Look, I've just come back from being invited for a couple of bottles of wine at what I would presume is a famous writer, given the stories told. Your issue is incredibly simple, either an event has occurred or you suspect will occur that will reduce your algo profitability. What you are seeing/sensing is shift that occurred a couple of weeks ago, the amount of time it takes to adjust your algos is longer than the time it takes for a new anomaly show up. You question is how do you adjust today for the long term, I'm still very switched on even after a bottle of wine to myself and grappa. The answer is, you can't unless you invert to manual entry and auto breakers. Now, even though I could I'm not going to waste my time understanding your issue to resolve it for a few weeks or months. You are looking for an incremental technical fix to a fundamental problem, simply because you have spent years perfecting your algos and presume you can continue doing the same. Here is the crunch, to do that you have to work harder to generate the same, meaning more backtesting. But you are already saturated, so your question is how do you skip 1002 from becoming a short, the answer is not what you want to hear, unless you put in a disproportionate number of hours testing you can't, only a trained human brain can reconcile the difference. There is a very specific reason I started Trading Experience and this is it, the inversion, there are now too many combinations to reconcile. Spend the hours to fix the algos, but spend more to adapt them. You can't, no one cares, one of the stories the writer told me is that after 25years managing 1billion sfr, they only care about the seat, not the person, the day he left he was outcast, not even a beer.
Hello, TDMA! It looks you know well about rules, will you, please, help us to find somebody in the USA ready to rise our claim - it is nine years we are in a war with Credit Suisse, Bahamas and later wit UBS Bahamas. One of the issues is that apparently and almost sure they never send our orders to the exchanges. I can give you all details. More of our war is on our website. Be well and thank you for any help!
I've had stock trades busted on me anywhere between 1 hour to 3 days later. The 3 days one was when IB forgot to notify me of the bust and it only became apparent after settlement day. These were instances where the prices were in the "clearly erroneous" range. The moral of the story is if you get a great fill, you need to sit on it (or call your broker) until you know it is real. I've also had instances where I've traded around earnings of a stock and I'll take the other side of a trade when I think I am right. And it turns out I am right, but then the counter-party busts after the fact due to "clearly erroneous" rules even though stocks are extra volatile around these news events. The exchanges do not seem to use any context in determining what is considered a bustable trade. Because of this, theoretically it seems like a viable trading strategy might be to bid a stock at "clearly erroneous" levels right before earnings are announced, get filled, and then wait for the earnings. If the earnings are great, then you make a profit. If the earnings are bad, request a bust. Can't lose! There have been instances that I have not been involved in where a stock exchange will after the fact declare a seemingly arbitrary price level to determine which trades are busted and which are not. There was a case a long time ago of someone posting a fake news story of either bad drug news or bad earnings which made the stock plummet. Traders shorted on the news and then covered for big gains only to find later that the exchange had busted all trades below a threshold price. The stock was halted and reopened close to unchanged. The traders who had covered into the panic now found themselves short at terrible prices after the reopening. Note that this kind of thing probably wouldn't happen to a stock like MSFT.