I keep seeing too many fancy looking charts with arrows and colourful lines and commentaries that would make you believe that you can compute where the price of bitcoin or any other cryptocurrency will move in the next (add whatever time period: 5min, 5hrs, 5days). Unfortunately, it's far from that. Read this and save yourself a lot of trouble and potentially money.
The idea of technical analysis
..is that you can look at historical trends, find patterns that repeat themselves and deduct future potential moves from it. It all sounds good on paper.
But guess what?
It's a 50/50 chance. A chart will either go up or down. You won't see a flat line.
So guess what nr. 2?
It's not that hard to get it right a few times. It's also not that hard to get it wrong a few times. It's a 50/50 chance. No amount of fancy charts and algorithm will hide that.
The problem with technical analysis of cryptocurrencies
Bitcoin, ethereum, litecoin or any other cryptocurrency do not move because of standard stock market reasons. They move because of a very wide range of reasons, mostly non-standard and external to any data that an algorithm can analyse.
Price movements are demand based but also, to a high level event & news based - good news, bad news, governments supporting or blocking the development, Vitalik getting married, false news, truthful news.
Unexpected development has a strong impact too - new companies entering the market, new technology, existing companies launching something new or doing big volume transactions.
Any news that touch this topic and have a wide reach will impact the behaviour of buyers and sellers and also the price.
Now show me an algorithm or a pattern recognition that can account for this? That can predict all these random developments?
Yep, you're right - there is none.
No algorithm can predict what South Korean government, Charlie Lee or any other key developer, US government or a hacker will do. And this is what often drives cryptocurrency prices. So stop lying to yourself with technical analysis.
Pattern recognition is very useful in daily life - we us it everywhere. It's also good to analyse big sets of data and draw conclusions. But it's useless in new areas that are prone to change because of factors from outside of their data set - like cryptocurrency trading. The dataset is the market data. But no algorithm has access to, nor knows how to analyse news, true or false, scandals, unexpected changes or random human herd behaviour.
No cup and handle, heads and shoulders or X amount of tops and bottoms shape can tell you what the world outside will do. Same as bottom of your coffee cup patterns or the patterns of your palm won't tell you what will happen in the future.
Confirmation bias, Texas sharpshooter fallacy etc.
- aside from not being able to analyse external factors that play a huge role in current crypto prices, human biases and fallacies do so too
- confirmation bias - you look at a chart today and guesstimate that it will go up tomorrow. Tomorrow comes, it does - Yay!, you were right, aren't you good. This happens 50 out of 100 times. But you tell yourself and everyone else, that it happens most of the time - because you comfortably ignore the many times it didn't work out. "Because there are other, external factors, and I can't be always right.". Well, truth is, you are not - most of the time. You might as well flip a coin to make your crypto buy/sell decisions for you.
- texas sharpshooter fallacy - because you guessed the price move right, you assume that your analysis is right. But the price moved in the direction you guessed not because the exact assumptions that your technical analysis used became reality. It moved there because of a million other reasons that you didn't even think about. But you give yourself a pat on the back - well done, you analysed it right.
it's just like shooting randomly at a board with no target. Then looking where did the bullets group the most and then you put the target there, claiming you hit the bull's eye many times.
same thing happens with technical analysis - you make a bunch of predictions, some work, some don't - but you discount the ones that don't and tell everyone about the ones that did. You hit the target.
Therefore - save ourself a lot of trouble (and money, and grey hair) and forget about technical analysis. Ignore all the fancy looking charts that people like to post around forums.
What to do instead?
- you can try your luck with active trading, but know that's much more stress and risk than slower forms of investing (buy & hold)
- or you can buy on dips and hold long (6months+). It's an exciting market and technology that is still in very early stages - therefore it will more likely grow in the long term, than fall down. The better the technology you choose, the more problems it solves, the better the team behind it, the better the chance of future growth
buy many small, new ICO's (tokens) and coins - many will fall, some will grow by a big percentage - those will grow and cover your losses too
What is your opinion? Do you share it with me or is your experience different? Let me know in the comments!