500k is a steep minimum for a strategy with no real money track record. I tried going this route about 5 years ago. My strategy required a lot less to get started so I traded my own account of about 50k for a year before I got my close family involved then my friends. At my peak, I was at about 2-3 million, but then it became difficult to grow past that. No big money institution is gonna give money to a one man show as the risk is just too great. I then looked into first loss deals and got involved with a prop firm and have been trading with them for a year. But the point is that you need a real money track record, even if it is with a small amount of money, before others will risk theirs. Once there is proof that your strategies work in real time, firms will be comfortable with you risking their money. Your best bet is to find a wealthy angel investor that's willing to plunk down the 500k so you can build a minimum 6 month track record. Then use that to set up a meeting with some credible prop firms armed with a track record. They might still require a first loss from you but at least you get more capital in the door. From there, start building some connections and possibly get a meeting with a hedge fund/family office etc and go from there. Alternative is to trade smaller with your own capital, but it's paramount that you get a real life track record asap!
Thanks for sharing your experience. Was your strategy involves stocks only, long and short, large number of stocks and fully automation or Manual trading? Why did it choke at 2-3 Million capital? (any more details if you may please?) I understand what you are saying and I see your point. It is too exhausting to do this all alone. I don't want to build a track record for myself (Man). It should be for the system (Machine). from my research of several years in back testing and automated system building, there only 3 ways to go in stock trading (not referring to options or any other instruments) 1. Trust in capitalistic system and invest in index ETF. one may go in cash and back in etf depending on market phases. 2. Fundamental research and connections with companies top management to have little inside info and investing in 20-50 companies at a time. (90% or more fund managers do this.) 3. large number of stocks portfolio trading using Fully automation. Very very few funds follow this. You can check quarterly holdings of large funds/managers and you will see how many unique stocks they are holding. For me, path is 1 or 3. and thanks for your reply.
I am a bit confused: if you are 98% automated, what’s exhausting assuming you don’t have to re-calibrate everything all the time? You said you don’t want to make this into a signal service, why not?
System taking positions after positions on stocks that you never heard of, and just the sheer volume of events happenings and trying to figure out and analyze what system is doing is all exhausting for individual. Especially when real money in on line. I have tried that on a smaller scale. I am technologist. My focus was only data gathering, backtesting, coding and creating automation framework. I never bothered about studying regulations and other Business side aspects. I just wanted to find quantifiable edge and I have done my job. There is other side to this game and that is Business side. Which I don't have and I am not much interested either. I will depend on Business side partner or consultants. I could visualize a lot more work on technology side in future when this scales up, also with changing nature of markets etc. I want to stay with technology side. It can not be be a signal service, unless you take all the signals. It is impossible to manage that many positions manually. Also the exits are calculated based on the portfolio situation dynamically and not static exits.
It sounds to me that your system has a high operational cost. If it requires 1~2 people full time to keep it running you should include 100~200 k USD in operational expenses per year. Which has to come from the profits of your trading system. I hope for you that your system still has good results after you have paid all these operational expenses.
I got stuck after I exhausted all of my immediate contacts. You may have better luck though as my strategy wasn't as fleshed out back then and I could have done a much better job presenting to new people what I was trying to do. Regardless, the overall message I got was that big money didn't want to be the biggest client in a small fund, similar to how a bank doesn't want to lend most of its money to one guy. The track record should be for the machine. You will need a way for your system to work with a capital base that you can provide it. As for #3, I think a lot of hedge funds do this over mutual funds. MFs are much more old school in how they approach investing (they do #2 mostly). #1 is just market timing and plenty of people do this, even if indirectly by trading what they believe will be the leaders of a bull/bear move.
Well, that’s what I was thinking: Let’s say you publish (electronically) the signals to client and the client decides how much to allocate etc. That process can be automated as well. Think of it as you provide the edge(s) and the client takes care of risk/position management. But, I don’t like below statement. There are not that many stocks that are really tradable, especially when you are talking billions(not that I have them or even know anyone who does)
The Risk management done at portfolio level is where substantial edge coming from. I can not quantify how much edge, but i am guessing it is extremely important. Personally I do not know all 5000 stocks and what is happenings with them. It may be allocating a very tiny amount to a small cap and bigger amount to a top 100 stocks, but it still counts as a Trade/event. system tries to find "inefficiency" that does not exists for longer time. also for very Big stocks like AAPL,AMZN the inefficiency is almost absent.