High Frequency Prop Firm in London looking for 4-8 traders

[london_hft]

Hi,

Don't know if this is the right forum for this, but here we go.

We have set up a new office in the City for our prop trading activities, with most of the infrastructure in place (direct connectivity to bank liquidity, dedicated lines, fail over connection, desks, etc)

We are planning to on-board 4-8 existing traders who have enough risk capital in order to be able to work full-time as traders and sustain a living (realistic expectations based on your living standard). Minimum 20,000 USD is required to open up an account for this type of trading.

We'll provide free training on usage of our software suites and the trading model, high frequency research data, real-time data from government servers, squawk box and 2 suites of research software for one of our high-frequency trading models (no coding required - we have done all necessary coding - it's a semi-automatic model) - these strategies will then run on a variety of FX ECN's across a variety of currency pairs.

Although it's a semi-automatic model, there will be a lot of number-crunching, research and excel-based work on the traders side ( so we are only looking for sharp guys).

We will extend credit via our existing lines with our Prime Brokers to the traders and progressively increase the intra-day credit to the traders who generate alpha - we'll also do a healthy split with the traders from the outset (70/30 to the traders benefit).

There will be a cost to run all software suits required and real-time economic data as well as one of our execution tools + desk space in the City (which comes down to around 800 GBP per month/per trader.)

Just so there are no misconceptions:

* We are not offering a salaried job.
* We will not initially seed anyone. Risk capital is required by each trader.
* We are not going to charge for our training.
* You will not be getting "trading signals" or have a fully automated box doing the work for you - actual work (research and execution) is required.
* It's not a simple strategy that gets taught - it's a model - your inter-market and intra-market research must spot new opportunities for you and us to execute on.
* There are no guarantees for anything.
* We are not looking for part-time people to come in and go as they wish. Every trader will be in the markets 1 - 8 times a month - but we still need a lot of research done during the weeks while at the office.
* This is not a market making program - we are on the buy-side.
* We are not looking for people with unrealistic expectations - there will be losses made, there will be months when you don't generate an income, you could potentially fail in generating profits while other traders might succeed.
* Understanding of financial markets, economics, FX, maths is very beneficial.
* You should be comfortable working with spreadsheets.

Best candidate is a successful trader who wants to add a HFT model to his/her existing strategies.

Any questions at all or to meet up for a meeting, just PM me (only serious enquiries please - all requirements are provided already).

Thanks.

/Alex

 

[Benj1981]

can i use my finspreads platform for this?

 

[Benj1981]

can up size to £2 per pip if signals are gud, but i mean realy gud!

 

[london_hft]

Hi Benj,

I believe that's a retail platform. You can't really execute high-frequency trades on that type of technology. Latency, trading costs (pricing, commissions, execution methodology), counter-party balance sheet, etc with similar brokers is just not made for HFT.

PM me if you have any other questions.

 

[BSD]

Actually at first glance this looks pretty interesting.

Unfortunately I couldn't program my way out of a paper bag if my life depended on that.

 

[timsk]

. . . Every trader will be in the markets 1 - 8 times a month . . .
. . . Best candidate is a successful trader who wants to add a HFT model to his/her existing strategies . . .

Hi Alex,
I'm confused by the two parts of your post that I've quoted, as being in the markets 1-8 times a month doesn't sit well with my understanding of high frequency trading. This isn't my area, so apologies in advance if this comment is another misconception or, worse still, plain daft!
Cheers,
Tim.

 

[london_hft]

Actually at first glance this looks pretty interesting.

Unfortunately I couldn't program my way out of a paper bag if my life depended on that.

Noted..

Keep in mind that coding is done. No coding skills required. You will have to be able to do "big data research" - which you'll be taught. A little bit of spreadsheet skills and simple maths is sufficient.

 

[london_hft]

Hi Alex,
I'm confused by the two parts of your post that I've quoted, as being in the markets 1-8 times a month doesn't sit well with my understanding of high frequency trading. This isn't my area, so apologies in advance if this comment is another misconception or, worse still, plain daft!
Cheers,
Tim.

Hi Tim,

No worries.

HFT (high frequency trading) is a quite broad category.

In that category you have two main participants:

1. Price Makers (liquidity providers, banks, prop firms providing liquidity to ECNs/exchanges) - their main business is to catch arb opportunities (buy something cheaper at one place and hedge it elsewhere - either inter-market or intra-market). That's not our category and the days of doing that manually or semi-automatically is long gone.

2. Price Takers (speculators) mainly trading on events (event-based trading) this could involve trading on news, economic data or similar. This is our category and it's mainly about getting information faster than most other market participants, interpret it correctly as often as possible and execute faster than most other market participants.

You need extensive infrastructure to be able to that. Considering your competition is spending millions of dollars a year in order to beat you. That's why there is extensive infrastructure behind it:

A). Getting data/stats as soon as it's released by being co-located in the same datacentres as the body releasing it (government bodies or news agencies).

B). Acting upon that data by having models in place that execute as fast as possible at multiple liquidity venues. In order to be as fast as possible here, we also keep our trading engines at the same datacentres as major ECNs/exchanges in order to lower latency as much as possible. This enables us to execute as fast as possible before the majority of our competition has time to react.

This is also high frequency as the execution of the trades happen extremely fast (keeping in mind that the length of a human blinking is probably around 200 milliseconds and execution by our models can be as fast as 2 microseconds for each deal - meaning in theory we can execute 100,000 deals in the blink of an eye). Obviously due to the inherent infrastructure of the market, we are unable to execute that many deals - but you understand the point.

High frequency doesn't necessary mean that you are in the market all the time, it just means that "you are executing very fast". How often you are in and out depends on your models, strategies and risk tolerance (sharpest models pick very high probability deals - which are not around that often).

Hope this clarifies a bit more.

 

[PualoP]

News trading isn't worth it. There aren't enough 'dumb' people providing liquidity over news events to make it worthwhile.

 

[random12345]

I disagree with your definition of high frequency trading, i.e. "it just means that you are executing very fast".

You are describing low latency algo trading. The type of trading you describe pre-dates HFTs significantly. Obvious semantics of the acronym aside, the IOSCO defines an HFT as having a high daily portfolio turnover as well as a high order to trade ratio, which it seems you don't intend to have even if you 'theoretically' could.

Are you really co-located with multiple ECNs as such a small outfit... I find that unlikely? Most UK FX OTC traffic goes through Equinix and it costs a lot of lucre to co-locate in their LD centres.

Also raises the point that if you have capital, why you are advertising here rather than efinancial. The latter are known to accept jobs with no pay ever since they let the sub tier London props advertise. Anyway, some of what you say sounds for real, but it's still damned unusual. Best of luck with it.

 

[random12345]

News trading isn't worth it. There aren't enough 'dumb' people providing liquidity over news events to make it worthwhile.

To the naked eye where you see 50 pts in 3 ticks yes.

Every time a limit gets hit against the flow of the news, who hoovers those cars up d'ya reckon?

 

[Forexmospherian]

I am trying to work out the upside for an independent retail trader with say $40 - 60K to even bother taking up your opportunity ?

You mention you only need $20k -but surely with you charging £800 per month for rent ( ie $1450 ) and then wanting 30% of the traders profit - immediately its not looking that good with a small capital account.?

Even with $50k and lets say an experienced retail trader doing say just 10% return per month consistently and compounding - in their first 3 months they will lose between $9 to $11k to your costs - and will not have this amount to assist them compounding?

I will say I rule myself out anyway as I am at retirement age ( an old dodger according to some ) - but I have been scalping FX approx 11 yrs and over 6 yrs on a full time basis ( well 25 -50 hrs a week for at least 9 months of the year) and I can easily make 30 -50% per month on a $50-70K account - but just don' t compound as i know my financial and psychological "wall" at under 20 lots max - and so have no dreams of taking a $100 k account up to $5 million plus - as I cannot do it ;-(

My question therefore is if a potential suitable trader with $50 k joins you - you are having probably 50% off his profits in total - and therefore you need to offer him a way of producing 20 -40% per month returns rather than what he might be making presently - for any win - win result

What do you therefore expect to get his monthly returns up to with all your HFT stuff and help etc ?

i know its a different ball game with traders with $100k plus to play with - but with you mentioning such a low amount of acceptable Capital to join - then you just have to get them doing over 20-30% per month minimum to survive and prosper - or are you really expecting all the traders to leave within 3 to 6 months anyway ??

OK many will be attracted by the "dream" and not having to have a first from Oxbridge etc etc but in reality is this like a ongoing commission only job "churn" that off a continual 6 traders through out the year produces something like $200k per annum for you - with at least 16 traders leaving and feeling rather peed off ??

Regards

F

 

[london_hft]

I disagree with your definition of high frequency trading, i.e. "it just means that you are executing very fast".

You are describing low latency algo trading. The type of trading you describe pre-dates HFTs significantly. Obvious semantics of the acronym aside, the IOSCO defines an HFT as having a high daily portfolio turnover as well as a high order to trade ratio, which it seems you don't intend to have even if you 'theoretically' could.

Are you really co-located with multiple ECNs as such a small outfit... I find that unlikely? Most UK FX OTC traffic goes through Equinix and it costs a lot of lucre to co-locate in their LD centres.

Also raises the point that if you have capital, why you are advertising here rather than efinancial. The latter are known to accept jobs with no pay ever since they let the sub tier London props advertise. Anyway, some of what you say sounds for real, but it's still damned unusual. Best of luck with it.

As mentioned, HFT is a very broad definition - market makers are pretty much the only HFT's that are in the market at "all times", you rarely have buy-side models such as our event-driven models constantly in the market.

Not being a market maker or LP and being in the market "constantly" would destroy any model (in terms of trading costs).

Yes, we are co-located at LD4 and NY4 close to the matching engines on the venues where we are participating. To entry level traders, we can get them in on LD5 at a low cost as we have some servers for other purposes there as well. We're also cross-connected directly to two of the venues.

We have considered efinancial - and other sources too - thought it would be interesting to see the response from this forum as it's well established. Obviously, we wouldn't get all these comments at efinancial - and for me starting such a thing up - it's important to hear other people's opinion and how this is viewed from outside my box... I think it's a great idea.. maybe others don't

Forexmospherian: I understand what you mean - and please do keep in mind that I'm writing what I'm about to write from experience and is not any advise or such....

As a speculator, it's very difficult (impossible) to look at things with the eyes of an accountant or set average returns in a linear fashion. Keep in mind that our strategies are heavily leveraged, however we are in the markets a very short time.

Returns to be generated is dependent on how you look at things:

If you start with 20K (example):

Are those 20K all your savings or is it your risk capital for this model (going in thinking I will risk all my 20k on this)?

How much you will risk and earn - depends on your personal circumstances and market volatility - there are no expectations on our end. All you can do as a speculator is do your best and then see what happens.

However, when it comes to how much these models HAVE generated:

We generally double risk capital once or twice a year - safe-guard profits periodically (once hit) and keep risk-level at a constant. 2008 and 2010 were remarkable where we did much more than that, due to heavily increased volatility.

The issue with these models, is that they're not as scalable as I would like to in the markets we are active in. That's why I need more guys so we can expand into other FX markets. Right now, time and liquidity only allows for 2 currencies - and we have them saturated as it is - and won't let new traders do those currencies.

The plan is that each new trader would be able to cover 1-2 other currencies eventually (so we don't step on each others toes and stay responsible for our own "desk pnl"). Obviously there will be a lot of research done on how we have to tweak the models to adapt to other currency pairs. Training will be provided for all of that.

Hope my tired head clarified this a bit more.. all the best guys!

 

[OrderFlowDashPro]

london, you say you've been a successful HFT trader since 2008? If that's the case one would think that your firm would be one of the biggest around, as it is well known that most HFT's fold within a few months.

Thought I'd add this as it is somewhat relevant and an interesting read http://finance.fortune.cnn.com/2013/08/30/latency-arbitrage-costs/

 

[london_hft]

london, you say you've been a successful HFT trader since 2008? If that's the case one would think that your firm would be one of the biggest around, as it is well known that most HFT's fold within a few months.

Thought I'd add this as it is somewhat relevant and an interesting read http://finance.fortune.cnn.com/2013/08/30/latency-arbitrage-costs/

Hi OrderFlowDashPro...

We're definitely not a big shop, just profitable - big shops are normally market makers - as some of them run very labour-intensive strategies (much more labour-intensive than our models). Although our models are labour-intensive, they are much more capital-intensive than any MM shops I know of.

Market making models are scalable and can grow into major enterprises (like JaneStreet, GSA and Optiver). Our models aren't that scalable in a single market - there's just not "enough" liquidity to hit - maybe we bring in some bright guys who teach us one or two things and make it happen. Who knows..

Right now, we just need to diversify into other currencies which we're not touching due to lack of resources (in terms of people).

Have a good weekend!

 

[OrderFlowDashPro]

London, Thanks for the info. Do you find that the edges you discover "erode" quickly? This was also suggested by the same study. That study suggested there were only a few successful HFT shops. New shops are initially profitable but within 8 to 12 months they tend to become unprofitable and quit.

 

[london_hft]

London, Thanks for the info. Do you find that the edges you discover "erode" quickly? This was also suggested by the same study. That study suggested there were only a few successful HFT shops. New shops are initially profitable but within 8 to 12 months they tend to become unprofitable and quit.

The problem with "edge" being discovered is one of the greatest issues of a successful model - there are those much more well-funded than us who hire armies of people to "front-run" models like us. So you have to be able to deal with it by organising your infrastructure around it.

Main problems are:

1. If you execute via same venues and same counter-parties your model will be "decrypted" by their guys. Front-running will lead to you getting slaughtered when you try to hit the liquidity.
2. You will run the issue of getting "piggy-backed" if you're too profitable through the same venues/counter-parties - which will lead to further absorption of liquidity - leading to lower vol upon the event (less profit to be divided between all participants).
3. Piggy-backing will potentially lead to false direction as too much liquidity is being soaked by you and those piggy-backing you, which will lead to the larger MM's preventing shorter-term direction to go to your favor by aligning the liquidity they push in to the market.

All of a sudden a model which is based on an extremely short time frame can take minutes or maybe an hour to realise to your favour.

Important thing is to have your deals spread evenly across multiple venues/counter-parties and not get "spotted" too frequently. Run multiple proxies and execute contradicting models via different venues - creating an overall confusion by your counter-parties and preventing them to piggy-back or front-run your models.

Most important for us is to just milk the cow and not slaughter it - that's the only way we think we have been able to keep the models going for a "long" time.

There are those who do the opposite of us, find an opportunity and hit it until other MP's "figure out" what they're doing - if you do that (like a very famous London shop did) - you better make sure you make enough money to support you for a life-time in a very short time frame.

Those guys (according to my sources) doubled their risk-capital almost on a daily basis, and it went on for almost 1.5 years until they cashed out at the peak. Very sharp guys!! For us it's not just about creating wealth but also generating a sustainable income for those involved. However the wealth they created is far above what we have been able to do. And don't take me wrong, would I have discovered the same arb - I would have done it too!! Maybe in the long-run - we can beat them at that.

 

[supremegizmo]

looks like a sophisticated broiler room re branding itself. HFT firms do not operate in this fashion? the whole post tells the story itself! I know very good HFT firms who are struggling to squeeze ticks due to current market conditions and here come a bunch of monkey offering a split .. stay away.. this sort of jobs to my experience negates the value of cv rather than adding value to your career!