Glossary

Cross-Sectional Momentum. Ranks assets by their return over a lookback window relative to the universe. The strongest performers in the cross-section are bought; the weakest are sold (or avoided in long-only mode).

Time-Series Momentum. Measures whether an individual asset's return over the lookback window is positive or negative, independent of other assets. A positive return = long signal for that asset alone.

A classification of the current market environment (Risk On / Cautious / Risk Off). Regime state gates which signals are actionable and at what allocation size.

The spread of returns across the universe at the rebalance point. High dispersion means assets are moving independently — fertile ground for momentum. Low dispersion means correlation collapse is underway; entries are blocked.

Annualised return divided by maximum drawdown. Our primary performance metric because it directly measures return per unit of worst-case pain — more relevant for leveraged crypto trading than Sharpe.

Annualised excess return divided by annualised return volatility. Useful tiebreaker but penalises upside volatility equally with downside. Used as a secondary metric after Calmar.

Average True Range. A measure of an asset's average price range over N periods. Used internally to size stop-losses in volatility-adjusted strategies.

Periodic payment between long and short holders on a perpetual futures contract. Keeps the perpetual price anchored to spot. High positive rates mean longs pay shorts; high negative rates mean shorts pay longs. A gate blocks entries when funding cost would erode signal edge.

The total number of outstanding futures contracts that have not been settled. Rising OI with rising price confirms trend; falling OI may signal weakening conviction. Used as a market-structure factor.

The ratio of accounts holding net long vs net short positions on an exchange. Extreme readings (crowded long or crowded short) are a contrarian input to the market-structure score.

A derivative contract with no expiry date. Unlike traditional futures it does not converge to spot at settlement; instead the funding mechanism keeps it aligned. All signals on MasterBitcoin target Bybit USDT perpetuals.

The ratio of notional position size to the capital deposited as margin. 2× leverage means a 10 % price move produces a 20 % gain or loss on the margin. Higher leverage amplifies both returns and drawdowns.

A pre-defined price level at which a position is closed to limit loss. Placed as a native exchange order so it executes even if the bot loses connectivity.

A stop loss that moves with the market as the position profits. Locks in gains while allowing further upside. The activation price is set at a fixed distance below the highest price reached.

The largest peak-to-trough decline in portfolio or strategy equity. Expressed as a percentage. The denominator of the Calmar ratio.

The percentage of closed trades with a positive P&L. A necessary but not sufficient metric — a 40 % win rate can be highly profitable with a good win/loss size ratio. Always consider alongside average winner vs average loser.

Maximum Favourable Excursion. The maximum unrealised profit a trade achieved before closing. Used to tune trailing stop placement and assess whether signals are capturing most of their theoretical edge.

Maximum Adverse Excursion. The maximum unrealised loss a trade suffered before closing. Used to optimise stop placement — if the MAE of winning trades rarely exceeds X %, stops tighter than X % would not have cut winners short.

The rate at which a signal's edge diminishes over time as markets adapt or as market structure changes. Short-lookback momentum signals tend to decay faster than longer-lookback ones. We monitor rolling OOS performance to detect decay early.

A confidence classification assigned to each signal. Tier 1 = highest composite score, passed all gates, validated regime. Tier 2 = good score, regime cautious. Lower tiers are informational only and not recommended for automated execution.