How Long Before A successful AT system

Discussion in 'Automated Trading' started by August, Jan 22, 2007.

  1. TraDaToR

    TraDaToR

    What is the purpose of intraday systems? take one of your system through a 200 optimization solutions report. Typical results would be 140 positive expectations for a daily timeframe and 50 positive expectations for a 30 min time frame...

    But since the very first result showing a 200% return is on 30 min, you will choose it...

    ( sorry i don't talk about ultra fast advanced arb systems for people who doesn't pay the bid-ask spread )
     
    #21     Jan 29, 2007
  2. August

    August



    In my limited testing thus far I have definitely found the intraday much more difficult to develop for.

    A huge phenomenon I've also found is that (using tradestation), once you enable searching intraday data - that even on a daily chart, your system profit is curtailed (presumably because of stops and such... I'm investigating this right now.)


    I really appreciate the honesty of people who can say "it isn't working" despite what must be an effort greater than I could mount myself.

    Is the consensus then that a daily system with moderate profits (beating market) is reasonable, but an intraday strategy to whomp the market is hard to sustain?
     
    #22     Jan 29, 2007
  3. TraDaToR

    TraDaToR

    That's a perfect summary of what I'm thinking at the time... But you can find better strat than mine... I don't know.

    Now, I am thinking that trading is "really" for wealthy people. If making 10-15 % per year is alright for you, then go ahead... I f you're comin' up with 10K( at least, I'm not going out with less, but with a little more ) like I was, you'd better do anything but trading.

    What you have learnt is the only valuable thing that's going out.
     
    #23     Jan 29, 2007
  4. TraDaToR

    TraDaToR

    Don't get me wrong august.

    I've run a lot of out of sample tests and I'm just saying that, considering the amount of expectations you are having at the first time( like I was ... ), system and automatic trading are reward-limited, BUT are DEFINTELY WINNERS on the tests I've run( just a little more than what a bank can do for you).

    The problem is only if it can content you...
     
    #24     Jan 29, 2007
  5. nitro

    nitro

    I cannot talk for Michael. Although I am privy to some of the strategies he has used in the past, and he did very well for many years, I don't know what he is doing now. All of those strategies that he used were a hybrid arbitrage/directional trading stragegy with a built in edge, coupled with killer comissions. Those edges have mostly dissapeard. To say nothing of the fact that he put together a talented team of people to make the systems work.

    I made a humble living trading equties and never had a down year trading them. If you start with say $300K and use 2:1 leverage, and you work real hard, you can make say somewhere around 20 - 30% after comissions. That is not a bad living, but given that you have to pay your own healthcare, and that you have to pay exuberant taxes, and that you probably put in 12 hour days when all is said and done, it ain't such a lavish lifestyle for the retail trader when you have another skill and do a line item comparison of trading against a job like programming. And that's the other kicker - coming up with $300K. If you go pro/prop, you get a chance to use large leverage on say $30K which gives you the $300K+ of money to put in play. In the hands of an experienced equities trader, that can be a good thing. You also should get good comissions which helps more than you can ever imagine.

    And trading futures? Well, I have yet to see a single retail trader make money trading futures consistantly. I have done it for small periods of time, but overall futures have crushed my account, resulting in a blowout recently.

    Best of luck.

    nitro
     
    #25     Jan 29, 2007
  6. nitro

    nitro

    Yes, all my ATS are intraday and are high frequency.

    nitro
     
    #26     Jan 29, 2007
  7. August

    August

    But so I understand... you said that your intraday strategies are NOT working? Yet all your strategies are intraday?

    Sometimes with the future traders I wonder if the random attritian factor is why you can have a site like worldcupadvisor with a handful of guys making 100% (with like 40% drawdown). Anyone who blows out would eventually disappear. By sheer randomness one of them is going to end up getting lucky every year. The ones who blow out disappear from the site (presumably) and your left only with the ones who have remained lucky. It's hard to know when something is luck or skill.
     
    #27     Jan 29, 2007
  8. nitro

    nitro

    My intraday ATS futs systems make money if I could get close to floor like comissions. That is definetely an edge, but one that is impossible for me to trade profitably, since I would either have to lease a seat, putting your nut now at -$1000+/M, or an average of about 3 NQ handles a day just to pay rent. This would only make sense if I was going to trade say 10+ contracts at a time, or I would have to find some other way to get great comissions without leasing a seat - nearly impossible to do.

    You can always tell when a system has an edge. Assuming no optimization or curve fitting, if you break even or better after comissions and paying the bid ask spread, that means you have found a system worth pursuing. The fact that it makes no money is not what you should focus on, but that it is treading water after being in the hole $4.80 to the broker, + the B/A spread to the market maker on the way in and on the way out, or another $10 for NQ. You are essentially $15 in the hole per contract the second you put on a trade. That means just to break even after taxes, you have to make a handle on NQ per contract consistantly. Just to break even!!!!!! There is no way around the B/A spread unless you are a MM. The only other place to get profits is to shave the comission to the broker. Think about that.

    You need to study statistics. To any confidence level, you can estimate whether someone (or an ATS) is lucky, or has skill, as long as the number of trials is "large" enough to extrapolate the statistic. You will always find lottery winners in any game...

    nitro
     
    #28     Jan 29, 2007
    kcgoogler likes this.
  9. TraDaToR

    TraDaToR

    Thanks Nitro for those precisions, I was really disappointed when I read your initial post.

    I'm trading futures and it is no doubt harder than equities. I've some systems on equities that I am watching real time results everyday and they are really more stable than any systems I traded on futures.

    I think it comes from the relative importance of the B/A spread on a typical day. I think( I'm not sure )equities have much more daily volatility in terms of minimal increments.
     
    #29     Jan 30, 2007
  10. RedDuke

    RedDuke

    Nitro,
    I hope you meant this as a joke. Futures are ideal for intra day/retail trading. I am doing it successfully.

    August,
    As far as automation, I am yet to see a fully automated system that consistatly works. Whoever can do it is a pinacle of trading. But kepp this thought in mind. it is very difficult to succeed in trading. And it is that much harder to automate it. I would suggest to you learning to do it manually, if you can not do it, you will not be able to automate.

    Regards,
    redduke
     
    #30     Jan 30, 2007