The Kirk Report has a great list of things its readers have learnt this year.

One that jumped out at me was “the importance of algos [i.e. algorithmic traders, using computers to make trades] in the market.” I am not taking issue with this line — I am sure algos are important in the market. But it made me think of a question people often ask me: as a discretionary trader, why do I think I can possibly compete with these wonderful machines?

The answer is simple: either I don’t, or they are thick.

  1. A lot of short-term algorithmic trading is just doing execution for people who used to buy or sell in a more straightforward way, getting a slightly better average entry or exit price (they hope) than they would get otherwise. This will not make a lot of difference to me: OK, I might not get quite such a good entry or exit, but since I trade on daily data and hold for days or weeks, that should not make the difference between being profitable and being a loser.
  2. Another kind of short-term algorithmic trading is run in order to make money for the people behind it — “quant hedge funds” and the like. These funds hold positions for a short amount of time and trade on short-term technicals. I am not competing with these funds. Again, my entry price may be a little better or a little worse, depending on what they are doing at the time, but how much difference does it really make?
  3. There are funds that use computers to take longer-term speculative positions. In this field, being a computer is not an advantage in itself (as it is in short-term trading — computers are much faster and consistent than people). And what are the computers doing? Is it that they are programmed by people who have hit upon some magical, highly complicated technical signal? I suspect not. Rather, I think they probably try to work out, from the market action, what technical signals are likely to work in a particular market, and then trade them.

    This is what I do as well. I trade breakouts, but only when there is a trend. If either I or the computer makes money out of a trend, then the trend must exist, and thus be available for either of us to trade. Moreover, I have two advantages over a computer in deciding whether there is a trend. I have an advantage in having a human being’s naturally superior pattern-recognition ability, which enables me to simply see a trend (a computer needs to be lengthily programmed by maths PhD’s just to manage a pale imitation of what a human can do). And I have an advantage in my knowledge of the newsflow and fundamentals that are driving the trend, of which most trading machines are totally ignorant. 

None of this is to say that I will automatically do better than a computer trader — indeed, on that score the evidence is against me. But I cannot see any reason why the presence of algos in the markets means that I will not in principle be able to make money.
Advertisements