So reading a few books is hard these days huh?
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
C# Developer Wanting To Learn Trading Strategy Development.
- Thread starter chrisparity
- Start date
So reading a few books is hard these days huh?
Read them again. It's how I make money every day.
Chrisparity, I'd be interested in your offer, but I can't PM you, so you'll have to contact me. I'm an experienced C# developer, and despite the pessimism of other participants in this thread, I don't mind taking someone up on an offer if all it will cost me is my time and if I stand to learn something of interest to me in the process.
As I said previously, the range responses and the "controversy" is so based on people's perspective of the world. To answer some Qs and address some comments.
1. I am not an experienced developer. I have been doing trading development for over 10 years both buy and sell side. I would not call myself a "great" developer, because I know what a really "great" developer can do...but to be an low latency trader is not only about your dev skills, it is trading knowledge, quant knowledge, etc, etc.
2. Any developer can learn. If any developer could learn trading, then the world would be full of traders who code for a living. Not that simple. What people fail to realize is that you can have the world's greatest assembly of technical talent, who can build the worlds greatest engine/platform/etc, BUT WITHOUT an alpha strategy it is just a great technical achievement. Until about 2 years ago, you could make money in US equity markets just by being technically superior and building a faster execution (eg. lower latency) system....but the days of 1 (or 2 programmers) building their own High Frequency system in US Cash equities and making immediate $$ are over. There is now too much competition and latencies are now down sub 10 micros (for the software).
To the individual developer who has posted lots of somewhat negative comments because he does not think it right for someone to take up my offer...from his posts it is obvious that he is an experienced trading system developer who makes good $$ as a consultant. Please remember there are others out there who do not have your domain knowledge or who want to focus on the PL (eg. making proft from trading) as opposed to making $$ from consulting on how to build a trading system.
1. I am not an experienced developer. I have been doing trading development for over 10 years both buy and sell side. I would not call myself a "great" developer, because I know what a really "great" developer can do...but to be an low latency trader is not only about your dev skills, it is trading knowledge, quant knowledge, etc, etc.
2. Any developer can learn. If any developer could learn trading, then the world would be full of traders who code for a living. Not that simple. What people fail to realize is that you can have the world's greatest assembly of technical talent, who can build the worlds greatest engine/platform/etc, BUT WITHOUT an alpha strategy it is just a great technical achievement. Until about 2 years ago, you could make money in US equity markets just by being technically superior and building a faster execution (eg. lower latency) system....but the days of 1 (or 2 programmers) building their own High Frequency system in US Cash equities and making immediate $$ are over. There is now too much competition and latencies are now down sub 10 micros (for the software).
To the individual developer who has posted lots of somewhat negative comments because he does not think it right for someone to take up my offer...from his posts it is obvious that he is an experienced trading system developer who makes good $$ as a consultant. Please remember there are others out there who do not have your domain knowledge or who want to focus on the PL (eg. making proft from trading) as opposed to making $$ from consulting on how to build a trading system.
Quote from chrisparity:
As I said previously, the range responses and the "controversy" is so based on people's perspective of the world. To answer some Qs and address some comments.
1. I am not an experienced developer. I have been doing trading development for over 10 years both buy and sell side. I would not call myself a "great" developer, because I know what a really "great" developer can do...but to be an low latency trader is not only about your dev skills, it is trading knowledge, quant knowledge, etc, etc.
2. Any developer can learn. If any developer could learn trading, then the world would be full of traders who code for a living. Not that simple. What people fail to realize is that you can have the world's greatest assembly of technical talent, who can build the worlds greatest engine/platform/etc, BUT WITHOUT an alpha strategy it is just a great technical achievement. Until about 2 years ago, you could make money in US equity markets just by being technically superior and building a faster execution (eg. lower latency) system....but the days of 1 (or 2 programmers) building their own High Frequency system in US Cash equities and making immediate $$ are over. There is now too much competition and latencies are now down sub 10 micros (for the software).
To the individual developer who has posted lots of somewhat negative comments because he does not think it right for someone to take up my offer...from his posts it is obvious that he is an experienced trading system developer who makes good $$ as a consultant. Please remember there are others out there who do not have your domain knowledge or who want to focus on the PL (eg. making proft from trading) as opposed to making $$ from consulting on how to build a trading system.
thanks for this reaction
but tell me, if you're succesfully trading, why won't u pay for a good programmer and be done with it.
Quote from traderslair:
thanks for this reaction
but tell me, if you're succesfully trading, why won't u pay for a good programmer and be done with it.
Well, I think the answer to that is quite obvious isn't it? Whilst anyone can come on a public board and claim anything, I have no reason to doubt OP's genuinity.
OP may be working at a HFT fund, but out of political, operational, entrepreneurial,... motives he does not/ cannot launch his strategy at his current employer. Paying someone to do the work is not feasible as developing a good software architecture takes time and is expensive - capital better spent on actually funding the strategy, paying/ leasing memberships, hardware,...
OP knows that a dev with domain experience wouldn't take him up on the offer, but also knows that there is a bunch of smart kids happy to jump on the chance for the promise of partaking in potential P&L and industry exposure. Honestly, 10 years ago I'd have taken OP up on the offer as well. What's to lose apart from a little bit of time? You gain experience and have the chance to learn something of practical value. At the same time you can walk away whenever you please, if you have a better offer.
Also, I agree with OP's notion that up until two years ago technological advantage would have been enough to make a killing in HF. I'd argue that just being co-located would have been enough for relatively simple strategies to work extremely well (that coupled with the nature of the markets in late 2008 and throughout 2009 to early 2010).
This is no longer enough as the space at that frequency is becoming crowded so that a number of funds move their strategies to "lower" frequencies (in comparison) but equip them with more smarts. If you look at the SEC flash crash paper, for instances, they're estimating the aggregate HF inventory half-life to be 130 secs (from memory) on average - which is no longer HF, really, from an inventory turnover time perspective (though ES is a different beast altogether).
They don't need to estimate. When you place an order you have to say in the datapacket you send if it's a trade initiated by a robot or by a human.
They know exactly.
They know exactly.