/biz/ - Business & Finance

>>21424568
I sat on cash all throughout 2018 until I saw this.

-to be continued.

>>21424568
>>21424895
I cashed out here, because the parabolic move needed a retest.

>>21424568
TA is expected value + groupthink/self-fulfilling prophecy + mean reversion/stat arb

>>21424568
Patterns are legitimate but weak indicators and need to have compounding confluence to mean anything. TA is more than just that.
And yes, it's extremely easy to fall prey to confirmation bias. When in doubt flip the graph upside down and see if you still agree with yourself.

>>21424985
But it's not in any way more accurate than choosing heads or tails. Unless you were to simply buy indices with the belief that they would go up, which they do in the US market.

>>21424985
this

>>21425081
Patterns only seem to be legitimate when there is a general trend. Therefore the pattern that forms is irrelevant to the direction. Particularly now that we had a giant sell off in March. Prices really had no where to go but up, so if you saw a bull flag there was a good chance your belief would be confirmed. But what about when there is a long term crab market like Japan?

as long as you try to recognize the bias.

i've been getting into trendspider, but it's expensive and they don't have a crypto-only version. regardless, i like being able to remove the emotion. sometimes i find my trendlines will move in whichever direction i FEEL like. if im in a buying mood, then ill come up with any reason to buy, etc. just pay attention to your bias without second guessing yourself. crypto isn't as predictable as common stock.

a lot of people argue that lagging indicators are useless in crypto, but i think RSI is important. i like watching it against the price depth. everyone should have their own style. whatever works.

>>21424568
bots will follow the patterns more than humans
so its two conversations really

>>21425187
>prices had nowhere to go but up
Not really. Check BTC Jack on Twitter, the dude publicly shorted 10.5k at the time with a 5k goal like a legend and posted his technical reasoning for why that dump had to happen while simultaneously being bullish on the overall trend and being long since then.

>>21425331

*** order depth

i meant order depth

>>21425331
I'm mostly talking about stocks, but the same holds true for crypto. And if going by feeling is just as valid as actual charting and study, then the system is no more than chance.

>>21425358
Bots will follow the patterns, but are the patterns worth following? That is the question. Overall it doesn't really seem so.

>>21425408
I'm not referring to crypto. And that could simply be a case of survivor bias.

>>21424568
Algo has been doing well, but TA only gives you a statistical advantage.

>>21425408
Are you trying to summon Satan over there?

>>21424568
TA is self-fulfilled by bots that use it to inform behaviour based on order books and arbitrage data. This only varies a bit these days because some bots are becoming sophisticated enough to predict the behaviour of other bots and exploit them. Years ago, it wasn't even this complicated.

Sometimes a discord raid will rock the boat, but only in small caps.

>>21425121
Yes it is. Get into it and be prepared to start throwing up in your mouth. It's better than 50%.

>>21425516
He had a following prior to the prediction and there's only so many times he can get it right before you have to admit TA probably is on to something.

>>21425408
No, he's likening crypto to wild rabbit population growth and hoping that enough other people and bots do it that he ends up being right. If enough people use TA, they form a conspiracy that moves the markets around until someone starts doing it faster or has enough liquidity for a contrarian play to catch them.

TA is a self fulfilling prophecy.

>>21425644
I have been in it. You're just as well served as buying into a directional trend. It works until it doesn't. Being in a rallying bull market for both traditional markets and crypto is giving even stronger biases toward long positions. If you set the timeline further out then longs win every time in the overall stock market. Crypto is to be determined.

>>21425121
You have absolutely no idea what TA is. There are 2 groups of people in existence with literally zero overlap. People who understand TA, and people who think TA does not work. It is not possible for a single human to fall into both categories.

>>21425762
Ok. Have 100 people trade for 10 years and beat an average of 9% return per year.

>>21425943
Trading is a zero-sum game. You aren't going to get good average return or even a positive median, but you're going to get a Pareto distribution that's decently consistent over the years.

>>21426038
Eventually all of them would be wiped out by attrition if they chose not to exit.