The only reasonable and even proven correct point of applying deep-learning is that the algorithm “learns” so-called “hidden patterns”, which escape (literally) from a biologically limited human intelligence.

The causality is so complex that even to list correctly the major factors (not missing anything relevant) is nearly an impossible task.

Social media creates whole waves of bullshit, including the idiotic memes about the US interest rates, and other poorly understood and dynamic macro-economic factors.

The common patterns which constantly re-emerge on the charts (of different timeframes), including the major indicators (such as RSI) create “waves” through social media.

But still, the major driving force that moves the prices is FOMOing in by degens, and the major winning strategy for institutions is (still) “trading against retail”.

They (the institutions) are constantly becoming algorithmically and technologically better at it. Look what this narcissistic amphetamine addict SBF did (running billions on some crappy untyped Python, lmao).

Wishful thinking

When we observer some emergent patterns on the chart we are usually getting lost in our wishful thinking, what has to happen next. This is, of course, bullshit.

The right edge

Everyone is a good trader at the middle of the chart of some historic data - here is where to buy, and there is the selling point. The problem is - we have to trade on “the right edge”.

Not tring to be right

When you observe that stupid people are doing stupid things (buying SHIB lets say), do not try to be rational and avoid to be triggered - just ride the wave, just like surfers do. Never argue with the ocean.

You do not have to “prove” anything or even to “be right about it”. Ride on the momentum they are creating by FOMOing. This is exactly how professional traders make money - they never argue with the crowd (or the market), they just freeride it while it lasts, and then keep looking for another ride.

Sort of hitchhiking the waves.

Amateurs

Amateurs are usually not aware of anything, except the current price, the last very few candles and their strong, overwhelming emotions about it. Not of a current range, “market”, phase and how the other people persieve the same charts - what is the current sentiment.

No wonder they literally have no idea what to expect, where “we” are in a cycle, within the current market condition (top, bottom), and what caused it (memes and FOMOing in by the degens).

So, they over-trade, made lots of mistakes, small and big, lost all the money and got washed out from the market (with an empty trading account). At least this is how I did it.

Professionals know “time in the market” (what to expect), or they think they know.

Greater fools

All the speculative trading is about dumping on a greater fool - someone who would buy your bags (at a higher price, or just buying the “top”). Everything else is mostly a marketing bullshit. Not just that, but every single degen is thinking that he will time it perfectly.

Psychology

Markets are driven by overwhelming emotions of market participants (including those on a sideline) and the price is being moved according to human psychology.

Even bots have been programmed with some implicit human assumptions about the markets.

Irrational exuberance

The biggest finding of the last century is that there is no assumed rationality, but emotions, memes and a crowd (a flock) behavior. They are all FOMOing in or they are all panic selling.

This is not just another theory. The fucking SHIB or DOGE or LUNC created a lot of dollar millionaires. Some shitcoins actually went 100x or even 1000x on high volume right before our eyes (we just were not aware of what we are actually observing, only some experienced traders were).

A person who knew “what is actualy going on” (FOMOing in by literal degens) and “what to expect” (a massive dump back to zero) could, at least in priciple, ride the both rides - one up, and one back down (clack, clack, you know).

Smart people at least would leverage-short all the meme shitcoins (ideally, with other people’s money).

What will NOT work (in principle)

Predictions

Statistical “inference”

Abstract models based on theories

Information loss

When we sum-up (and averaging) into a bigger “candle” we lose information.

A day-candle, for example, loses all the intraday movements, including activities from the major timezones (US West Coast, Asia, Europe).

Capturing information

Some events could be observed and properly captured only on daily chart for a multiple years timeframe.

8-hour candles would, probably, do it too, or even some custom 6-hour ones (better reflecting the major timezones).

Regular events

  • The US macro economy monthly data releases.
  • Quarterly earnings reports by the major corporations.
  • etc.

Irregular major events

  • FED interest rate decisions (the actual changes)
  • Musk’s tweets about Dogecoin
  • etc.

The best we can hope for

The whole point is to make informed decisions, to augment an intuition with the actual supporting (or disproving) data.

A grossly over-simpilified scenario is to identify a major FOMOing into some meme shitcoin and be prepared to short the top.

Shorting overheated shitcoins is costly (due to constantly being stopped-out by another a smaller and smaller waves of FOMOing).

But once a position is secured, it could print all the way down through /lower lows.

The level of an entry may (with the high probability) never be touched again (at least until the next bull market).

We exit as the market bottomed-out (due to “funding rates” and other slow bleeding).