What cross-sectional momentum is (and why it actually has an edge)
When someone says “momentum” they usually picture buying whatever is going up. That is time-series momentum: you look at one asset over time and ride its trend. Cross-sectional momentum is something else, and it is where a measurable edge actually shows up.
Compare, do not chase
Cross-sectional means comparing many assets against each other at the same moment. Instead of asking “is this coin going up?”, you ask “which ones are going up more and which less than the rest RIGHT NOW?”. You go long the strongest in relative terms and short the weakest. The bet is not on the market, it is on the difference between one and another.
Why that helps
By going long and short at the same time, you neutralize a good part of the broad market move: if everything falls together, your shorts cushion your longs. What stays exposed is the dispersion between the strong and the weak, which is exactly where the edge lives. It is one of the most documented anomalies in the academic literature, in stocks and in crypto, precisely because it is hard to arbitrage away completely.
Why it survives where classic technical analysis does not
Most candlestick patterns, indicators and “secret systems” do not survive an honest out-of-sample validation: they are overfitting. Cross-sectional momentum, properly built, does leave a residue of edge after costs and out-of-sample. Not because it is magic, but because it captures a real and persistent behavior of the markets.
It is not magic, and we say so
The edge is modest and decays: you have to measure it over short horizons, control volatility (vol-target), filter by liquidity and revalidate it continuously. It will not make you rich overnight, and it will have bad runs. But it is real, and it is verifiable, which is more than almost everything else offers.
This is the strategy family that survived our validation, and you can see the walk-forward study and every closed trade, win or lose.