Goomboo's Journal

[Bebop]

Real time stats (at time of posting):

http://i40.tinypic.com/xqfhfr.jpg

Applying the Goomboo system (I forgot the offical name of the system) it would seem too early to go long because there is no crossover yet (at 3pm)....

I can see the crossover at 10am(ish), followed by a large volume of red. This red represents traders going short in response to the crossover right?

Can anyone speculate out loud here, and just describe what might be happening at the 1pm-3pm. Whis is the cause of all the Green. I speculate one thing -- that it can't be traders on the 21/10 because theres no crossover yet..

[proudhon]

Real time stats (at time of posting):



Applying the Goomboo system (I forgot the offical name of the system) it would seem too early to go long because there is no crossover yet (at 3pm)....

I can see the crossover at 10am(ish), followed by a large volume of red. This red represents traders going short in response to the crossover right?

Can anyone speculate out loud here, and just describe what might be happening at the 1pm-3pm. Whis is the cause of all the Green. I speculate one thing -- that it can't be traders on the 21/10 because theres no crossover yet..

It can be traders reacting to the 21/10 crossover that prompted the shorting.  Over time 21/10 trading will become less and less profitable now that it's been made a topic of discussion here.

[RyNinDaCleM]

Real time stats (at time of posting):

Applying the Goomboo system (I forgot the offical name of the system) it would seem too early to go long because there is no crossover yet (at 3pm)....

I can see the crossover at 10am(ish), followed by a large volume of red. This red represents traders going short in response to the crossover right?

Can anyone speculate out loud here, and just describe what might be happening at the 1pm-3pm. Whis is the cause of all the Green. I speculate one thing -- that it can't be traders on the 21/10 because theres no crossover yet..

I can be traders reacting to the 21/10 crossover that prompted the shorting.  Over time 21/10 trading will become less and less profitable now that it's been made a topic of discussion here.

This, and the contrarians! If everyone does the same thing, the price would plummet indefinitely. Or rise for that matter.

[Timbo925]

---> So if wont work if everyone uses it --> it will work --> everyone uses it --,
'----------------------------------------------------------------------------'

 

[Serge]

It can be traders reacting to the 21/10 crossover that prompted the shorting.  Over time 21/10 trading will become less and less profitable now that it's been made a topic of discussion here.

or maybe someone just needed to cash out some btc

[Goomboo]

Short, average price $6.21

[Goomboo]

Hello Goomboo, do you have any experience trading stock in an industry you are an expert in? Or do you ignore such factors all together?

I ignore pretty much everything but technical analysis.  I won't trade stocks that are pennies per share or going bankrupt.  I mostly do forex though.

[Bebop]

or maybe someone just needed to cash out some btc

or..

someone had $6.17 flagged, and when the drop was at 6.17 it triggered a long/leverage/whatever is called, buy up of btc.

http://i41.tinypic.com/9sbjox.jpg

Now the mystery buyer is dumping his loot, and the price drops.

I speculate that its not the expliciit 21/10ers though, because they are still patiently waiting for the next crossover.

(all speculation).

[EDIT] Goomboo, this system is uber-intriguing. I've posting all this stuff here in your thread, because you are the teacher of teh system, but if its too much, let me know.. I can shift on out and make a new space. Although.. I think I'm done for now in any case.

[jojo69]

pssst...hey guys

if you had been using a 5-15 crossover you would have got out at 6.36, perhaps earlier if the MACD had made you jumpy already

my point is, fuck around with the charts and find a system that YOU think you believe in for this market

[Serge]

or maybe someone just needed to cash out some btc

or..

someone had $6.17 flagged, and when the drop was at 6.17 it triggered a long/leverage/whatever is called, buy up of btc.

......

Now the mystery buyer is dumping his loot, and the price drops.

I speculate that its not the expliciit 21/10ers though, because they are still patiently waiting for the next crossover.

(all speculation).

just a weekend ping pong session

[Goomboo]

What are your thoughts on RSI? It seems like another simple yet popular indicator for overbought / underbought

Hey Mav,

I don't like oscillators / overbought / oversold kind of stuff in this type of market because currency markets typically trend.  An oscillator basically tells you when something is too expensive or too cheap - in a trend this means you would be buying all the way down and selling all the way up.

To demonstrate my point, here's a quick strategy I've created in NinjaTrader and backtested on EUR/USD daily candles since 2000.  It's from a concept called "The RSI is Wrong" written about in Active Trader Magazine.  Basically it says when the RSI goes over 70, you should be buying because it's a breakout and when it goes under 30, you should be selling for the same reasons.

Here's a picture of when it trades:



As you can see, this essentially tries to short bottoms or go long on tops betting that the trend will continue.  If we were using the RSI as it was originally intended, we would be trading against the prevailing trend - which is typically a no-no unless your strategy / risk management calls for it.

Here is a chart of the performance of "The RSI is Wrong" since 2000:



It's not the best strategy, but it is profitable - so there is money to be made in using the indicator to determine breakout rather than reversal.  This is why I'm cautious of using indicators as advertised.

[Goomboo]

Oh, and one more word about leverage!

I do use Bitcoinica, but I DO NOT use more than 2.5 times leverage.  My method is a trend following system which simply means I'm looking for a new trend to start and I want to be trading in the direction of that trend as long as it remains.  Trend following systems typically only have a 30% profitability ratio which means that if you trade, you have a 70% chance of being wrong and losing money on that trade.  If you are using 10:1 leverage and are wrong 3-5 times in a row, you're bankrupt!

Leverage is great in some situations, but it is a double-edged sword.  The beautiful thing about leverage from my perspective is that it allows individuals to practice fixed-fractional money management on lower timeframes.  Basically this means that leverage allows you to adhere to your risk management system by giving you buying power.

An example of an appropriate use of leverage:
-You have a $1,000 account.
-You are willing to risk 2% on a trade.  This means that if you're wrong, you lose $20.
-You know that a logical stop / place for you to exit if you are wrong is $.30 away from the market price.
-This means that you should buy > Dollars At Risk / Price move > $20 / $.30 = 66 bitcoin.  If you didn't have the money to buy 66 bitcoin, leverage finds a use.

Yea money management is just as important as a strategy.  I also recommend people back-test.  Do you use any oscillators with your EMA signals?

Hi dropoutbox,

I have found that the more indicators I add to a strategy, the more confusing / curve-fitted it becomes.  The premise behind me using just two moving averages is that I believe BTC is in a trend more often than in a trading range, so I should be trading accordingly.  Also, I posted a few minutes ago showing that from a pure mechanical standpoint, oscillators are typically wrong in calling the immediate trend direction.

[Timbo925]

Maybe a stupid question but what do you do when you sleep?  Do you set some positions to liquidate or something else?

[Goomboo]

Much gratitude and respect Sir Goomboo.

I notice on the first red arrow, the 10/21 didn't actually cross-over, but just met together for a moment. In retrospect, it was the time to sell. So I guess this means the lines don't strictly need to cross-over, but a coming-together can be enough to initiate a sell.

Now look, again (right now as I type this), the 10/21 have met... lets see if the strategy plays out.

Also Goomboo, the question which is probably on the tip of a few tongues right now. Pure skepticism -- I'd like to ignore it, but curiosity has the better of me. Although you instructed people not to use your 10/21 forumla, its almost impossible to expect anyone would not use it (Diffusion of responsibilty). If someone (such as youself) had a hunch that a bunch of people would be working explcitly with the 10/21 system, is there a way that you could use a counter-measure to make extra profits from this manipulation of the trend?

(this question is asked with much respect by the way).

Hi Bebop and thank you for the kind words,

If you look closely at your chart, you will see that a buy was generated before that sell...it was brief and you lost money on it, but you should have bought and then sold short.

I COULD be buying from all the people who are selling and selling to all the people who are buying.  But frankly, that's too much effort and more risk than I'm willing to take for two reasons:

1.  I don't bet against the trend and the moving averages determine intermediate trend for me
2.  Let's say I were to try and absorb everyone's volume.  What if I don't have enough money?  10k bitcoin sell into the market and I can only buy 9k.  Now I'm in a 9k position and prices are going against me.  I would have to "puke out" and sell all of my bitcoin to protect my money, further profiting those who are following the trend.


About the 10/21 combo - there's nothing special about it.  It's just two moving averages.  You can use the 9/14, the 12/26, the 21/50...basically whatever you're comfortable with.  The idea here isn't a strict "you must use this!", but rather a concept to help people be objective about prices.

I've decided to go ahead and test this strategy across other data to demonstrate why I wouldn't trade against it.  I believe that in order to act successfully in the financial markets, you must backtest your edge and verify that it holds up in the past.  I have backtested the 10/21 moving average crossover in NinjaTrader to demonstrate its usefulness.

Here is a chart showing the entries and exits as we have discussed previously in this thread.  I just noticed that I backtested simple moving averages, but frankly there is very little difference in the two:



Below is a chart showing the equity curve in percentage gain/loss basis across the past 10 years using a simple moving average crossover system on daily candles in the EUR/USD:



Now here's a page of the statistics on how frequently it was profitable, etc.  There are several things to note here.  The first is that profit is only "$.71".  This strategy only trades $1 per trade (we're just testing entry / exit efficiency, not any sort of risk management).  This means that over 10 years you made 7,100 pips...how much money you actually made depends on how many lots you traded.  Next, look at the win percentage.  On average, you would have won 41% of your trades.  This means that every time you took a trade, you more than likely would lose.  Think about that for a minute.  In order to be successful in this business it isn't about how many times you win / lose, but how you handle those gains and losses.  This is an important realization for all of you who decide to trade the 10/21 strategy - if you have a trade on right now, you more than likely will lose money on it, but if you stick with it, in the long run you will probably be profitable.

[Goomboo]

Also Goomboo, the question which is probably on the tip of a few tongues right now. Pure skepticism -- I'd like to ignore it, but curiosity has the better of me. Although you instructed people not to use your 10/21 forumla, its almost impossible to expect anyone would not use it (Diffusion of responsibilty). If someone (such as youself) had a hunch that a bunch of people would be working explcitly with the 10/21 system, is there a way that you could use a counter-measure to make extra profits from this manipulation of the trend?

(this question is asked with much respect by the way).

That's what I meant when I said this isn't going to work for long.

Thanks for the feedback proudhon,

Here's what I suggest - if you think that this won't work for long, then I invite you to trade the 9/20 crossover...be 1 step ahead of the rest of us.  The numbers really don't matter.  What really matters is that it is a simple system which follows the trend.

[Goomboo]

Also Goomboo, the question which is probably on the tip of a few tongues right now. Pure skepticism -- I'd like to ignore it, but curiosity has the better of me. Although you instructed people not to use your 10/21 forumla, its almost impossible to expect anyone would not use it (Diffusion of responsibilty). If someone (such as youself) had a hunch that a bunch of people would be working explcitly with the 10/21 system, is there a way that you could use a counter-measure to make extra profits from this manipulation of the trend?

(this question is asked with much respect by the way).

That's what I meant when I said this isn't going to work for long.

So if you will short when there is a crossover (and many ppl do this). This creates a extra downward momentum so the way to make a profit from this is to also sort yourself. So in a way if many ppl use it. Don't really see how you could make extra profit of this.  If the market goes up again there is a new crossover so you if you this system, you will sell your shorts and go long.
(I'm not an expert just a enthusiastic noob.  )

Yep, that's right Timbo925,

Trend following often works because other trend followers trade using the same concepts.  Basically money just consistently piles into one side of the market and a trend begins.

[Goomboo]

Also Goomboo, the question which is probably on the tip of a few tongues right now. Pure skepticism -- I'd like to ignore it, but curiosity has the better of me. Although you instructed people not to use your 10/21 forumla, its almost impossible to expect anyone would not use it (Diffusion of responsibilty). If someone (such as youself) had a hunch that a bunch of people would be working explcitly with the 10/21 system, is there a way that you could use a counter-measure to make extra profits from this manipulation of the trend?

(this question is asked with much respect by the way).

That's what I meant when I said this isn't going to work for long.

So if you will short when there is a crossover (and many ppl do this). This creates a extra downward momentum so the way to make a profit from this is to also sort yourself. So in a way if many ppl use it. Don't really see how you could make extra profit of this.  If the market goes up again there is a new crossover so you if you this system, you will sell your shorts and go long.
(I'm not an expert just a enthusiastic noob.  )

Right, until enough people do this, then when there's a crossover I watch everybody short, I put a few thousand in bitcoinica on 10:1 margin and squeeze everyone out.

Good point. But most important thing is not working met a big leverage. Because of the way it works on bitcoinica, the risk is to high. To protect you from this it would be quite easy. If your short on 6.5 just set a stop order on 6.7. This way you liquidate your position when the market turns against your position. This of course will lose you money. But it also protects it.

I was also wondering about the values to use for the moving leverages. Maybe the gap between the everages should be variable compared to the margins between the sell and buy price. (Because this is a direct loss your taking when selling/buying fast on bitcoinica). So if the sell/buy price is 6.13/6.25 the margin is 0.12 you should be using something like 10/22 for the moving averages.
Totaly not sure about this but it seems if theire is a great gap between buy/sell, the higher difference in your 'moving avarages' lines will only make you jump sides when it is clearly time to switch. (You need to be sure because a change of direction 'cost' more money)
On the other hand when there is almost no margin you can switch possibly cheap so your should switch faster to maximize profits.

This is just a theory but seems logical in a way to me.  

edit: your avatar is exactly the face i'm making while writing this.
edit2: Or maybe divide the margin by 2 and make this the difference between the moving averages

Hey if that makes sense to you and it has been shown to be profitable in the past, then go for it!  I do suggest that you keep it simple though and don't confuse the trade.  You'll find in this type of trading that people tend to just add and add and add to the system until they have so many conflicting signals they are paralyzed and unable to trade.

[Goomboo]

Real time stats (at time of posting):



Applying the Goomboo system (I forgot the offical name of the system) it would seem too early to go long because there is no crossover yet (at 3pm)....

I can see the crossover at 10am(ish), followed by a large volume of red. This red represents traders going short in response to the crossover right?

Can anyone speculate out loud here, and just describe what might be happening at the 1pm-3pm. Whis is the cause of all the Green. I speculate one thing -- that it can't be traders on the 21/10 because theres no crossover yet..

I would be very cautious in assigning a motive to price action.  Rather than saying "this happened because ___", I would just say "this happened, here is how I am going to respond to it".

[Goomboo]

Also Goomboo, the question which is probably on the tip of a few tongues right now. Pure skepticism -- I'd like to ignore it, but curiosity has the better of me. Although you instructed people not to use your 10/21 forumla, its almost impossible to expect anyone would not use it (Diffusion of responsibilty). If someone (such as youself) had a hunch that a bunch of people would be working explcitly with the 10/21 system, is there a way that you could use a counter-measure to make extra profits from this manipulation of the trend?

(this question is asked with much respect by the way).

That's what I meant when I said this isn't going to work for long.

So if you will short when there is a crossover (and many ppl do this). This creates a extra downward momentum so the way to make a profit from this is to also sort yourself. So in a way if many ppl use it. Don't really see how you could make extra profit of this.  If the market goes up again there is a new crossover so you if you this system, you will sell your shorts and go long.
(I'm not an expert just a enthusiastic noob.  )

Right, until enough people do this, then when there's a crossover I watch everybody short, I put a few thousand in bitcoinica on 10:1 margin and squeeze everyone out.

You could do that - several times if you want.  For a trader who is managing his risk, he will survive and continue to trade.  Eventually a trend will continue, down or up, it doesn't matter.  Larger money will break your squeeze and trend followers will be following the trade.  This happens every day on the market - the struggle for trend continuation / reversal.  Rather than choosing a side, I prefer just to stick with something simple and focus on survival.

I want to remind everyone at again that more than likely you will lose when you trade.  Check some of my previous posts - this system at BEST wins 40% of the time and loses 60% of the time.  Proudhon is right that a reversal may happen because people will be aware.  However, with trend following, your winning trades are typically much larger than your losing trades.

[Goomboo]

Real time stats (at time of posting):



Applying the Goomboo system (I forgot the offical name of the system) it would seem too early to go long because there is no crossover yet (at 3pm)....

I can see the crossover at 10am(ish), followed by a large volume of red. This red represents traders going short in response to the crossover right?

Can anyone speculate out loud here, and just describe what might be happening at the 1pm-3pm. Whis is the cause of all the Green. I speculate one thing -- that it can't be traders on the 21/10 because theres no crossover yet..

It can be traders reacting to the 21/10 crossover that prompted the shorting.  Over time 21/10 trading will become less and less profitable now that it's been made a topic of discussion here.

If it does become less profitable, then traders should start trading the 9/20 cross and be one step ahead! :p...you see where I'm going here?  My point isn't to teach a number - the number isn't that special.  It's a concept.  Find a way to determine the trend, manage your risk, take the trade.