Momentum Trading — Catching the Market's Strongest Moves

Momentum is an observation so simple it's almost suspicious: what's been rising strongly statistically tends to keep rising — at least for a while. In academic finance, it's one of the most thoroughly studied market anomalies, documented since the 1990s across dozens of markets. In retail trading — it's one of the most overused words, thrown into every course alongside "smart money."

This article separates the two: what the measurements actually say about momentum (including on BTC), how to concretely measure the strength of a move, how to turn that into rules — and when buying strength turns into buying the top.

What Momentum Trading Actually Is

The idea: the market has inertia. Price moves aren't fully random — information spreads gradually, funds build positions over weeks, and the crowd joins with a lag. The result: a strong move attracts more capital and sustains itself, until the fuel runs out.

Momentum is played two ways:

• Time-series momentum: you look at a single market — if its own return over the last N periods is positive/strong, you're long; if it's weak, you stand aside or go short. This is the approach closest to trend following.
• Cross-sectional momentum: you compare many assets and buy the relatively strongest ones (e.g., the top 10% ranked by 3-month return), skipping the laggards. This is how classic academic momentum works, and it's a viable approach to a basket of alts against BTC.

There's also an intraday variant — momentum day trading, playing the day's strongest moves on volume. This is the most demanding version: algorithmic competition, costs and slippage eat up most of the theoretical edge, as we discuss honestly in the article on day trading.

What the Numbers Say

Momentum on BTC: according to cited QuantifiedStrategies backtests, Bitcoin has historically been a distinctly momentum-driven market — simple trend-momentum strategies on the daily timeframe (buy strength, exit on loss of it) were profitable over multi-year data, unlike many contrarian strategies, which performed poorly on BTC. That fits the intuition: crypto is a market of narratives and retail flows, and markets like that tend to trend.

Classic momentum (stocks): dozens of academic studies since the 1990s have documented a momentum premium of several to a dozen or so percent a year on "buy winners, sell losers" portfolios over a 3–12-month horizon. It's one of the few anomalies that survived publication — though with periodic, painful crashes (so-called momentum crashes), usually during sharp market reversals.

Why does this even work, and why doesn't it disappear? The most convincing explanations are behavioral: investors underreact to new information (price reaches its new value gradually, not instantly), then herd into a move that's already visible, and mental accounting makes them sell winners too early and hold losers too long. These are biases baked into human nature, so they're hard to arbitrage away to zero — which explains why momentum has survived three decades since the first publications. But watch the flip side: the same herding means that when the crowd reverses, it reverses all at once — hence momentum strategies' characteristic, sharp crashes during market turns.

Caveats without which these numbers would just be marketing: the BTC results come from backtest summaries — verify them at the source and run your own numbers before using them. The momentum premium can be "thin" after transaction costs, and its crashes arrive suddenly. And as always: past results don't guarantee future ones.

How to Measure the Strength of a Move — the Tools

A comparison of the most popular momentum indicators (conclusions consistent with LuxAlgo's overview):

Practical rule: one momentum indicator plus one trend-strength filter is enough. RSI, Stochastic and CCI, calculated on the same candles, will show nearly the same thing — three agreeing oscillators aren't "triple confirmation," they're one piece of information in three wrappers. The simplest and most honest measure of momentum remains plain N-day return — no smoothing, no magic.

How to Apply It Step by Step

An example time-series momentum system for crypto (D1):

1. Strength measure: the return over the last 90 days (ROC 90). Additional filter: price above the EMA50, with the EMA50 sloping upward.
2. Entry signal: ROC 90 > 0 and price breaks the 20-day high on volume above average. Volume confirms that someone besides you is buying the strength (how to play breakouts).
3. Stop: 2x ATR(20) below entry — momentum needs room to breathe, not a tight stop.
4. Exit: loss of momentum, not a "target level" — e.g., a close below the EMA50, or ROC 90 falling below zero. An additional warning sign: RSI divergence at the top of the move.
5. Risk: 1% of capital per trade (position sizing).

Numerical example on ETH (illustrative): a $10,000 account, 1% risk = $100. ETH has ROC 90 = +38%, price $2,600 above a rising EMA50 ($2,450). A 20-day high breakout on volume 1.4x average. ATR(20) = 90, stop = 2,600 − 180 = 2,420. Risk per ETH: $180, position = 100 / 180 ≈ 0.55 ETH (notional ~$1,430). The move runs to $3,050, then ETH closes the week below the EMA50 at $2,890 — exit with a profit of 0.55 × 290 ≈ +$160 (1.6R). If the breakout had died immediately: a −$100 loss and waiting for the next signal.

Cross-sectional variant (simpler than it sounds): once a week, rank 10–20 liquid coins by ROC 90. Go long only in the top 2–3, on the condition that BTC is above its own EMA200 (a filter for the whole market's regime). Falling out of the top ranks = close the position. Zero prediction — pure rotation of strength.

In the crypto variant, it's worth measuring alt strength against BTC, not just in dollars: an alt rising slower than BTC has no momentum of its own — it's riding on borrowed fuel and will fall harder the moment Bitcoin catches its breath. An ALT/BTC chart rising together with ALT/USD is a sign the strength is real, not borrowed from the whole market.

When It Doesn't Work and the Most Common Mistakes

• Market reversals are momentum's execution. The biggest historical momentum-strategy crashes came during sharp reversals — what was strongest fell the hardest. A regime filter (BTC above/below EMA200) eases the problem but doesn't eliminate it.
• Buying strength after the fact. Entering after five green candles in a row "because there's momentum" is often buying the local top. The system enters on a defined signal (breakout, ranking), not on the emotion of "the move is getting away from me."
• A sideways market grinds up signals. When there's no trend, ROC oscillates around zero and breakouts reverse. ADX below ~20–25 is information: momentum has nothing to do here — consolidation is played differently.
• Meme coins and garbage liquidity. The biggest momentum in crypto belongs to assets where you can't sensibly place a stop: spread, slippage and pump-and-dumps can eat the entire edge. Strength without liquidity is a trap, not a signal.
• Indicator overload. Five oscillators measuring the same thing give an illusion of confirmation and decision paralysis. Speed measure + trend filter + volume — more usually doesn't help, it just adds reasons to break the rules.
• Ignoring costs. Intraday momentum generates a lot of trades; on an expensive exchange, commissions and spread can flip the sign of the result. Before you trust a backtest, add in your exchange's real costs.

Momentum and mean reversion are two opposing forces of the same market: inertia and gravity. The art isn't picking the "better" one, it's recognizing which one currently governs your market and your timeframe — and that's decided by measurement, not conviction.

The moral: momentum is real, measured, and works on crypto too — but it pays irregularly and can bite hard on a market reversal. Before you trade strength with real money, check on the data how your rule behaved in the last crash. If you don't know — you don't have a system yet, you have a slogan from a course.

FAQ

Rafał — Krypto Bez Ściemy

Trader and founder of Strefa Tradingu. He’s been breaking crypto down on YouTube for years — no hype, no signals, with a focus on market structure and risk management.

kryptobezsciemy.plInstagramYouTubeKlub Zero