DG, Not only that, but Don states: ".....they seemed to prove a pretty reliable strategy. And I have yet to see anyone else have a strategy (in real time) that works nearly as well." Then he states in another thread: "But now to the ironic part of this?..you have actually committed a crime that helps me out?.you get all the downside, I get the upside.But, I'm not really upset....and when I explain their errors, then it will even make the strategy look better." So Don is making some claim that he was injured sufficient to take legal action, and out of the other side of his mouth he is saying that the post proves the validity of his service and this is benefical to his business? Don should run for public office, he seems to have all the requisite skills.
Tony, Do keep that risk element in the front of your mind at all times. These are some of the questions I ask myself: (I'll use your numbers in my example). Can you afford to get a 3 sigma fill day? (I'm guessing that might be around 45 fills....a really rough guess on my part). Three days ago was one of those for me. Now, what if 40 of those go against you? How do you stop out and how much slippage will you expect? Worst case, you might assume that you average a .15 loser on each stock traded (winners and losers). That'd be a negative $6 day. So doing 100 shares of each would cause a $600 hit to your P&L, and 1000 shares, $6,000. Size accordingly. Also, if your approach results in a 60% system (I'm guessing again), then also consider that it's not outside of "normal" bounds to have five or six losing days in a row. Say that a trader's 1 sigma day is 20 fills, with 13 losers. Again, I'm speculating that the 1 sigma average trade is -0.07. That would be a ($.91) day. For five consecutive days, you'd be looking at a ($4.55) drawdown. Again, at 100 shares that's only $455, but at 1000 shares, that's $4,550 against your P&L. My advice would be to size accordingly, and don't give up when you see those extreme events. The same is true when you see those five winning days in a row.... Obviously, this is a subject near and dear to my heart. I, as well as many traders that I know, run automated programs that do opens on over 100 stocks, in some cases, many times that amount. The approach is a good one, but I'm always thinking about those points above.
Thanks for the remarks Dave. My sizes are still fairly small, but they are scaled from 100 to 400 shares. Today was much less enthralling. 17 fills again 9 losers 6 winners 2 scratch But was still gross and net positive for the morning. Worst trade, BAX, 300 shares, gapped -.29 against me. ( I didn't give proper weight to the news ) I sure as heck don't want 10 of these trades. Best trades were HCA, UNH, BSX. Worst news yet was the sizable sums I left on the table by out thinking my auto exiter. Would have been up a couple of hundred bucks more if I had just let the system do it's work. At least I didn't go into complete mental lock up today. We will hit it again tomorrow and see how it goes.
I have thought about doing the many many stocks approach. What scares the hell out of me is the day when you get filled on a hundred or several hundred stocks on the LONG side, and a 9/11 event happens and I can not find a bid and spoos lock limit down, and the exchages close down. Although "unlikely," boy I just don't see how a firm let alone a trader can survive that kind of event. My guess is that a firm like Bright, since most of the calculations are similar, are getting filled with tens of thousands of shares over hundreds of stocks on any given day IN THE SAME DIRECTION. The asymmetrical risk to the firm has to be enourmous. If I were doing the spray and pray method of trying to get filled on hundreds of stocks, I would risk like crazy to go short, but I would limit my longs. nitro
I was looking around this thread to see if there have been any discussions on the seasonality of opeing orders and could not find anything. I have heard that opeings work the best during July and August, but do not know if this is true. I remember last summer that they did work really well, but was it due to the extreme volatility in general? I had some success with openings last year in general, but then had problems with them in Jan-Feb. I quit doing them until starting back up in May with a revamped strategy (I hold winners much longer than I used to, only do one side of the market each day and also vary my share size).
I haven't seen a pattern to seasonality, but have seen some correlation to the size of the opening gaps. During periods of heavy volatility like last july and october, we were getting several opening gaps of more than 1% and I was getting a lot of big win days ($2-4k). The last few months as general market volatility has come down, the opening gaps are smaller and the big wins are fewer. But the consistency of the methodology hasn't changed for me. Average win rate, winning days/month still about the same. The fill rate is a bit lower, and I've added more stocks to my list to compensate.
4 fills, 3 winners. +$500 (and that would requrire 8000 shares or about $50 in commish....not bad. Do
Do you ever use pre-market Instinet on DIA and SPY to judge how volatile the open might be? I've noticed that when pre-market volume is high I tend to do better on the open.