Throughout the course of this thread, we have discussed the importance of creating, testing, and using discipline to execute your strategy. In this post, I will detail a key component of trading systems, which has received very little attention.
For your trading system to be a success, it really has to make sense. If you are unable to describe your method on the back of a business card to the average person in less than 30 seconds, your system is probably nonsense. Take for example the following situation in which you are asked to explain your system to a potential investor:
Situation 1 – The Moving Average Crossover
“So, what is your trading method?”
“I trade a moving average crossover. Basically, I have a theory that price increases and decreases in relatively predictable trends, so I try to be long when price is increasing on average and be short when price is decreasing on average. I have found that this method allows me to almost always be positioned with the trend. I have backtested it, and it has proven profitable across nearly every financial market for the past 150 years.”
Situation 2 – The Fibonacci Gann Master
“So, what is your trading method?”
“I trade Fibonacci retracements, expansions and Gann angles. I have a theory that there is a hidden order to the markets and price moves in the same mathematical pattern that a few things in nature exhibit. Basically, it’s super complex but don’t worry, I’ve read a book or two. If you’re interested, I’ll allow you to pay me for my analysis…but be warned, I am not actually making money using it. Also, I haven’t actually tested it, so…”
Where would you entrust a $100,000,000 investment?
Throughout my life, I have found that the true masters of fields are those who are able to express advanced concepts in terms that the common person can understand. Trading is no different. Without fail, the most successful traders I have met have been those who are able to take their decades of experience and logically present their method in less than 30 seconds to the listening ear.
The serious trader should ask himself if his method actually makes sense.
Topic: Goomboo's Journal - page 28. (Read 281459 times)
I don't agree with you. I believe that people overstate their ability to see the "effects" of developments in markets. Having Bloomberg and CNBC on the trade floor makes you pretty skeptical of "cause and effect" in the markets. "Markets are down 2% today on fears of Europe". Really? The reporters know the collective reason why hundreds of thousands of individuals bought and sold? Even better is when the market ends the day positive following a loss: "Markets are up 1% today on European optimism". Nonsense. I've even seen the Bloomberg terminal spit two news articles in the same day telling me why the markets moved up and down, each just a rephrasing of the same news. It's human tendency to assign a cause to something and when it comes to trying to say why trillions of dollars and hundreds of thousands of people collectively behaved a particular way is just silly. It's human arrogance at its finest: "I know why it happened, I am smart and in control."
I was referring to the events which actually have the potential to move the market. Specifically the cases of bitcoin Reddit or wordpress accepting bitcoin will have a large affect on the user base and transactions over time, not only because of their effect individually, but in this case because it set a precedent for others to follow. I am talking long term fundamentals, not some BS that a TV station tries to cook up to maintain ratings.
News is perceived differently depending on the juncture of the market. It is not the cause, only the trigger.
Also, some of those things you listed did nothing, or they were reversed immediately. An example is the DHS/Dwolla news and the VCs/the conference. You could create a bigger list of events that ended up in that category.
Also, some of those things you listed did nothing, or they were reversed immediately. An example is the DHS/Dwolla news and the VCs/the conference. You could create a bigger list of events that ended up in that category.
In my short journey with the bitcoin market, I've noticed two trends with 'news'. There is the kneejerk reaction to the news, and the reality of the news. The reality sets in after the immediate impact and often has no bearing in the medium - long term. The kneejerk reaction usually impacts only short term expectation unless the reaction was rooted in reality. Both impact the savvy trader in their own way. The buy & hold group can optimize their position as a result, but are not likely to be impacted too greatly.
I like the maths mixed with a little social confidence or feeling . Realistically, I'm likely just fooling myself with both.
Goomboo, this is one of the premier threads in the forum. Thank you!
TL;DR: http://xkcd.com/904/
The amount that they move the price is not necessarily quantifiable, its rather qualitative in nature. Compounded one can see the effect these events have on the price overall.
I don't agree with you. I believe that people overstate their ability to see the "effects" of developments in markets. Having Bloomberg and CNBC on the trade floor makes you pretty skeptical of "cause and effect" in the markets. "Markets are down 2% today on fears of Europe". Really? The reporters know the collective reason why hundreds of thousands of individuals bought and sold? Even better is when the market ends the day positive following a loss: "Markets are up 1% today on European optimism". Nonsense. I've even seen the Bloomberg terminal spit two news articles in the same day telling me why the markets moved up and down, each just a rephrasing of the same news. It's human tendency to assign a cause to something and when it comes to trying to say why trillions of dollars and hundreds of thousands of people collectively behaved a particular way is just silly. It's human arrogance at its finest: "I know why it happened, I am smart and in control."
As I have said before, traders are not paid to know why something happened: they are paid to make money.
Many venture capitalists do not use hard data, they use their gut, which is based on knowledge and experience. There is often data behind their feeling, but they do not analyze it in detail before acting - they act as soon as they know the moment is right (or more often than not for the less successful of them after the moment has passed because they either did not trust themselves to act without data or did not see the opportunity).
Back in grad school I took an entrepreneurial class and a VC firm talked about funding new ventures. He basically expected 70-80% of investments to lose everything, 5-10% to double his money, and 1-5% to return 10-100 times investment. This sounded an awful lot like trading to me - managing risk and only approaching the highest probability opportunities. At the core, pure investing and trading is basically after the same thing, they just pursue it differently.
(as you pointed out, investing to date has actually been more profitable than your system when utilized perfectly)
Buy and hold hasn't "made" more than traders of the moving average crossover. In fact, buy and hold has "made" nothing. If they sat throughout the bubble and the subsequent crash and are still sitting, chances are, they will "make" nothing. You have only made money when a trade is closed and you actually have profit in the bank. If you are content to watch the value of your account fall by 81% in 7 days, you will be content to watch it dwindle to zero.
We see the world through different lenses. Best of luck with your investment.
If you made an automated trade every time the EMA 10 and EMA 21 lines crossed during the current flat period, you would very likely lose money due to the high 0.25-0.6% MtGox fees. So what's the best strategy to mitigate this effect of a very flat market?
1) Turn bot on/off when the market is flat, but you'd miss the beginning of the next volatile spike due to some new report.
2) Instead of trading exactly when the 10/21 lines cross, build some hysteresis in, for instance the two averages must be at least 0.6% apart - but you'd lag getting into or out of any large price swings, decreasing profitability.
3) GoomBoo said he's only trading daily. Perhaps at that timescale there are fewer false-positive triggers?
I feel that the current flat market is very different than the high volatility over the previous months.
1) Turn bot on/off when the market is flat, but you'd miss the beginning of the next volatile spike due to some new report.
2) Instead of trading exactly when the 10/21 lines cross, build some hysteresis in, for instance the two averages must be at least 0.6% apart - but you'd lag getting into or out of any large price swings, decreasing profitability.
3) GoomBoo said he's only trading daily. Perhaps at that timescale there are fewer false-positive triggers?
I feel that the current flat market is very different than the high volatility over the previous months.
Thank you for the questions, I'm glad you've learned from the thread. Here's my thoughts on your questions:
- If you are being chopped up (trading too much / too many false signals), I suggest moving to a higher timeframe. As you mentioned, I am trading daily for this reason. The large spread (difference between bid and ask) and large commission (MtGox fee) reward strategies which don't engage in too much trading. The actions of trading (crossing the bid-ask spread and paying commissions) really add up at the end of the day.
- I can't speak about the trading bot you mentioned (I wasn't involved in its creation or maintenance), but if the 10/21 is too active, I suggest lengthening the period of moving averages used (provided you have backtested the combination and found it to be within your risk tolerance).
- Waiting until some threshold is reached is a popular method of improving moving average crossover systems. There are hundreds of variants of the original strategy and I definitely encourage you to test these concepts.
One indicator is Google Trends for "bitcoin". Previously it was a good correlation. Though it recently fell 80% without a proportional fall in BTC price.
Two words: Lagging Indicator
Any lagging indicator doesn't indicate anything but the past.
One indicator is Google Trends for "bitcoin". Previously it was a good correlation. Though it recently fell 80% without a proportional fall in BTC price.
These indicators do in fact often move the price in a stable way. The reason we have gone from 13 to 120 is not because of trading, but because of fundamental differences in the infrastructure and user/ business adoption of bitcoin.
Which indicators? How often do they move the price? What is stability? How do you know that infrastructure drove the price up 1000%? If you don't have a ready answer with hard numbers to these questions then you are relying on unsubstantiated speculation.
As discussed in this thread, professional-grade technical analysis creates, tests, and trades a set of technical rules. Professional-grade fundamental analysis is no different. If you want to talk about serious fundamental analysis, you have to quantify and test it as well. At its core, fundamental analysis is about the study of supply and demand and the majority of market participants fail to do just that...study supply and demand.
Here's some examples of a good place to start with legitimate fundamental analysis:
- "Demand for the commodity is 10% above the 1-year average and supply is 5% below the 1-year average - historically, this leads to price increasing by an average of 20% over the next week 70% of the time"
- "Imports are down 15% year over year while harvests are up 15% over the same time period - historically, this leads to price decreasing by an average of 15% over the next month 65% of the time"
- "Money managers added to their long positions by 10% while supply has dropped by 20% - historically, this leads to price increasing by an average of 15% over the next year 85% of the time"
Can't you see how there is a distinct difference between the nonsense spread on this forum and the above examples? This is fact telling rather than story telling. In each of the above, supply, demand, and the price response have been fully quantified and measured. Not only has the method been backtested, but it also makes intuitive sense. If you want to make money from trading fundamental analysis, you have to tell facts, not stories.
Your post should have read "these indicators do in story move price..."
The reason we have gone from 13 to 120 is not because of trading
Are you sure? Can you prove this? How sure are you that it isn't me and some of my trading partners systematically extracting profits from the market? You cannot prove it and it is very bold to say "the reason" for anything. Above all, traders must remain skeptical and objective - not engaged in story telling.
This is precisely why I caution everyone against asking "why". It's a deadly question and breeds a false understanding. Unless you are a major player in the market, you truly have no idea why something happened. I'm not paid to know why something happened, I'm paid to make money.
The absolute lack of any tangible fundamental information that influences the price of BTC means that you should seriously consider not asking or trying to answer why, what, or how prices move.
I'm surprised no one has challenged me on this comment yet.
This means that all of the "China is coming online...BUY", "Gold is down...BTC UP!", "The fair value of BTC is X" threads are absolutely worthless. It's pure unsubstantiated speculation which deludes and confuses the participants. If you are relying on Chinese buying, flight from gold, or a magic "fair value" to emerge, you are simply investing. You have no science, no method, and no plan. You're holding onto hope and hope is one of worst enemies of the objective trader.
FTFY
Goomboo, I largely agree with everything you say in this thread regarding trading, but please dont discount the value of fundamental analysis when it comes to investing (as you put it). These indicators do in fact often move the price in a stable way. The reason we have gone from 13 to 120 is not because of trading, but because of fundamental differences in the infrastructure and user/ business adoption of bitcoin.
Just read through the GoomBoo Journal over the last two days after hearing about it from one of the trading bot threads https://bitcointalksearch.org/topic/chrome-browser-extension-mtgox-trading-bot-67591 Thanks very much, GoomBoo, for posting this advice and answering everyone's questions.
However, 'bomtis' above highlighted something that I'm curious about anyone's thoughts on:
If you made an automated trade every time the EMA 10 and EMA 21 lines crossed during the current flat period, you would very likely lose money due to the high 0.25-0.6% MtGox fees. So what's the best strategy to mitigate this effect of a very flat market?
1) Turn bot on/off when the market is flat, but you'd miss the beginning of the next volatile spike due to some new report.
2) Instead of trading exactly when the 10/21 lines cross, build some hysteresis in, for instance the two averages must be at least 0.6% apart - but you'd lag getting into or out of any large price swings, decreasing profitability.
3) GoomBoo said he's only trading daily. Perhaps at that timescale there are fewer false-positive triggers?
I feel that the current flat market is very different than the high volatility over the previous months.
However, 'bomtis' above highlighted something that I'm curious about anyone's thoughts on:
2) what triggers the trends? it seems that since i uploaded fiat into my account ( nothing more then i can withstand to lose ) the market has come to a stop. Prices dabble around $122 but nowhere near anything profitable
If you made an automated trade every time the EMA 10 and EMA 21 lines crossed during the current flat period, you would very likely lose money due to the high 0.25-0.6% MtGox fees. So what's the best strategy to mitigate this effect of a very flat market?
1) Turn bot on/off when the market is flat, but you'd miss the beginning of the next volatile spike due to some new report.
2) Instead of trading exactly when the 10/21 lines cross, build some hysteresis in, for instance the two averages must be at least 0.6% apart - but you'd lag getting into or out of any large price swings, decreasing profitability.
3) GoomBoo said he's only trading daily. Perhaps at that timescale there are fewer false-positive triggers?
I feel that the current flat market is very different than the high volatility over the previous months.
News is perceived differently depending on the juncture of the market. It is not the cause, only the trigger.
Also, some of those things you listed did nothing, or they were reversed immediately. An example is the DHS/Dwolla news and the VCs/the conference. You could create a bigger list of events that ended up in that category.
Also, some of those things you listed did nothing, or they were reversed immediately. An example is the DHS/Dwolla news and the VCs/the conference. You could create a bigger list of events that ended up in that category.
The only news that really triggered a significant move and attracted an important quantity of new money was "Bitcoin broke ATH and is going up up up"
Everything else, including the Cyprus situation which seems so important for everybody, didn't do shit in the grand scheme of things, apart from affecting super short term the price because of speculators already in this game overreact to every Bitcoin related press release.
News is perceived differently depending on the juncture of the market. It is not the cause, only the trigger.
Also, some of those things you listed did nothing, or they were reversed immediately. An example is the DHS/Dwolla news and the VCs/the conference. You could create a bigger list of events that ended up in that category.
Also, some of those things you listed did nothing, or they were reversed immediately. An example is the DHS/Dwolla news and the VCs/the conference. You could create a bigger list of events that ended up in that category.
The absolute lack of any tangible fundamental information that influences the price of BTC means that you should seriously consider not asking or trying to answer why, what, or how prices move.
I'm surprised no one has challenged me on this comment yet.
This means that all of the "China is coming online...BUY", "Gold is down...BTC UP!", "The fair value of BTC is X" threads are absolutely worthless. It's pure unsubstantiated speculation which deludes and confuses the participants. If you are relying on Chinese buying, flight from gold, or a magic "fair value" to emerge, you are simply gambling. You have no science, no method, and no plan. You're holding onto hope and hope is one of worst enemies of the objective trader.
QFT
"Absolute lack of tangible fundamental information influencing the price of BTC"
Beep! Beep! This just in:
• ...Pirate is closing shop...
• ...GLBSE is closing shop...
• ...Blockchain reward halving...
• ...Wordpress accepting Bitcoin...
• ...Reddit accepting Bitcoin...
• ...Cyprus on the brink...
• ...SilkRoad under sustained DOS, future uncertain...
• ...DHS seizing MtGox Dwolla account...
• ...Union Square Ventures...
• ...Founders Fund ...
I fully agree. Fundamental analysis is not useful with Bitcoin.
I wouldn't go that far.
Short term fundamental analysis is not useful with Bitcoin.
The absolute lack of any tangible fundamental information that influences the price of BTC means that you should seriously consider not asking or trying to answer why, what, or how prices move.
I'm surprised no one has challenged me on this comment yet.
This means that all of the "China is coming online...BUY", "Gold is down...BTC UP!", "The fair value of BTC is X" threads are absolutely worthless. It's pure unsubstantiated speculation which deludes and confuses the participants. If you are relying on Chinese buying, flight from gold, or a magic "fair value" to emerge, you are simply gambling. You have no science, no method, and no plan. You're holding onto hope and hope is one of worst enemies of the objective trader.
QFT
I fully agree. Fundamental analysis is not useful with Bitcoin.
The absolute lack of any tangible fundamental information that influences the price of BTC means that you should seriously consider not asking or trying to answer why, what, or how prices move.
I'm surprised no one has challenged me on this comment yet.
This means that all of the "China is coming online...BUY", "Gold is down...BTC UP!", "The fair value of BTC is X" threads are absolutely worthless. It's pure unsubstantiated speculation which deludes and confuses the participants. If you are relying on Chinese buying, flight from gold, or a magic "fair value" to emerge, you are simply gambling. You have no science, no method, and no plan. You're holding onto hope and hope is one of worst enemies of the objective trader.
1) are there any other trading strategies that actually work? i can't think about any other methods that would make a profit as safe as the one described here and all the strategies i think of are based on emotions ( i _believe_ the price will go up )
There are virtually unlimited trading strategies that work. The purpose of this thread was to show how even the simplest trading strategies yield profit if applied with discipline and risk control.
Some advice for making a strategy would be to write the following on a piece of paper:
- Entry - when will you enter the market
- Exit - when will you exit the market
- Trade size - how much you will trade
- Test your strategy using historic data
The most important step here is testing. You'll find that if you haven't adequately quantified your methodology, you will be unable to test using historic data. Unfortunately, the vast majority of market participants are not willing to put forth the effort to actually backtest their theories.
2) what triggers the trends? it seems that since i uploaded fiat into my account ( nothing more then i can withstand to lose ) the market has come to a stop. Prices dabble around $122 but nowhere near anything profitable
"What" and "why" are surprisingly deceptive questions. If your goal is to be a technical trader who relies on objective analysis, these questions are deadly. If you are seeking to be a fundamental trader, these are important questions, with one very important caveat: you must test how the market has historically behaved under similar conditions.
The absolute lack of any tangible fundamental information that influences the price of BTC means that you should seriously consider not asking or trying to answer why, what, or how prices move.
The key to successful trading is what I've repeated throughout this entire thread: create, validate, and trade a method with tight control of risk and emotion.
Jump to: