 17:06:33] BitMEX_Sam : Yeah, some models have an Nvidia card in addition to the built-in Intel card
[2016-02-13 17:06:45] BitMEX_Sam : It's more trouble than it's worth, the switching is buggy
[2016-02-13 17:06:46] sleger : i have the extra gpu but no such issue
[2016-02-13 17:06:46] Strobelov : Right, mine has that.
[2016-02-13 17:06:58] BitMEX_Sam : Maybe just got unlucky with it.
[2016-02-13 17:07:08] BitMEX_Sam : I'm sure some small proportion of these chips come out bad
[2016-02-13 17:07:17] Strobelov : i think mine's the second revision of the first gen retina MBP. The one before Haswell.
[2016-02-13 17:08:18] Strobelov : Sometimes my display backlight comes on, but no picture when resuming from sleep. Not often, but a bit glitchy.
[2016-02-13 17:08:44] tscha : anybody here running linux on his mbp?
[2016-02-13 17:08:59] justinlooking : t450s debian reprazent
[2016-02-13 17:08:59] Strobelov : They may have done some hacks stuff to make it resume from sleep so quickly. Might not always work reliably.
[2016-02-13 17:10:18] sleger : i have a linux in parrallels
[2016-02-13 17:10:18] Strobelov : Nah, thought about installing Linux, but I mostly do Java dev these days, and IntelliJ is far more polished in Mac OS.
[2016-02-13 17:10:34] Strobelov : Especially with high dpi.
[2016-02-13 17:11:44] tscha : Strobelov: really? Damn, I use IntelliJ a lot. Thought about getting a mbp and getting rid of osx
[2016-02-13 17:12:03] REKT : Liquidated short on `XBT24H`: Buy 6250 @ 390.31
[2016-02-13 17:12:30] Strobelov : tscha: I haven't tried it on Linux in a couple years, honestly, but it felt second-class back then.
[2016-02-13 17:14:34] Strobelov : I'll say one thing for the Mac platform: Mac developers seem a lot more disciplined about following the platform UX guidelines. Feels very unified.
[2016-02-13 17:15:04] tscha : Strobelov: seems like they implemented hipdi support for linux too now: http://blog.jetbrains.com/idea/2015/07/intellij-idea-15-eap-comes-with-true-hidpi-support-for-windows-and-linux/
[2016-02-13 17:15:48] Strobelov : Probably part of the reason high DPI felt so polished yeas ago when Windows and Linux support was a joke.
[2016-02-13 17:16:07] Strobelov : tscha: well, that's an improvement to be sure.
[2016-02-13 17:21:12] Strobelov : How's Swing's text rendering on Linux these days? That'll play a part too.
[2016-02-13 17:23:39] tscha : no clue. guess i'll have to do a lot of hardware research. but it's not urgent atm so easy going.
[2016-02-13 17:25:18] Strobelov : I'd find someone local running Linux on a rMBP and take it for a spin.
[2016-02-13 17:26:11] Strobelov : Or resign yourself to the possibility that you'll wind up falling back on OS X.
[2016-02-13 17:26:25] Strobelov : Great hardware, though.
[2016-02-13 17:27:28] tscha : Strobelov: nah, before i'll use osx i'll get a different notebook :)
[2016-02-13 17:27:39] sleger : my linux VM is centos so cant comment on what you guys are talking about
[2016-02-13 17:27:49] sleger : no UI installed
[2016-02-13 17:29:12] Strobelov : tscha: to each their own :)
[2016-02-13 17:33:00] Strobelov : The only thing that bothers me about OS X is the smug, sanctimonious attitude of much of its user base.
[2016-02-13 17:33:34] Strobelov : Then I remember that Windows and Linux users are just as bad.
[2016-02-13 17:39:50] tscha : Strobelov: that reminds me of this: https://i.imgur.com/gyU0me4.png
[2016-02-13 17:42:07] Tulip_stefan : I use windows, but i don't deny that linux is a better development platform.
[2016-02-13 17:42:34] Tulip_stefan : and OS X. What do you mean i need to buy a mac to develop phone applications? No thanks...
[2016-02-13 17:46:53] j8 : you also need to pay annually for a developer, and they take a 30% cut in the app store
[2016-02-13 17:47:06] j8 : * for a developer account
[2016-02-13 17:47:45] Tulip_stefan : yeah, exactly. shut op and don't take my money, i'll take it elsewhere.
[2016-02-13 17:48:59] j8 : was pretty annoyed when i found out i needed to pay $100 to run a program i wrote on a device i own
[2016-02-13 17:57:11] Tulip_stefan : there seems to be a lot of market makers looking to the orderbooks....
[2016-02-13 17:57:34] justinlooking : ^^j8, *$100 per year, fuckers
[2016-02-13 17:59:21] j8 : yeah seriously
[2016-02-13 18:03:11] Tulip_stefan : I really, really need to make a bot to mess with those market makers here.
[2016-02-13 18:03:17] Tulip_stefan : they are so obvious...
[2016-02-13 18:03:43] habibi : nice u talking about this
[2016-02-13 18:03:43] habibi : :)
[2016-02-13 18:21:48] tscha : Tulip_stefan: free money then for you :)
[2016-02-13 18:22:21] Tulip_stefan : meh, honest wage for my time.
[2016-02-13 18:24:40] Tulip_stefan : It's not like my time is free or anything.
[2016-02-13 18:28:33] j8 : hey @Tulip_stefan, you ever use zmq in your bot operation
[2016-02-13 18:28:54] Tulip_stefan : nope.
[2016-02-13 18:28:58] Tulip_stefan : it's single process.
[2016-02-13 18:30:59] j8 : i see, that probably won't be an issue for me
[2016-02-13 18:31:19] Tulip_stefan : now i'm confused.
[2016-02-13 18:31:41] j8 : zeromq?
[2016-02-13 18:32:55] Tulip_stefan : i'm confused about the statement that that won't be an issue for you. But my brain fails to parse what 'that' refers to :P
[2016-02-13 18:33:20] j8 : the fact that it's single process
[2016-02-13 18:33:56] BitMEX_Sam : Being single-process isn't an 'issue' :)
[2016-02-13 18:34:05] BitMEX_Sam : If properly architected it can be quite an advantage in terms of complexity
[2016-02-13 18:35:54] Tulip_stefan : my program is even single threaded, but i have a few branches with very resource heavy machine learning that runs in a separate thread, i designed it from day 1 so that that would be easy to do.
[2016-02-13 18:37:06] tscha : Tulip_stefan: you use machine learning for your bot?
[2016-02-13 18:37:49] Tulip_stefan : I spent quite a lot of time researching machine learning for my bot but failed, no strategies involving machine learning are currently live.
[2016-02-13 18:38:55] mrp1nk : @Tulip_stefan; what did you try to predict and which models did you tried so far if i may ask?
[2016-02-13 18:39:20] Tulip_stefan : machine learning is quite... black-box in this area. My current strategies have very solid math fundamentals and are easy to reason about. When i was still doing machine learning, i'd occasionally happen that i lost a large amount without any idea why.
[2016-02-13 18:40:03] mrp1nk : probably due to "concept drift"
[2016-02-13 18:40:22] Tulip_stefan : i used a stacked model of bayesian regression and knn for pattern detection and then finally used gradient boosting trees to aggregate multiple results together.
[2016-02-13 18:40:46] mrp1nk : nice, interesting...
[2016-02-13 18:40:57] Tulip_stefan : it worked on very low frequency on okcoin, i predicted the price 10 seconds into the future and went long or short using limit orders.
[2016-02-13 18:41:40] Tulip_stefan : i had a month without losses across 4 accounts with slightly different settings, around 0.7% per day. Then one day i had 30% loss and i never figured out why.
[2016-02-13 18:42:00] rapidtrades : how can u not figure out...
[2016-02-13 18:42:29] Tulip_stefan : as i said, it's very black box. You feed data to a model and it does.. stuff. No real feedback.
[2016-02-13 18:42:49] rapidtrades : dam
[2016-02-13 18:42:54] mrp1nk : here an interesting related article http://www.wired.com/2016/01/the-rise-of-the-artificially-intelligent-hedge-fund/
[2016-02-13 18:43:02] rapidtrades : Tulip_stefan: also LOL
[2016-02-13 18:43:32] Tulip_stefan : at least with my current algorithm, i can reason about the profit per trade (actually, that's one of it's parameters, profit per trade) and it matched up extremely well with the reality.
[2016-02-13 18:44:01] mrp1nk : would you mind to share which data resp. features do you tried?
[2016-02-13 18:44:16] mrp1nk : (as input)
[2016-02-13 18:44:49] rapidtrades : that sequence seems about right....if u have constant 0.7% winning days with no losses u have to know u'll pay the piper eventually
[2016-02-13 18:44:56] rapidtrades : it's just how markets work
[2016-02-13 18:45:38] Tulip_stefan : that's not true. My current algorithm makes good profits averaged over a month, and it works and keeps working wel, exceeding my expectations about the algo lifetime.
[2016-02-13 18:45:47] rapidtrades : it's either high win % with the low r/r ratio or the reverse...u can't cheat the market
[2016-02-13 18:46:03] Tulip_stefan : features: i used only price delta's.
[2016-02-13 18:46:04] tscha : Tulip_stefan: so your model wasn't based on a rnn?
[2016-02-13 18:46:07] Tulip_stefan : no.
[2016-02-13 18:46:15] mrp1nk : imo a strong set of features is more important than the particular model choice with respect to accuracy
[2016-02-13 18:46:34] mrp1nk : tscha: rnn?
[2016-02-13 18:46:48] Tulip_stefan : i tried many different features, price delta's where the best predictors by far.
[2016-02-13 18:47:01] tscha : I am not an expert, but apart from rnns, machine learning models aren't well suited for time series AFAIK
[2016-02-13 18:47:05] BitMEX_Sam : recursive neural network: here be dragons
[2016-02-13 18:47:08] rapidtrades : Tulip_stefan: what is ''price delta'' and how do u use it
[2016-02-13 18:47:08] tscha : mrp1nk: recurrent neural networks
[2016-02-13 18:47:25] BitMEX_Sam : Ah right, recurrent
[2016-02-13 18:47:53] Tulip_stefan : price delta: [the price difference between T-10 and T-9, the price difference between T-9 and T-8, .....]
[2016-02-13 18:48:12] rapidtrades : ah ok...so u follow price trends basically?
[2016-02-13 18:48:28] mrp1nk : ok, whoa, i see you guys are quite experienced in ML... didn't expected that
[2016-02-13 18:48:39] mrp1nk : (no offense)
[2016-02-13 18:48:42] tscha : Tulip_stefan: what inputs did you have?
[2016-02-13 18:48:56] tscha : features*
[2016-02-13 18:49:15] Tulip_stefan : i used price delta's as input for bayesian regression/knn and fed that to a gradient boosting tree.
[2016-02-13 18:49:24] tscha : nothing else?
[2016-02-13 18:49:30] rapidtrades : ah got it @Tulip_stefan
[2016-02-13 18:49:41] Tulip_stefan : nothing else.
[2016-02-13 18:49:42] rapidtrades : That's exactly what I was thinking as well
[2016-02-13 18:49:55] mrp1nk : did you tried stuff like RSI MACD etc?
[2016-02-13 18:50:22] rapidtrades : I was being sarcastic if it wasn't clear
[2016-02-13 18:50:38] Tulip_stefan : i tried many things, price delta's where the best by far. I actually had some nice plots with feature importances and there where clear dismissing returns with price delta's farther back.
[2016-02-13 18:50:57] Tulip_stefan : RSI/ MACD is complete bullshit, especially on the time resolution i was working with.
[2016-02-13 18:51:13] rapidtrades : 1 sec?
[2016-02-13 18:51:38] Tulip_stefan : i was predicting 10 seconds ahead but placed orders immediately.
[2016-02-13 18:51:49] tscha : You should use many inputs as you can at the same time, the more data you have, the better. Use a rnn with deltas of log return, price spread, high/low, volatility, volume (buy, sell), number of trades etc. etc. etc... even day of the week and hour of day are dependent factors
[2016-02-13 18:52:02] tscha : in fact, you need the orderbook and read the data out of there
[2016-02-13 18:52:16] Tulip_stefan : i don't like using orderbook data.
[2016-02-13 18:52:25] rapidtrades : yeah once u start optimizing that much it's a recipe for disaster @tscha
[2016-02-13 18:52:26] tscha : but it's important
[2016-02-13 18:53:33] mrp1nk : i suspect that this will make the accuracy worse due to overfitting
[2016-02-13 18:53:36] tscha : i mean, orders have influence on the price, buy/sell walls etc..... also, the frequency of the changes in the orderbook
[2016-02-13 18:53:43] Tulip_stefan : i mentioned black boxes. If there is anything that;s a black box, it's RNN...
[2016-02-13 18:53:48] laisee : simple strategies can be tested, otherwise its betting on all factors somehow.
[2016-02-13 18:53:55] rapidtrades : mrp1nk: exactly
[2016-02-13 18:54:11] tscha : you just need a lot of data
[2016-02-13 18:54:26] rapidtrades : laisee: yes, the more shit you put in, the more the odds increase that it will break down once one of the factors goes off
[2016-02-13 18:54:27] tscha : and a good machine :)
[2016-02-13 18:55:17] tscha : well you can make it multi-layered, so whole orderbook just counts as 1 feature at the final layer
[2016-02-13 18:57:05] mrp1nk : Tulip_stefan: Thank you very much for sharing what you tried so far!
[2016-02-13 18:57:47] tscha : But as long as normal strategies are working and your profit is +, you shouldn't worry much... most likely just a waste of time to cope with ML in Bitcoin.
[2016-02-13 18:57:54] Tulip_stefan : There is also this paper which i used as most of my groundwork: http://arxiv.org/abs/1410.1231
[2016-02-13 18:59:00] Tulip_stefan : tscha: i don't agree. My current algorithm is simple, but i think there are opportunities to improve details of it using machine learning.
[2016-02-13 18:59:32] mrp1nk : @Tulip_stefan: thx!
[2016-02-13 18:59:37] Tulip_stefan : the big hedge funds also use machine learning, i think. Buy those are powered by armies of phd students...
[2016-02-13 19:01:06] rapidtrades : I would guess less then 20% use strictly algo trading and of those prolly 1/20 use machine learning
[2016-02-13 19:01:10] tscha : Tulip_stefan: i just think that there is a lot of room left for your profit without ML
[2016-02-13 19:02:31] tscha : so economically it would be wise to focus on that instead of trying something where you are not even sure if it adds substantial profit (even if implemented correctly), especially since you mentioned your time is valuable
[2016-02-13 19:02:46] tscha : just trying to give advice, but you can of course do what you want :)
[2016-02-13 19:03:49] rapidtrades : most programers would be better off with learning how to trade...or try to....he's missing obvious red flags
[2016-02-13 19:04:04] mrp1nk : rapidtrades: are you? (Disclaimer: I don't use bots)
[2016-02-13 19:04:09] rapidtrades : at least demo that shit to see what's up
[2016-02-13 19:04:28] rapidtrades : mrp1nk: I am a manual trader
[2016-02-13 19:04:56] rapidtrades : but most of my systems have strict rules
[2016-02-13 19:05:13] Tulip_stefan : implement those rules in a bot and take the rest of the week off. :)
[2016-02-13 19:05:33] Tulip_stefan : you can even backtest your rules, to see if they really work.
[2016-02-13 19:05:57] rapidtrades : most of them they run on daily charts so learning how to code is not worth it for me at this point
[2016-02-13 19:06:11] rapidtrades : oh don't worry, I've backtested everything....by hand
[2016-02-13 19:06:40] Tulip_stefan : respect.
[2016-02-13 19:07:13] mrp1nk : ok one remaing question: anybody here successfully using ML in their bots?
[2016-02-13 19:07:35] rapidtrades : Tulip_stefan: :)
[2016-02-13 19:08:45] Tulip_stefan : http://algobit.org/ claims to use machine learning but to be honest i'm not sure if it really works or they're just lucky.
[2016-02-13 19:09:42] Tulip_stefan : i like how their mouseover on their trend prediction says 'ema'....
[2016-02-13 19:11:00] tscha : Tulip_stefan: how do you know if they are successful? I just see data from april 2015
[2016-02-13 19:11:30] mrp1nk : thx. yes, minimal info given at this site
[2016-02-13 19:11:49] Tulip_stefan : I don't know anything that's not on their site.
[2016-02-13 19:12:33] Tulip_stefan : sadly, in 2 years i haven't met anyone else that openly discusses their profits... lame.
[2016-02-13 19:13:00] habibi : u took them all
[2016-02-13 19:13:00] habibi : :)
[2016-02-13 19:16:29] tscha : Tulip_stefan: mabye nobody is making profits except you :)
[2016-02-13 19:17:03] Tulip_stefan : i doubt that.
[2016-02-13 19:17:17] mrp1nk : i think also some kind of sentiment/news analysis may be an important thing to consider to catch "black swan" events
[2016-02-13 19:17:38] Tulip_stefan : that's another thing where an army of phy students can come in handy...
[2016-02-13 19:44:23] Mangalica : or an army in general
[2016-02-13 20:36:33] Tulip_stefan : bahhh i really can't get bitmex into my bot platform without creating ugly code everywhere.
[2016-02-13 20:37:15] Tulip_stefan : the other 13 exchanges i support have exactly the same API, but that doesn't really work for bitmex.
[2016-02-13 20:37:35] tscha : what is so different?
[2016-02-13 20:39:42] Tulip_stefan : take the xbt chain. I would like my bot to say 'buy 0.1 XBT24H for 390". Internally, i would translate that to the amount of contracts.
[2016-02-13 20:40:25] Tulip_stefan : but the problem is that the relation amount of contract * price / amount of btc doesn't hold. Since contracts are expressed in usd.
[2016-02-13 20:40:38] Tulip_stefan : if i swap buy and sell it works sorta, but it creates confusion elsewhere.
[2016-02-13 20:42:20] Tulip_stefan : #contracts * price = #BTC, i meant.
[2016-02-13 20:43:00] tscha : wait, why doesn't the relation hold?
[2016-02-13 20:43:07] j8 : i think you can post orders using `simpleQty` in btc
[2016-02-13 20:44:11] j8 : maybe only on testnet
[2016-02-13 20:45:24] j8 : `simpleOrderQty` actually.
[2016-02-13 20:46:09] tscha : Tulip_stefan: "Internally, i would translate that to the amount of contracts." --> Why does that not work? I don't understand
[2016-02-13 20:47:02] Tulip_stefan : give me a moment, my head hurts.
[2016-02-13 20:48:57] j8 : you just need a function that uses a different formula to convert to contracts depending on the instrument
[2016-02-13 20:49:00] tscha : you said you want to buy 0.1 XBT24H for 390, then just calculate 0.1 / (390 * 0.00001) and you get your ~26 contracts
[2016-02-13 20:53:37] j8 : the `instrument` endpoint has fields like `isQuanto`, `multiplier`, `underlying`, `quoteCurrency`, `settleCurrency` that you can use also.
[2016-02-13 21:06:32] BitMEX_Sam : `simpleOrderQty` is an option as well, at some point soon we'll post a lot more documentation about how it work