Reality based coin-tosser method that beats 95% of traders in the world.
- Thread starter Whisky
- Start date
It's a good thread Whisky. I appreciate this a lot.
There hasn't been a good thread like this for awhile. As well, there as many good trader around as before. There are still some good traders, but they aren't as willing to share some of their knowledge like how Acrary, Anetoken, and many other did.
I doubt Baron is going to ban you just because some whinny traders believe their "secret" was exposed.
There is really no secret. Everything is in archive, on the net, or on other message boards.
PA
There hasn't been a good thread like this for awhile. As well, there as many good trader around as before. There are still some good traders, but they aren't as willing to share some of their knowledge like how Acrary, Anetoken, and many other did.
I doubt Baron is going to ban you just because some whinny traders believe their "secret" was exposed.
There is really no secret. Everything is in archive, on the net, or on other message boards.
PA
It seems that "understanding price action" hasn't helped you much...as yesterday you were running Montecarlos for a 50-50 coin tosser and posting them on this 95% LOSERS thread.
When are you going to run the Montecarlos with the modified EXITS suggested?. Eh DOUCHE2?
Not man enough to apologize either. uh?
I bet you take your losses like a DOUCHE2 as well. Are you into watersports as well as it seems you like pissing on others, or that's only when you are too nervous to place a trade?.
This is something I need to put the work and research into as it will likely produce better returns... as such I'm open to suggestions.
Right now I allocate capital to models relatively evenly based on my end-of-day total fund balance (I have to do this as capital moves in and out with some frequency, usually monthly, sometimes weekly). The models are segregated by dynamic so as to avoid correlation, like vola. arb&breakout and several forms of mean reversion. They each get allocated a percentage for the concept - i.e. several models can fall into one concept and at the moment I trade 3 distinct concepts with several execution models grouped under the concept. Then the fixed per-trade capital is set based on that percentage. I then limit a concept from taking too many trades, i.e. to never exceed a set amount of the total fund base. Sometimes I miss trades as a result, but, there are days when I'm glad I didn't take those extra trades, i.e. I try to never be overwieght capital in any one concept for any given day.
An example: say I have 100k AUM and 3 concepts with 2 execution models each. I allocate 33k for each concept at 1.5k per trade each and no execution model can take more than 11 trades, each concept no more than 22 trades.
I monitor a ton of products each day, so this is all fully automated. Somedays everything fires and somedays I get just a few trades.
So as the fund grows the trade sizes grow, its basically fixed fractional and if I can figure out something better and simulate it... well, I haven't figured out a good way to do that yet...
Per the order flow, I use buy/short stops @ market almost exclusively, and I get "picked off" quite often as it is
. Limits, and I've tried to use em', have never worked well for me. Foremost, I want to get filled. Limits always fill the losers and *usually* the winners. Mike
What is there to understand, really? Price goes up, price goes down, and price goes side way.
What else? Trend identification? Consolidation identification? Reversal identification? Impact of fundamentals on TA?
PA
Have you considered allocating capital at different leverage levels?.
i.e.: A portion at x% where is x is low, so that account equity does not "swing" too much and compounds more slowly. Let's call it the "income" account.
and then a portion at y% that you attempt to trade as close as possible to f?.
You'd be amazed how equity for account 2 can surpass account 1 relatively fast , even if you only assign 5-10% funds to account 2. But it's very swingy. So at some point you may want to re-allocate.
That should not be too hard to simulate.
On the order flow, there may be brokers that pay you rebates up to your commission rates for giving them the orders, but I'm not 100% sure how that is working these days. Worth investigating as your costs seem to be a very high %.
On the second account you could make it even simpler by using your 40% 4-1 method alone, and assigning risk as close to f as possible, starting with a 5-10% total asset allocation. If the frequency is high enough your results can be really interesting.