It's tough to find documentation.... I don't think hedge funds or other institutions record this information of failed strategies and then provide it to a central authority. But there is a huge turnover rate and periodic tweaking... so the stat's are under interpretation. One indirect piece of evidence is quant funds in general. Renaissance was not doing very well early in the year: http://zerohedge.blogspot.com/2009/04/renaissance-underperforms-s-by-17-in.html
I have been at the same prop shop for 10 years. The ONLY thing I have heard since I got here is that boxes will spell the end of old fashioned price discovery and "click trading" (whatever the f*ck that means). I've had staff in my own company tell me it's done. Well, i'm still here, profitable and staying positive (mentally). Boxes will do their thing, but i'm not in it to generate profits on fast price fills and such. I use some TA, but mostly fundamentals and support and resistance. I guess I am "seat of my pants" in a way, since I don't automate anything. I'm here everyday, but still enjoy myself at work and out of the office. This sh*t won't change. It might short term, but i'm trading for the long. We have been taught good risk management and how to make a living at this. I have no plans to go back to what I was doing before. They will have to drag me away from my screens with a team of wild horses.
Just a comment on black and grey boxes. There seems to be a big difference between having a good, smart programmer develop a system based on what they have determined should work in the markets...and having good traders with solid strategies who take those strategies and automate them. It does take excellence in both sides of auto trading development. I remember a few years back, a really nice guy from another industry paid a programmer to develop a system, hired a couple of people to monitor the system (always a good idea, LOL)... and did extremely well ...for about 2 years. He failed to keep up with changes, and although made great money, had to shut down that particular operation. My trading groups, who are "trades first, programmer second" do very well...because they understand that it take constant tweaking to make money. Adapting is always key IMO. FWIW, Don
Can you describe what "profitable" means, is ut 25k 75k 200++k ? Thank you Maybe describe up day down day 2k?
Just like any strategy, it depends on the trader implementing/trading it - and in the case of auto trade systems, it not only takes a very skilled TA analyst but also a very skilled programmer. Much easier to just manually trade a system with an edge so long as the trader possesses superior psychology and discipline. Auto trading is def not the way to go
90% of TA does not work. Computer programs that make money are the one's that have sophisticated algorithms and engage in some sort of hedging like arbitrage or pairs trading. They don't outright speculate like most of us daytraders do. A layman example is this: going long SPY and short QQQQ. The S&P 500 (red) is stronger than the piece of sh!t Nasdaq (yellow) today. Of course you have to believe in the concept of relative strength in order for this to be profitable. Intuition to me is priceless. Not until a computer becomes more sophisticated than a human's brains, then we will still have a possibility of profit. (Please don't bring up the chess example where the human loses... way less variables)
This was supposed to be how prop's scam people now its turning into T/A Please take that to PM, no big deal but there are thousands of ways to look at market.