Momentum
Recent direction tends to carry on, for a while. The same idea speeds up neural-network training, drives a family of trading indicators, shapes how traders add to winners and fuels the crowd. Follow it, and see where it stops paying.
-
ML Math
Optimizers
Momentum in training: keep part of the last step's direction, so the optimiser rolls through small bumps instead of stalling at each one.
-
Technical Indicators
Exponential Moving Average
The same running memory applied to prices: an exponential average weights recent moves most, so it turns with the trend sooner than a simple average.
-
Technical Indicators
MACD
The gap between a fast and a slow average measures momentum directly. It widens when a move accelerates and narrows as it fades.
-
Technical Indicators
Rate of Change
Momentum at its plainest: the percentage change over N bars. A rising rate of change says the move is speeding up.
-
Risk & Portfolio
Pyramiding
Traders act on momentum by adding to positions that are working, in smaller tiers as the trend gets older.
-
Market Psychology
FOMO
The crowd's version: rising prices pull in buyers who fear missing out, and their buying carries the move further, until it stops.
-
Pattern Essays
The Random Walk
The sceptic's stop. If prices were a pure random walk, past direction would say nothing about the next step. Momentum strategies bet that they are not, and the evidence is mixed.
Momentum is memory: of the gradient, of the price, of the crowd. It helps while the direction holds and hurts at the turn, which is why every version comes with a way to let go.
Add up enough small, independent nudges and the same curve appears — in sample averages, in price bands, in risk models. Follow it across four fields, then see where it quietly stops being true.
Extreme results tend to be followed by less extreme ones. It looks like a force pulling things back; it is arithmetic. Watch the same idea turn into trading signals, portfolio rules and model penalties.
Every field has the same problem: real structure mixed with randomness. These stops are the tools for pulling them apart — and for checking you have not invented a pattern that was never there.
When a system's output feeds back into its input, small nudges can snowball. Crowds, prices, volatility and even deployed models run on loops like this — and they tend to stay quiet until they don't.
The data you can see is often the data that survived. Patterns found in it can be an artefact of what went missing. These stops show how the trap works, and the habits that keep you out of it.
Many systems change all at once when a line is crossed. A neuron fires, a classifier says yes, a language model drops its unlikely words, a price breaks out, a stop-loss sells. Each stop is a threshold, and each has to answer the same question: where to draw the line.