Sunday 9 September 2012

First Year in the Jungle - Part 1

Its been a little over 1 year since leaving a good job in finance, 1 year and 1 month exactly since my employment ended and dam... so much has changed since then. To give some insight, back in 2010 I joined a front office connectivity group for a large bank building some cool tech. The plan was work my ass off, starting in technology, and slowly migrate towards the "trading" side of things - how ever you define trading. How green was I ? so green that I thought had joined a HF prop fund inside a bank... yes bright florescent green with blinking lights.. holy cow. It took about a month and half before finally admitting to myself what connectivity means, what a broker does, how we make money, our place in the food chain and most importantly its the *clients* who do all the trading - was a jarring week.


Fast forward 18months, the project built, rolled into prod, making money all is good and im there thinking how do I get to "trading". By this stage had acquired enough info to know trading at that point means UHF/HF stat arb prop trading and for some reason thought market making had no relation.. hmmm.. yes green. Worse yet there wasn`t anything remotely close available at the bank and thus took an interview at a well known, well respected quantitative fund in the area.

The interview was the worst ive ever had, like ever - my failed teenage interview for a gas station attendant went far better. Problem was I insisted on applying for a trading position (no interest in pure tech) with my tech credentials getting me the interview and my complete lack of skills flunking it in spectacular fashion. What do you mean add some numbers and take a percentage of it, without paper, without a calculator, no computer, no nothing just use your grey matter? ... oh and we are timing you. It was a skill last practiced when I was 10 years old memorizing multiplication tables for a school test and a gold star, after all that's what computers are good at? right? .... maybe.

This should have been a dude wtf are you doing slap-in-the-face, as translated into coding skills, I couldn`t even write hello world in BASIC let alone C or .

Since then after many weeks, months of banging my head against a wall staring into the abyss is, trading, even at the UHF level which in my opinion is the most demanding on the tech side vs the "trading" side is still a 100% completely different profession. What I some how missed is the mirage it presented being so similar to pure tech, because the tools are the same.

Think about that for a second, maybe its blindingly obvious to everyone and I`m a naive chump (rhetorical question lol) Yet how many professions use a screwdriver and a hammer? Hell I use one from time to time, however the difference between a plummer and electrician is very large but they both use a hammer and screwdriver.


Trading is an art, don`t confuse it with mathematics, computer science or the ability to use a hammer.

.... to be continued

6 comments:

  1. What are the concrete trading skills you are missing? Market micro-structure?

    ReplyDelete
  2. my biggest learning curve is understanding all the relationships within a symbol & between symbols & between markets & external events e.g. news.

    Guess this is what people call market micro-structure.

    ReplyDelete
  3. If trading is an art, then what's the point of building models etc.? There are those who are successful in trading (and based on the models).

    btw. Emanuel Dermans book:
    http://www.forbes.com/sites/francinemckenna/2011/11/14/a-review-of-emanuel-dermans-new-book-models-behaving-badly/

    Nice to read you again...

    ReplyDelete
  4. Yet finding a good model is not an art?

    ReplyDelete
  5. "art" means for me "I don't understand why". If you examine and measure something.. discover, it's not an art, it's a science. We can think that this is art, when our alpha isn't stable.. quickly changing and therefore its very difficult to exploit. In theory a UHF exchange arb gives you more stable alpha than mid-freq models. But your advantage depends on the speed - you're faster than someone.. tomorrow someone is faster than you.. and your alpha disappears (still exists but you lose money). Important is the experience and understanding of the process.. and then it will not be an art. So understanding... and then a question: could I __really__ exploit this alpha?

    ReplyDelete
  6. Wonderful..way to go.. May the force be with you...

    Gaurav

    ReplyDelete

Note: only a member of this blog may post a comment.