Using Margin for More Positions

Discussion in 'Trading' started by Corso482, Feb 2, 2003.

  1. iceman,

    let me explain. i've got a swing trading system that i've been paper trading for the past 2 years, refining it the whole time. The thing has some great ratios and statistics and has backtested well for the past 7 years (i'd test longer, but my data doesn't go back that far). Anyway, I'm not trying to boast my paper trading or my system, but I am trying to explain why I asked the original question.

    I want to divide my account into 10th's because I've found that to be about the optimum risk level for my desired volatility and returns. The problem is that the system often generates more than 10 open positions at once. So I started thinking of how I can open double the number of positions without cutting my position size in half. So the solution I came up with is to double my account through margin, but instead of doubling up my position size, just have more of the same position size, i.e.

    without margin:

    -- ten 10k positions in a 100k account, 5% stop loss

    with 2:1 margins becomes:

    -- twenty 10k positions in a 100k account, 5% stop loss

    The risk level per trade is the same, still 10k positions w/ 5% stop loss, only now I can open 20 positions at a time instead of 10, which solves the problem.

    I was trying to see if this use of margin is dangerous, which I don't think it is, given I still have a 5% stop loss on all my same sized 10k positions.

    To illustrate the impact of freak gaps on the margin strategy, I brought up the "lose all your money but no debt thing"-- it's not an accomplishment, but it shows this use of margin isn't crazy either.

    As to your comment on diversification, my system isn't really diversified. All of the stocks are highly correlated and are either all long or all short. I first establish a bullish or bearish short term bias for the dow, then I accordingly either long or short a group of highly correlated to the dow stocks (i could also use nasdaq, s&p, etc). So in that sense, the strategy is concentrated.

    I agree with your comments about diversification, though. In my testing I've found that establishing a 'market neutral' stance, with a balance of longs and shorts, leads to medicore returns, with the big cap stocks I'm testing anyway. And I'd rather not deal with illiquid small cap stocks, so i've given up on being market neutral or truly diversified.

    Anyway, I'm very satisfied with my system and it's results. I just want to be able to open more positions at once with it. So, for my purposes of wanting to take on double positions without cutting my position size in half, I think the use of margin may be viable. I'm just checking what others think.
     
    #21     Feb 2, 2003
  2. So you're saying that you've ONLY paper traded, haven't actually started trading this system yet?

    If so, I would really advise against using margin at first. I mean, if the system is so great, why not try it on a cash-only basis for a while and see if it performs as it did on paper. If so, then great... start using margin and get rich. But if not, then at least you'll learn from your system's mistakes a bit more cheaply than if you started out on margin.

    The few times I've used margin out of greed and impatience, I've regretted it. It has worked out better when I did it from a more considered & rational motivation.
     
    #22     Feb 2, 2003
  3. Yep, only paper traded, and I don't intend to use margin when I decide to go live, at first anyway.

    On a side note, the paper vs. real trading shouldn't be that different. I only deal with liquid (at least 2mm a day volume) stocks and it's a swing system, meaning executions don't have to be terribly accurate.
     
    #23     Feb 3, 2003

  4. so why not trade 10 positions with 20K. How did you select the $ weighting. Given the confidence in your system and testing, why not identify 10 high probability trades and put more into each one given your stated desire to potentially utilize margin. Not suggesting this necessarily but just discussing. Or maybe 5 positions of 20K and keep some powder dry to add more to the best performers, if indicated.
     
    #24     Feb 3, 2003
  5. I've found dividing the capital into 10th's makes for a volatility and return level I'm most comfortable with. Deciding on it was very unscientific, I just experiemented, but I did make sure the risk to my total account stayed small.

    As far as concentrating more money into fewer 'high probability trades', I haven't been able to refine my selection any further. In other words, I've given up on finding the perfect entry criterea that will give me a 90% winner rate. So I cannot identify any more high probability trades than I'm already finding.

    IMHO, with swing trading anyway, such entry criterea don't exist. I've found some criterea that give me an edge with about 1.50 winner/loser ratio. Unfortunately these criterea often generate more than 10 buy signals at once. I've also found, through backtesting, that if I reduce the number of positions I can take on, then my returns are always reduced-- It's often from those periods when the most signals are generated that the biggest returns come. Given there's no way of pin pointing which of the positions would have turned out better than others, i.e. 'high probability', i figure I'm better off just entering into them all.

    As well, doubling up on fewer positions with margin would make me more volnerable to freak gaps, which are an issue with swing trading.
     
    #25     Feb 3, 2003
  6. 20 positions ? So your risk per trade at time t can approach 10%. Don't count on diversification because there is no diversification in a hypercorrelated assets space especially in crisis situation where individual risk of assets then behave like magnetic materials all aligned together towards the same direction. 10% is high, probabistically 10 losing streaks is not rare when cumulating time or number of trials.

     
    #26     Feb 3, 2003
  7. stokhack

    stokhack

    Seems if all positions are net long or net short, margin is not a good idea because of event risk. A mix of long and short will help you sleep at nite, but does lower overall returns.
     
    #27     Feb 3, 2003
  8. Hey Corso,

    I'm starting a paper-trading Broker Dealer.

    If you can control your risk well, I'll get you a great deal on commissions.

    Hmmmmmmmmmmmmmm. Diversifying your paper trades...........

    Now THAT is risk control, man.
     
    #28     Feb 3, 2003
  9. Haha, very funny. Give me crap all you want about paper trading.
     
    #29     Feb 3, 2003
  10. At least you can take some ribbing..
     
    #30     Feb 4, 2003