oh come lebing
trading the one day cross would have been a stellar strategy looking back, and Goomboo has harangued readers to DYODD all along
Topic: Goomboo's Journal - page 4. (Read 281394 times)
As there is a lack of skepticism around motivation in this thread, I thought I should weight in here with a few cents.
Traders are in this business to make money. Even if they are "helping to educate", they still have their own motivation for doing so. I'm not saying that what Goomboo is doing is wrong, but I am saying that he is probably making a lot of money off of the people that are blindly following his advice in this thread.
For example: a trader that uses trading signals (especially bots) to execute trades regardless of other factors is highly predictable. In trading, probability is everything and if one is able to isolate exactly when others are going to be making trades, there is a huge amount of money to be made doing so because you can stack the orderbook to counter such trades. In this thread, he outlines exactly how to use predictable variables to find your own strategy. There are generations of programs that have been developed over the last 30 years that can detect these strategies (using pattern analysis) and bet against them to take the money that would have been made from these strategies.
With that in mind, be safe out there and happy trading.
Traders are in this business to make money. Even if they are "helping to educate", they still have their own motivation for doing so. I'm not saying that what Goomboo is doing is wrong, but I am saying that he is probably making a lot of money off of the people that are blindly following his advice in this thread.
For example: a trader that uses trading signals (especially bots) to execute trades regardless of other factors is highly predictable. In trading, probability is everything and if one is able to isolate exactly when others are going to be making trades, there is a huge amount of money to be made doing so because you can stack the orderbook to counter such trades. In this thread, he outlines exactly how to use predictable variables to find your own strategy. There are generations of programs that have been developed over the last 30 years that can detect these strategies (using pattern analysis) and bet against them to take the money that would have been made from these strategies.
With that in mind, be safe out there and happy trading.
I've got 2 month / 30 min period stats from bitcoincharts for bitcoin/usd price. Then I've wrote a soft that calculates ema based on this data, essentially emulating trading bot. Using your strategy with 10/21 this what I've got:
Starting price for btc was $925. I've started with 1 btc or 925 usd for simplicity.
There were 128 virtual trades made, when ema lines where crossing. After that I've ended up with 1.2085 btc == $731, when last price for btc was $605.
Kind of disappointing result. One could just take $925, buy btc for $605 and get 1.53 btc. So, what's the point?
Starting price for btc was $925. I've started with 1 btc or 925 usd for simplicity.
There were 128 virtual trades made, when ema lines where crossing. After that I've ended up with 1.2085 btc == $731, when last price for btc was $605.
Kind of disappointing result. One could just take $925, buy btc for $605 and get 1.53 btc. So, what's the point?
If I read this correctly, you lost 21% (fiat) across 2 months while price dropped 34%. Two comments on this:
1) It sounds like you are choosing a very aggressive timeframe for analysis. This is a trend following system - that means that it loses during periods of market inactivity. Trading on such a short timeframe will expose you to a lot of choppy action. I suggest re-running your studies on longer timeframes. For example, I trade the daily crossover. Over the last two months, there have been only a handful of trading signals.
2) As to your comment about disappointment in trading vs. buy and hold, here is a post from last year:
Profits are in dollars. I'm a trader, I'm only after the kind of profits that I can grow in other markets.
As mentioned, no matter what system you use in a market which tends to increase on average, buy and hold will nearly always outperform. The buy and hold people are sitting at +2000% vs. the +1700% of the system. The difference is that the system has actually banked these profits whereas if BTC were to collapse, the buy and hold investors would...hold.
This has been the recurring debate throughout the year and a half that this thread has been in existence: buy and hold vs. active trading. To restate my previous post:
It really boils down to your philosophy of the markets: are you an investor or trader? Investors believe in a concept and invest in that concept (buy and hold). Traders are after profit while maintaining the ability to live and fight another day (systematic/discretionary trading).
As mentioned, no matter what system you use in a market which tends to increase on average, buy and hold will nearly always outperform. The buy and hold people are sitting at +2000% vs. the +1700% of the system. The difference is that the system has actually banked these profits whereas if BTC were to collapse, the buy and hold investors would...hold.
This has been the recurring debate throughout the year and a half that this thread has been in existence: buy and hold vs. active trading. To restate my previous post:
It really boils down to your philosophy of the markets: are you an investor or trader? Investors believe in a concept and invest in that concept (buy and hold). Traders are after profit while maintaining the ability to live and fight another day (systematic/discretionary trading).
You are doing great here. Do you have a website?
No, I don't have a website.
@Goomboo, a few questions to you:
- Your average win/lose ratio? (number of trades won/number of trades lost)?
- Your average risk: reward ratio? (when you win, it is how many times as much as a loser trade?)
- In general how many trades do you take per day/month/or week? I don't know if it is called high frequency or not
- Your average win/lose ratio? (number of trades won/number of trades lost)?
- Your average risk: reward ratio? (when you win, it is how many times as much as a loser trade?)
- In general how many trades do you take per day/month/or week? I don't know if it is called high frequency or not
Greetings and welcome to the thread,
- On average, I win 32% of all trades
- My average losing trade results in a .79% loss of account equity
- I do not set profit targets, but my best winners have historically been 5-20 times what I'm willing to risk on the trade
- In general I target 1-3 trades per week and I'll hold positions as long as the trend remains. Losing trades are cut within a few hours and winning trades last days, weeks, or months
- I don't know anything about the bots, sorry :/
- I primarily trade the currency markets
My advice to someone wanting to get into this business but not willing to put in the work necessary to beat the competition - don't. The S&P 500 historically returns around 11.96% per year - in the long run, the casual participant is significantly better off by buying an index ETF (US: SPY is good) and focusing on the day job. 66% of currency traders consistently lose money in every single quarter. The odds don't favor speculative traders.
This is my personal opinion - I'm not a registered financial advisor. Best of luck.
You are doing great here. Do you have a website?
I've read this post an got very excited. I'm new to this trading business, so it was all interesting new stuff for me.
So I researched a matter of EMA a little, and then began testing.
I've got 2 month / 30 min period stats from bitcoincharts for bitcoin/usd price. Then I've wrote a soft that calculates ema based on this data, essentially emulating trading bot. Using your strategy with 10/21 this what I've got:
Starting price for btc was $925. I've started with 1 btc or 925 usd for simplicity.
There were 128 virtual trades made, when ema lines where crossing. After that I've ended up with 1.2085 btc == $731, when last price for btc was $605.
Kind of disappointing result. One could just take $925, buy btc for $605 and get 1.53 btc. So, what's the point?
So I researched a matter of EMA a little, and then began testing.
I've got 2 month / 30 min period stats from bitcoincharts for bitcoin/usd price. Then I've wrote a soft that calculates ema based on this data, essentially emulating trading bot. Using your strategy with 10/21 this what I've got:
Starting price for btc was $925. I've started with 1 btc or 925 usd for simplicity.
There were 128 virtual trades made, when ema lines where crossing. After that I've ended up with 1.2085 btc == $731, when last price for btc was $605.
Kind of disappointing result. One could just take $925, buy btc for $605 and get 1.53 btc. So, what's the point?
Fees included?
Switch to 4h time-frame or bigger.
wtf these post is called GoomBoo 
I've read this post an got very excited. I'm new to this trading business, so it was all interesting new stuff for me.
So I researched a matter of EMA a little, and then began testing.
I've got 2 month / 30 min period stats from bitcoincharts for bitcoin/usd price. Then I've wrote a soft that calculates ema based on this data, essentially emulating trading bot. Using your strategy with 10/21 this what I've got:
Starting price for btc was $925. I've started with 1 btc or 925 usd for simplicity.
There were 128 virtual trades made, when ema lines where crossing. After that I've ended up with 1.2085 btc == $731, when last price for btc was $605.
Kind of disappointing result. One could just take $925, buy btc for $605 and get 1.53 btc. So, what's the point?
So I researched a matter of EMA a little, and then began testing.
I've got 2 month / 30 min period stats from bitcoincharts for bitcoin/usd price. Then I've wrote a soft that calculates ema based on this data, essentially emulating trading bot. Using your strategy with 10/21 this what I've got:
Starting price for btc was $925. I've started with 1 btc or 925 usd for simplicity.
There were 128 virtual trades made, when ema lines where crossing. After that I've ended up with 1.2085 btc == $731, when last price for btc was $605.
Kind of disappointing result. One could just take $925, buy btc for $605 and get 1.53 btc. So, what's the point?
Anyone aware of a hedge fund (filed with SEC) trading BTC for profit? It seems that the handful of existing funds are acting as price exposure for investors.
Are you recommending to buy/sell as soon as they cross of to wait until there is a +.25% before buying/selling?
The system calls for buying and selling at the crossover. The .25% you are referencing is the commission for trading that is charged in that example above.
Here's an update on the performance of the hourly trading system under the highest Mt. Gox commission of .6%:
http://thetradequest.com/bitcoin/EndOfJulyUpdate.png
A couple of things worth noting:
Here's a breakdown of the monthly performance of the system, assuming a .6% Mt. Gox commission:
http://thetradequest.com/bitcoin/MonthlyReturns.png
Here is a graph showing the performance of the system assuming the lowest Mt. Gox commission of .25%:
http://thetradequest.com/bitcoin/EndOfJulyUpdateLow.png
A few things worth mentioning:
http://thetradequest.com/bitcoin/EndOfJulyUpdate.png
A couple of things worth noting:
- The system performance before any trading expenses is a 2,437% gain since the beginning of the year
- This assumes no slippage from crossing the bid-ask spread
- This assumes no commissions
- This assumes no delays in execution
- When factoring in a .6% Mt. Gox commission, trading the 10/21 crossover resulted in a net gain of 484% since the beginning of the year
Here's a breakdown of the monthly performance of the system, assuming a .6% Mt. Gox commission:
http://thetradequest.com/bitcoin/MonthlyReturns.png
Here is a graph showing the performance of the system assuming the lowest Mt. Gox commission of .25%:
http://thetradequest.com/bitcoin/EndOfJulyUpdateLow.png
A few things worth mentioning:
- Paying the lower commission of .25% resulted in a net gain of 1,278% since the beginning of the year
- This commission currently only applies to accounts trading more than 500,000 BTC per month: as trading size increases, systematic performance will vary due to increased slippage
Are you recommending to buy/sell as soon as they cross of to wait until there is a +.25% before buying/selling?
Oops, I misunderstood your question. I haven's checked that "raw data" quality.
Is the data incorrect if I have gone to: http://www.bitcoincharts.com/charts/bitstampUSD# and taken the data from "Load raw data" button aswell?
Time period "All data".
Time period "All data".
bitcoincharts' API gives you at most 2000 trades and it seems like they limited API queries to some of the most recent data (maybe days or weeks, haven't thoroughly tested that). The data given is in the correct order.
Yep, logic is pretty strange. What I found is that %Profit is close to
1 - 0.99*1.3971 = 0.383129
1 - 0.99*1.3971 = 0.383129
Yes, you need to calculate fee at every trade as it would depend on the trading sum. As far as I understand, fee is subtracted from the resulting sum of a trade.
E.g. buying 100 btc will result in 99.5 btc as a final sum. Selling 100 btc for, say, 800 USD, will result in 79600 USD.
E.g. buying 100 btc will result in 99.5 btc as a final sum. Selling 100 btc for, say, 800 USD, will result in 79600 USD.
replying to myself previous messages, in case someone would find it helpful
I have made a .hst 1 minute file compatible with btc-e's MT4, here is it: https://drive.google.com/file/d/0B3f4gdrrQPzZMS0wem1rLTd6TXc/edit?usp=sharing
here is a csv 1 minute file that can be imported in MT4 by pressing f2 - > import https://drive.google.com/file/d/0B3f4gdrrQPzZSWFHSzJNa0Z3UWM/edit?usp=sharing
Mt4 has a built in script to convert 1 min candles to all other kinds it is called period converter
but the problem with backtesting MT4 bots is that btc-e MT4 terminal miscalculates commission for all data older then 24.11.2013, the commission can get 10 times higher
If anyone is succesful in testing MT4 bots (also called expert advisors) and somehow overcome those difficulties please share how to do that.
I have made a .hst 1 minute file compatible with btc-e's MT4, here is it: https://drive.google.com/file/d/0B3f4gdrrQPzZMS0wem1rLTd6TXc/edit?usp=sharing
here is a csv 1 minute file that can be imported in MT4 by pressing f2 - > import https://drive.google.com/file/d/0B3f4gdrrQPzZSWFHSzJNa0Z3UWM/edit?usp=sharing
Mt4 has a built in script to convert 1 min candles to all other kinds it is called period converter
but the problem with backtesting MT4 bots is that btc-e MT4 terminal miscalculates commission for all data older then 24.11.2013, the commission can get 10 times higher
If anyone is succesful in testing MT4 bots (also called expert advisors) and somehow overcome those difficulties please share how to do that.
Thanks for the very insightful shares. Haven't had time to look through them all, but looks great so far!
Btw, You should use TradingView.com to chart & share your trades!
Btw, You should use TradingView.com to chart & share your trades!
@Goomboo, a few questions to you:
- Your average win/lose ratio? (number of trades won/number of trades lost)?
- Your average risk: reward ratio? (when you win, it is how many times as much as a loser trade?)
- In general how many trades do you take per day/month/or week? I don't know if it is called high frequency or not
- Your average win/lose ratio? (number of trades won/number of trades lost)?
- Your average risk: reward ratio? (when you win, it is how many times as much as a loser trade?)
- In general how many trades do you take per day/month/or week? I don't know if it is called high frequency or not
Greetings and welcome to the thread,
- On average, I win 32% of all trades
- My average losing trade results in a .79% loss of account equity
- I do not set profit targets, but my best winners have historically been 5-20 times what I'm willing to risk on the trade
- In general I target 1-3 trades per week and I'll hold positions as long as the trend remains. Losing trades are cut within a few hours and winning trades last days, weeks, or months
- I don't know anything about the bots, sorry :/
- I primarily trade the currency markets
My advice to someone wanting to get into this business but not willing to put in the work necessary to beat the competition - don't. The S&P 500 historically returns around 11.96% per year - in the long run, the casual participant is significantly better off by buying an index ETF (US: SPY is good) and focusing on the day job. 66% of currency traders consistently lose money in every single quarter. The odds don't favor speculative traders.
This is my personal opinion - I'm not a registered financial advisor. Best of luck.
I've read both, that Gekko is and isn't compatible with BTC-E.
Do you know if it is compatible?
Do you know if it is compatible?
True: https://github.com/askmike/gekko/blob/master/README.md
Jump to: