Modeling novice traders can be done with a fairly simple algorithm. Make subtle changes to the algorithm 100 times and presto...there is your team of losing traders. Use a significant standard deviation channel around the equity curves and use the top of the channel to close out the inevitable minority of lucky "traders". Ride the rest. Works great...been doing it for years.
You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
The Most Sure-fire Hedge Fund Possible
- Thread starter brettman9
- Start date
Modeling novice traders can be done with a fairly simple algorithm. Make subtle changes to the algorithm 100 times and presto...there is your team of losing traders. Use a significant standard deviation channel around the equity curves and use the top of the channel to close out the inevitable minority of lucky "traders". Ride the rest. Works great...been doing it for years.
Quote from Random.Capital:
It won't work, because taking the "wrong" entry position is not why most traders fail. 98% of the time, across all markets, you can enter either long OR short and come out ahead.
It's choosing the exit point that is the problem, and that is impossible to "invert".
Chuck Lebeau used a similar logic in teaching his new futures traders in the 80's. He had them give him - in his own personal account - the worst trades they could think of (ie, If they were sure corn futures were going to die, he got long corn, etc).
And by properly handling the exits, he showed them exactly what you're positing: he made money, on average. So your point is well taken.
But I don't follow you in terms of how this would work.
If what novice traders believe will be a good idea, by and large, leads to losses, the inverse will have to be true if the buttons are mis-labeled. The entries and exits are both inverted if the buy and sell buttons are. An account that loses 20k in a week would have to be up 20k in the same period if the buy and sell orders were inverted.
Quote from StLouisTrader:
I think ur a little late on this one. It's called the FX bucketshop model. Good idea tho. It does appear to work well in practice.
Indeed. Good point. And they don't have legal problems and they don't soften the blow on the way out.
In fact, the whole industry is in part already using this idea to some extent in one manner or another.
Quote from ecritt:
Modeling novice traders can be done with a fairly simple algorithm. Make subtle changes to the algorithm 100 times and presto...there is your team of losing traders. Use a significant standard deviation channel around the equity curves and use the top of the channel to close out the inevitable minority of lucky "traders". Ride the rest. Works great...been doing it for years.
Maybe I'm not getting it, but it seems the algorithm would have to be pretty complex.
Care to elaborate?
Quote from brettman9:
You start a hedge fund. You hire 100 novice traders. You match their capital 3:1. You set up a trading floor with individual trading stations.
BUT - and here's the kicker - you have a trading interface with the buttons switched. Everytime they think they are buying, they're actually selling. They see their p/l * -1 (i.e. when they think they're having a bad morning, and they start to trade emotionally and lever up to get it back, they're actually having a hugely good day).
When they think they have lost all their original money, you let them go and hire replacements, telling them that you absolve them of their losses and refund them their original money (which, of course, is a small portion of the huge gains they have made for you).
The details of exactly how to set it up are unimportant...the key point remains: If you believe that trading is hard, on average, and the odds are that most traders will fail, at least at first, then you also believe, ipso facto, that the inverse of those statements is also just as true for an inverted trading interface (ie, trading is easy, on average, and the odds are that most traders will succeed, at least at first). Their decision-making would be a consistent money machine.
I think this would actually work in practice (except for possible legal issues).
I also believe that it is proof, for the trading nihilists who seem to show up on this website from time to time, that consistently winning trading is possible.
You're the OP and this is your thread. So you get to do what I suggest and make the money.
Set up a circular room with screens facing in and great trader seating facing out. On each trading setup alternate the markets traded.
Recruit only proven detractors and guys who criticize others. Use TZ, T28, ivanovich, bighog, RN and T^^^ as models. You will only need about 25 to set world records.
The mechanics of the room are very important. The trading desks move clockwise and the seating center disk moves counterclock wise. The indexing of the floors is set up so that periodically one or the other moves but never by waiting the same amount of time. Trading desks always go in the market arbitrarily when the desk is moved. These 25 traders are moved every so often too and they can be moved more than one position in a move. The movements are as they say "random" or at best "chaotic". Use chaotic.
Traders can ONLY do EXITS of the trade they found already underway. Then they sit until they are moved to another station to do their criticizing and exiting. Make them wait and watch after they fucked up.
You can always be setting the scene (remotely) and be mind fucking these guys as you wish. The bigger the assholeness they exhibit, the more you fuck with their minds as they arrive at each station.
Their job is to only outsmart the trade they find underway with their critically honed detractor minds.
As the week goes by the trading hours change relative to regular trading hours. To keep these people upset with their peers, change the seating daily so different people sit next to each other. always make the following day an hour longer too. By Friday you will really be churning out the profits.
Flipping a switch to do the same or opposite could be optional; probably using trending and rangebound as the criteria would be a refinement you could do after a while.
Do profit sharing artificially. Keep decreasing the share as each week goes by.
Naming the organization is tricky. Smartass, Inc would appeal to the employees, probably.
Newbies aren't any fun. The criticizers and detractors is where the highest bullshit level is located. Exits is the place where they really screw up too. LOL......
This could replace Cramer, I bet......
Or you could spend your time and effort just sizing up in your own trades and making more. Just a thought.
No way it would work.
The market is tougher than random. Good entries and exits exist for a small fraction of the time. When you understand that you will understand why jack's thought experiments do not equal real world profits.
The opposite of a bad traders is probably still a bad trader.
The market is tougher than random. Good entries and exits exist for a small fraction of the time. When you understand that you will understand why jack's thought experiments do not equal real world profits.
The opposite of a bad traders is probably still a bad trader.
dumb idea
b
i bet if you eliminated spread, commission, ecn fees, and other fees then most trades would just break even.
also, you would have to hide any level 1 or 2 and time and sales from them. if the bid and ask are 20.01x5000 and 20.05x2000 and the person thinks he's buying 1000 shares and he sees the bid turn into 10.01x4000 instead of the ask turning into 20.05x1000 the jig would be up. and imagine the look on a trader's face if he marketed in 2000 shares on a thin stock with only 100 shares at each level. you'd create a big spike on the chart... in the opposite way!
and i can't even begin to think about how illegal it would be
b
i bet if you eliminated spread, commission, ecn fees, and other fees then most trades would just break even.
also, you would have to hide any level 1 or 2 and time and sales from them. if the bid and ask are 20.01x5000 and 20.05x2000 and the person thinks he's buying 1000 shares and he sees the bid turn into 10.01x4000 instead of the ask turning into 20.05x1000 the jig would be up. and imagine the look on a trader's face if he marketed in 2000 shares on a thin stock with only 100 shares at each level. you'd create a big spike on the chart... in the opposite way!
and i can't even begin to think about how illegal it would be