Pages:
Author

Topic: The flipist method - page 3. (Read 13945 times)

legendary
Activity: 1615
Merit: 1000
March 16, 2012, 05:37:04 AM
Well, that was a longer time on the sidelines than usual! Now I'm back in again:
Short 19.2 BTC @ 5.3116   
legendary
Activity: 1615
Merit: 1000
March 11, 2012, 03:25:38 PM
I liquidated my position, taking a loss of 1.12 USD. Now it's the waiting game again, with no position.
legendary
Activity: 1615
Merit: 1000
March 08, 2012, 08:14:46 AM
I've gone long again, 20BTC @ $4.8983   

Thanks for the link, stochastic, I'll have a look when I have time. One reason I'm not interested in any more involved trading than this is I have little enough free time as it is...
hero member
Activity: 532
Merit: 500
March 06, 2012, 02:51:05 PM
Liquidated my long position when the profit was showing $3.72 or 3.760% with a swap of -$0.13.

I'm now holding no position with 105.20559 USD in my account.


Stochastic: sorry I didn't get around to replying earlier.

As you point out, no system will turn a losing strategy into a winning one. This is a losing strategy. That's kind of the point, actually. I'm a complete moron when it comes to trading, but you don't have to take my word for it - the strategy I'm applying is also completely moronic. It's a monkey's trading strategy.

The speculation forum is full of, well, wild speculation and curious methodology. I'm providing a base to compare against, a guaranteed bad strategy. If you can't beat me, you should get out of the game.

In case you are still interested here is a link to a pdf download of a money management book.
legendary
Activity: 1615
Merit: 1000
March 06, 2012, 05:59:32 AM
Liquidated my long position when the profit was showing $3.72 or 3.760% with a swap of -$0.13.

I'm now holding no position with 105.20559 USD in my account.


Stochastic: sorry I didn't get around to replying earlier.

As you point out, no system will turn a losing strategy into a winning one. This is a losing strategy. That's kind of the point, actually. I'm a complete moron when it comes to trading, but you don't have to take my word for it - the strategy I'm applying is also completely moronic. It's a monkey's trading strategy.

The speculation forum is full of, well, wild speculation and curious methodology. I'm providing a base to compare against, a guaranteed bad strategy. If you can't beat me, you should get out of the game.
legendary
Activity: 1615
Merit: 1000
March 03, 2012, 04:47:17 PM
Stochastic, thanks for your input. I'll formulate a reply tomorrow, too tired now.

Just updating since I forgot to earlier:

I'm long again, 21 [email protected]
hero member
Activity: 532
Merit: 500
March 02, 2012, 03:31:22 PM
Even if your balance does average to $100 over time, this is good practice for deciding on a strategy and sticking with it, despite what your emotions will tell you.



Well... not trading at all is a strategy too, if you look at it that way. I expect I won't be able to outperform that one in the long term.

Quote
the trading system is simply a vehicle to give you a positive mathematical expectation on which to use money management.

You need to get your money management correct.  Let's say we do a coin toss and every time it lands on heads I give you $2 but every time it lands on tails you give me $1.  Your expected value per coin toss is $0.50, but your optimal bet is not $1, it is 25% of the money you are willing to trade with.  So if you started with $1 then you would bet $0.25 each time.

So for this example the optimal f is 0.25.  For this example the equation is:

f = ( ( b +1 ) * p - 1 ) / b

f = the optimal fixed fraction
b = the ratio of amount won on a winning bet to the amount lost on a losing bet
p = the probability of winning the bet


Doing some backtesting you see that the player before you had a win and loss string of:

-1, +2, -1, +2, -1, +2, -1, +2, -1, +2

10 bets, 5 winning, 5 losing or a total of $10 won and $5 lost.  If you originally bet $1 then you would have made a 90% return.

Here,

b = $10 / $5 = 2
p = 0.5

so,

f = ( ( 2 + 1 ) * 0.5 - 1 ) / 2 = 0.25


What this f value means is that if you bet 0.25 of your total stake then you will leave with the optimal amount of winnings.  So for example if you only had $1 your optimal bet is $0.25.  If you had $10 your optimal bet is $4.  If you bet too much, for example your whole stake of $1 then you could quickly lose all your money and can't bet anymore.  If you bet too little then you don't win enough money to make up for the times you lose.

For your flipist strategy, you won't have only two outcomes.  You can lose or win any amount of money.  So the equation is a little more complicated and you need to do backtesting.  The backtesting will show if there is a positive expectation to your strategy.  If there is no a positive expectation then no type of money management will help you win money.  It is like those people that go to the casino and think their "system" works.  In a casino they have a negative expectation of winning so nothing they can do, besides not gambling, will lead to a positive expectation.

Then with that positive expectation you are able to see which f is optimal.  You run the equation below with f in a range from 0.01 to 1 and see where the geometric mean reaches its peak.

G = { Sum( 1 + f * (- Trade i / Biggest Loss ) ) } ^ (1/N)

- Trade i is the profit or loss (with the sign reversed) on the ith trade
Biggest loss is a negative number of the largest drawdown
N is the total number of trades


With your current strategy you are making 3 decisions.
1) you flip a coin and based on the flip you buy, sell, or do nothing. 
2) you decided how much to risk
3) you decided to reinvest any profits
legendary
Activity: 1615
Merit: 1000
March 02, 2012, 11:54:17 AM
Even if your balance does average to $100 over time, this is good practice for deciding on a strategy and sticking with it, despite what your emotions will tell you.



Well... not trading at all is a strategy too, if you look at it that way. I expect I won't be able to outperform that one in the long term.
donator
Activity: 2058
Merit: 1007
Poor impulse control.
March 01, 2012, 05:58:16 AM
Even if your balance does average to $100 over time, this is good practice for deciding on a strategy and sticking with it, despite what your emotions will tell you.

legendary
Activity: 1615
Merit: 1000
March 01, 2012, 05:47:33 AM
I've divined from my readings of random.org that the exchange rate will go  up, down or sideways. In the face of uncertainty, I've
liquidated my short position at a loss of ~2.37% and now hold no position. We'll wait and see!

my balance: "$101.52193"
legendary
Activity: 1615
Merit: 1000
February 27, 2012, 05:37:52 AM
I've liquidated my long position , at a P/L of  -$1.11, or -1.066% and swap of -$0.09

This leaves me with $103.81294 in my account.

The flips seem to be pretty action-oriented lately, and the same trend continues. I went short 20.5 BTC @ $4.8934
legendary
Activity: 1615
Merit: 1000
February 26, 2012, 09:58:56 AM
I could have sworn I posted here but apparently I forgot to click "post".

on the 24th I liquidated my long position at a profit of around 13% IIRC and went short, liqudating that position yesterday (didn't have time to come write here as I was on the move) at a profit of around 0.4% and went 21 BTC long @ 4.9461. Currently I have $104.98754.
hero member
Activity: 714
Merit: 504
^SEM img of Si wafer edge, scanned 2012-3-12.
February 22, 2012, 04:45:59 PM
Looking good, current bitcoinica bid at $4.48.

Don't count your hens until they've hatched... I don't know when I'll be liquidating again.

I wonder how many people actively trading here have done better than I have with this method, and how many have done worse.
Daytrading: no.
Arbitrage: Better.
legendary
Activity: 1615
Merit: 1000
February 22, 2012, 04:44:16 PM
Looking good, current bitcoinica bid at $4.48.

Don't count your hens until they've hatched... I don't know when I'll be liquidating again.

I wonder how many people actively trading here have done better than I have with this method, and how many have done worse.
hero member
Activity: 714
Merit: 504
^SEM img of Si wafer edge, scanned 2012-3-12.
February 22, 2012, 04:37:49 PM
I've liquidated my short position and went long. The liquidation netted me a 3.668% loss, or 3.42 USD. I'm now 20.5 BTC long @ $4.3963
Looking good, current bitcoinica bid at $4.48.
legendary
Activity: 3066
Merit: 1147
The revolution will be monetized!
February 22, 2012, 03:41:02 PM
Chaos math is not primarily to do with randomness, in fact chaotic systems run in a clockwork universe (for our purposes).

It is more that a chaotic system has a large change in output for a terribly small change in input. In the real world we cannot measure these inputs with infinite precision making the output unpredictable.

Markets are chaotic because there is no way to know all the actors and their motivations. We cannot predict that tomorrow a bitcoin proponent will win the lottery and pump a load of advertising into the world.

Equally... some large player with 50k coins could have his finger over the mouse button right now and reading this post distracts him for 30 seconds, long enough for the phone to ring and him to leave his desk. Small input large output.

Mathematical chaos is not literary chaos.
Well stated.
legendary
Activity: 1615
Merit: 1000
February 21, 2012, 05:20:57 AM
I've liquidated my short position and went long. The liquidation netted me a 3.668% loss, or 3.42 USD. I'm now 20.5 BTC long @ $4.3963

hero member
Activity: 532
Merit: 500
February 16, 2012, 06:55:08 PM
Quote
I tend to believe markets, like complex systems in general, are mostly chaotic and unpredictable


that is one of the main problems I have with libertarians and many people who dont understand math. Chaotic, does not equate to perfect randomness. A tree growing is chaotic, but in the end, the end result will be very tree like. If it was random, sometimes a tree would look like a bear. The only thing chaotic math says is that your answers will be in probabilities rather than exact numbers. THAT DOES NOT MAKE IT UNPREDICTABLE, just less exact.

Think of it this way.  The normal non libertarian weather man, gives us weather reports. He gives us the probability it will rain based on what happened the last time conditions where this way. It does mean he is never officially wrong. But generally speaking if he says there is an 80% chance of rain, it most likely will rain that day.

with your totally random method.. say flipping a coin, you are going to be wrong about half the time, way way way way way worse than the weather man and his chaotic system.


it drives me absolutely nuts to read milton or mises and listen to them claim that chaos is totally unpredictable and as such totally useless in economics. Do they not grab a fucking umbrella when the weather man says 80% chance of rain? why the fuck do they think he gives percentages? Chaotic math is not the every day math we are used to, just like realitivity described a time we are not used to in our every day non light speed traveling lives. That doesnt make chaotic math wrong or useless or equated with randomness.

I think the problem is that people think stochastic or randomness means that a person cannot predict was is going to happen.  Really the definition of stochastic is a system whose behavior is non-deterministic.  A deterministic system will always produce the same outcome with the same inputs.  A non-deterministic system can produce different outcomes using the same inputs, but some outcomes maybe be more likely to occur than other outcomes.
legendary
Activity: 1904
Merit: 1002
February 16, 2012, 12:24:23 PM
Chaos math is not primarily to do with randomness, in fact chaotic systems run in a clockwork universe (for our purposes).

It is more that a chaotic system has a large change in output for a terribly small change in input. In the real world we cannot measure these inputs with infinite precision making the output unpredictable.

Markets are chaotic because there is no way to know all the actors and their motivations. We cannot predict that tomorrow a bitcoin proponent will win the lottery and pump a load of advertising into the world.

Equally... some large player with 50k coins could have his finger over the mouse button right now and reading this post distracts him for 30 seconds, long enough for the phone to ring and him to leave his desk. Small input large output.

Mathematical chaos is not literary chaos.

+1

Also @joulesbeef... why are you bringing libertarians into this.  Misunderstanding math is not politically discriminatory.  Also, you don't understand chaotic math as well as realnowhereman. Your point about patterns is valid, but that's more fractal related.  To your credit, these two subjects are often studied together because they are analyzed with similar tools, and sometimes appear together.
hero member
Activity: 504
Merit: 502
February 16, 2012, 11:08:12 AM
#99
Chaos math is not primarily to do with randomness, in fact chaotic systems run in a clockwork universe (for our purposes).

It is more that a chaotic system has a large change in output for a terribly small change in input. In the real world we cannot measure these inputs with infinite precision making the output unpredictable.

Markets are chaotic because there is no way to know all the actors and their motivations. We cannot predict that tomorrow a bitcoin proponent will win the lottery and pump a load of advertising into the world.

Equally... some large player with 50k coins could have his finger over the mouse button right now and reading this post distracts him for 30 seconds, long enough for the phone to ring and him to leave his desk. Small input large output.

Mathematical chaos is not literary chaos.
Pages:
Jump to: