Everything starts with this question, because if it had an answer nothing else would matter. So it has to be asked first, and asked seriously.
The answer is no
The direction of the next candle, from past price alone, comes out indistinguishable from a coin toss. Not “weak”, not “needs more data” — level with chance, on every instrument and every timeframe tried.
The deepest reason is the simplest: if direction were predictable from past price, it would have been predicted away long ago. This is a market with thousands of clever people and faster computers than yours looking at exactly that data.
But being right half the time breaks even, surely?
No, and this is the step nearly everyone skips. There are costs, and costs do not shrink with the timeframe. The size of a bar does.
Median move of one bar, measured on real closes, against a 4 basis point round trip — generous for a liquid instrument.
At one minute, BTC’s typical bar is smaller than the cost of trading it: you would need to be right 99.9% of the time to break even. On the daily, 51.4%. Same skill, two universes.
Put the two together: the shorter the frame the more it demands, while what you actually have stays pinned near 50%. On the left of that chart the flat line never reaches the curve.
Why this is particularly cruel to a private trader
Look at what gets sold to you. One-minute scalping courses. Five-minute signal groups. Every broker nudges you shorter, because shorter means more trades, and they earn on each one.
In other words: you are pushed onto the left half of that chart — where the break-even bar sits at a height nobody clears.
A fund does not stand there. It stands on the right, or it makes money in a way you cannot copy at all — through size, through cable latency, through management fees. None of that is available to you. The right half of the chart is.
This is not a dead end
The no is a sieve. It removes everything that depends on knowing direction — and leaves exactly the things that do not. What is left is the next question.