I think that's fine for the occasional one off questions. They sound generous. But I think it wood be hard for me, personally, to expect people to give up two weeks of their time doing something extremely tedious without a fairly solid indication that there is reasonable potential for successes.
I hadn't thought of the perceived value issue. I'm ok with discussing general principles of manual and automated trading. With much more focus on specific data / matching engine issues of different exchanges. I think learning about how computers 'see' data and react to each other has the potential to help people avoid certain problems in manual trading. At least they'd know which products to stay away from vs which ones present good manual opportunities. That's worth something. Teaching a specific manual trade on a specific product would be crazy. 1) we'd be direct competitors 2) most of these can't be taught in 2 weeks 3) the ones that could be taught in 2 weeks are way too lucrative 4) I'd have to charge a lot of money to compensate for dilution. At least $1M. 5) your perceived value point would come into play.
I don't think the system has to be anything special. I read an interview with Ed Seykota where he said he taught a simple moving average crossover system. He spent most of his time and effort getting the students to follow the rules. The hard parts in learning to trade are managing emotions and following rules. The classic example of training traders was the Turtle program. As I recall, the initial training took two weeks. The Turtle story is interesting because the students as a group did pretty well. It's my belief that careful screening of the candidates was an important factor. But it was also important that their living expenses were covered, and they were supplied with capital and rules to follow.
Garachen, I commend you putting yourself out there with a very novel approach. Just a few first-blush reactions: 1. Unless you are dealing with an experienced trader who is pliable and devoid of self-destructive habits, I would have to agree that 2 weeks is likely unrealistic. 2. Floor traders used to do something very similar to this with their clerks. Only the clerks were put under contract and it was something akin to indentured servitude for a year on a profit split - and this was after the clerk had worked for the trader for at least a year already. And floor traders paid their clerks $50 per day usually. And clerks had second jobs. But having said all that, many very successful floor traders were taught that way. 3. Great deal for the student - I have my doubts, however, that this would be an appealing proposition for the mentor. Much more downside than upside in it for him, unless I am missing something. My concern would be that you would have 850 students and no mentors. I just don't picture the upside for the mentor - especially if his window to profit from the student is only two weeks. And students always start out trading small. Am I misunderstanding this ? Please tell me otherwise. 4. Some electronic prop firms here in Chicago used to teach newbies from scratch - but they usually cut them off quickly if they ran into difficulties. Also, the firm would unfortunately have a room full of traders that were all doing the same thing; and when things got bad a firm could just kill itself with the exposure. There were also three to five year contracts that the traders were under. Again, indentured servitude. That worked fine from the late '90's to the early to mid-2000's and then all of that went away very quickly. Scalping with a mouse became dangerous sport when good automation became viable. Great start ! All difficult issues require sorting out. Well done.
Maybe ET can set up a mentorship matching system. It could work like angieslist/taskrabbit or something like that. Potential mentors can advertise what services they would provide, track record, fees, types of students desired, etc. Mentors would be required to provide evidence of their track record directly to the student. Mentors and students can leave each other feedback. Fees are arbitrary and up to the mentor. I assume in many cases this wouldn't be up-front payment, so however the mentor wished to structure it. I think instead of enforcing a particular restriction on mentor's track record, just let market forces dictate who will get students and what the mentor can charge based on his own trading track record and student feedback. The benefit of doing this on ET is that in addition to feedback, there'd be some additional track record of the mentor and student's posting activity and history that can be used for evaluation. Perhaps students can also advertise that they are looking for a mentor and the profile of the mentor they are looking for, what they are willing to spend, etc. For example, a student might not want a full 2-week training course, but maybe have some particular questions they want answered and are just looking for a couple of hours of a mentor's time, etc. Anyway, not sure something like that would work. I think the big issue is whether a highly-qualified mentor could be properly incentivized at terms that a student would be willing to accept. I suspect very few successful traders have even entertained the notion of mentoring someone unless it was an unusual circumstance. I'm also not sure every profitable trader would be willing to share trading records/tax forms/etc. I suppose that could be worked out between mentor and student (e.g. maybe a mentor who is paranoid would be okay with letting the student peruse trading records for a few minutes in front of the mentor, etc.)
That's a cool idea. Almost like a dating service. It's a good thing we don't work together. It doesn't take much effort to get me to try something crazy.
So here's my idea. Please do what people here do best. Critique. Best to point out the flaws now before I or someone else goes through with it. First, I'd have to screen people for personality fit and technical ability. This is probably the most painful part. Then, I'd get 5-7 computers with 2 big monitors each and hook them up to my network but in a separate office place. I'm guessing this to cost about $50K. Upon arrival, I'd give people access to any tax, clearing and financial statements I have for up to 2 hours. I'd answer questions about them. At this point, if anyone felt uncomfortable proceeding I'd pay for their return flight back home. The rest of the day we'd discuss some aspects of manual trading, exchange data, speed and matching algorithms. I'd describe general characteristics of automated trading that I've seen work. Then I would give them access to our tick data. It's the full direct access feed from several exchanges. With nanosecond time stamps. I could point people in the direction of products that I think are good for automated trading and what features might be worth investigating. Or they can look at whatever they want. After a few days of this at least someone should have found something interesting. We would code up the strategy and start back testing and tweaking. Might do a few different strategies. Class could vote on which one to continue with and then spend several more days of tweaking and optimizing. At the end, they'd get to keep a copy of the strategy part of the code and take it with them. If it looks good, I'd run the strategy and pay out 20% of the first 6 months to the class. My guess at probability of success (annual numbers). I'm pretty confident in my accuracy here: $0. 30% $100,000 or less. 5%. (I wouldn't keep running this) Up to 300,000. 10% 300-500k. 20% 500-700. 20% >700. 15% So, my expected gross profit is around 300k/year per class. Expenses Payout to class. 30k One time setup costs. 50k My time / opportunity cost. 100k Some meals. 3k Not like I'd be making a killing, but it would help brake up the monotony of trading without feeling like I'm totally wasting my time. And, while we wouldn't focus much on manual trading directly, some of the same principles apply and there could be some crossover. Going through this experience certainly shouldn't be detrimental to someone's trading.
Yeah, opposite of a prop firm who just provides leverage and makes money off commission. But 20% with no expenses is actually pretty inline for a trading company. Usually, they structure a higher payout but include (arbitrary and high) expenses. Works to roughly the same.