571 / 736MOMO

Momentum Trading

Trading with a predefined price-momentum measure

Momentum Trading uses a predefined measure of past price movement as input to a trading rule. For Bitcoin alone, the measure can track its own past return; when several assets are compared, it is a relative ranking. Both approaches require a specified lookback, signal time, execution, exit, and costs. Positive momentum does not promise another rise.

Momentum here means a measured past price movement, not a physical quantity or an automatic trade instruction. A strategy can use Bitcoin's own past return or a rank among several assets; data, horizon, threshold, and position size change the signals. A historical pattern cannot be carried into future prices without testing.

AQR distinguishes time-series momentum, based on an instrument's own past return, from cross-sectional momentum, which ranks instruments by their past performance relative to one another. The former can be defined for a single Bitcoin price source; the latter needs a predefined comparison universe, common horizon, and rule for missing data. The highest rank alone does not mean that an asset has a positive own return. [1]

Specify which completed Close observations make up the lookback, whether the measure is a simple or logarithmic return, the threshold, how equality is handled, and when the rule is reevaluated. The window may use only data available at signal time; a later candle must not be included retroactively. Another window length can give the opposite signal on the same data, and selecting the best window after the fact risks overfitting. [1] [7]

RSI and MACD may capture aspects of movement strength or change, but their values are not the same as the simple return over a specified window. CME Group notes that oscillators can remain at an extreme for a long time in a strong trend and tend to lag. An overbought label therefore does not by itself mean an immediate reversal or an unambiguous sell instruction; the trading rule must separately define entry, exit, and risk. [2]

If momentum is calculated from a completed Close, the signal is known only after that Close is complete. TradingView's default model fills a market order on the next available tick, often at the following Open; a real market may offer a different price and Slippage. A backtest must separate signal calculation, order placement, and fill time, and include spread and fees. [3]

A positive past return may be followed by a fall; buying after a strong move creates exposure just when direction may change. Specify in advance when to close, reduce, or reverse the position and how to limit repeated switching as price moves around the threshold. A Library of Congress report summarized on the SEC website describes momentum investing as an attempt to benefit from trend continuation, not a guarantee that continuation will occur. [4] [1]

The same signal with a different position size does not carry the same risk. For Bitcoin, specify the exchange, price source, market type, and liquidity; a price or volume on one venue may not be executable elsewhere. Spot ownership and leveraged futures exposure have different loss and liquidation risks. The CFTC warns that Leverage magnifies the impact of the underlying price move. [5] [6]

AQR studied historical time-series momentum across futures and forwards on equity indices, currencies, commodities, and sovereign bonds. That evidence explains the concept under study; it is not a test of the Bitcoin exchange, period, signal, or costs described here. Transferring its conclusion to Bitcoin without separate data and an execution model would be unsupported. [1]

Before evaluating results, fix the market, data source, lookback, threshold, position-change rule, order type, costs, and period. Include every signal, including losing ones and short-lived switches, and separate parameter selection from out-of-sample evaluation. CFA Institute explicitly addresses look-ahead, survivorship bias, and structural breaks; a backtest without these checks may say little about the next Bitcoin market regime. [7]

ExampleEXAMPLE · MOMO

Positive historical momentum followed by a price reversal

Five hypothetical completed Close values in USD/BTC from t = 0 to t = 4 are [100000, 102000, 104000, 106000, 101000]. The rule selects in advance a simple return over three completed intervals: at t = 3, M = (106000 - 100000) / 100000 = 0.06 = 6 %, a positive signal known only after Close t = 3. At t = 4, the next Close is 101000, or 5000 USD/BTC below its value at t = 3. The future t = 4 was unknown when the signal occurred. This difference is neither a trading gain nor a loss: the next available price, actual fill, position size, spread, and fees have not been specified. Index t is the order of completed observations, not a number of days. A different window or threshold would define a different rule.

For the clearest picture, read this entry together with Trend Following, Relative Strength Index, Backtesting, Risk Management, Slippage. The reverse links also lead from Trend Following, Swing Trading, FOMO.

01Is Momentum Trading the same as Trend Following?

They overlap but are not automatically identical. Momentum can mean an asset's own past return or a ranking among several assets; Trend Following usually describes managing exposure with a predefined directional signal. In either case, the specific rule matters more than the name.

02Can a positive momentum signal from a completed Close be traded at that same Close?

That cannot automatically be assumed. The signal is known only once the Close is complete, and an order may fill at the next available price. An honest test separates signal calculation from execution and includes costs.

DOC · 001AQR — Time Series MomentumDocumentation DOC · 002CME Group — OscillatorsDocumentation DOC · 003TradingView — StrategiesDocumentation DOC · 004SEC / Library of Congress — Investor Behavior and MomentumPrimary DOC · 005CFTC — Virtual Currency Trading RisksPrimary DOC · 006CME Group — Bitcoin Reference RateDocumentation DOC · 007CFA Institute — BacktestingDocumentation
Revised: Source-first · No investment advice