Guppy Multiple Moving Average

The Guppy Multiple Moving Average (GMMA) pairs two clusters of exponential moving averages—one tracking short-term trader activity, the other revealing institutional investor conviction. Where the clusters align, the trend is unified; where they diverge, a change in conviction is brewing.

Why Guppy designed two clusters, not one

Most traders default to a single moving average: the 50-day, the 200-day, the occasional custom length. This assumes the market moves as one. It doesn’t. Daryl Guppy, an Australian trader, observed that markets contain two populations with radically different time horizons and stakes. Short-term traders operate on minutes to days; they chase momentum, enter on noise, exit on fear. Long-term investors hold for quarters to years; they care about earnings, assets, competitive moats. A single moving average averages away this conflict.

The GMMA separates them. One cluster of three, five, eight, and fifteen-day exponential moving averages (EMAs) captures what traders are doing right now. The other cluster—thirty, thirty-five, forty, forty-five-day EMAs—shows what longer-horizon players believe the trend actually is. Where they converge, both camps agree the trend is real. Where they split, something is broken.

The two clusters and what they signal

Short-term cluster (3, 5, 8, 15 days): These EMAs are sensitive, reactive, responsive to intraday and swing-trader positioning. When they bunch together and point upward, traders are collectively bullish and momentum is live. When they flatten or invert (shorter ones above longer ones), trader conviction is waning. They are the nervous system of the market.

Long-term cluster (30, 35, 40, 45 days): These move more deliberately and resist noise. They filter out the chop and consolidation that confuse day traders. A long-term cluster pointing upward means institutional capital and position holders believe the asset is in a bona fide uptrend. A downward-pointing cluster is a long-term seller’s market. This cluster is the spine of the trend.

Reading the four patterns

Alignment (bullish): Both clusters slope upward, short-term cluster above long-term cluster. Traders are early, institutions confirm. Continuation likely.

Alignment (bearish): Both clusters slope downward, short-term cluster below long-term cluster. Selling is coordinated across timeframes. Weakness persists.

Bullish divergence: The short-term cluster turns up and separates from a still-flat or downward long-term cluster. Traders are buying; institutional conviction has not yet awakened. Early warning of a potential trend shift—or a whipsaw if institutions never join.

Bearish divergence: The short-term cluster rolls over while the long-term cluster remains elevated. Traders are exiting; institutions have not yet acknowledged the shift. A trailing edge or a bottoming flush depending on the next institutional response.

Practical entry and exit rules

Many traders use GMMA for trend-following entries. A common rule: enter long when the short-term cluster crosses above the long-term cluster and both clusters are rising. This removes early-entry noise and waits for institutional confirmation. Exit when the clusters begin to flatten or invert, signalling weakening conviction.

In ranging markets, GMMA produces whipsaws—the clusters converge and diverge without a sustained trend. Guppy himself advised traders to abandon the system in choppy consolidations and wait for a clear separation and directional commitment. This is not a weakness but a feature: GMMA is a trend filter, not a reversal hunter.

For shorter-term traders, some watch only the short-term cluster and use the long-term cluster as a macro bias. For longer-horizon investors, the long-term cluster matters far more; the short-term cluster is visual noise. Adjusting the lookback windows (using 2, 4, 6, 10 for scalpers or 40, 50, 60, 70 for swing traders) rescales the system to your timeframe without changing its logic.

Limitations and common misuse

GMMA lags price. By definition, a moving average is always behind the market it measures. On sharp reversals or gap moves, GMMA confirms the trend change well after the move is half over. It is not a predictive system; it is a confirmation system.

GMMA also assumes that separating short-term and long-term players is economically meaningful. In highly correlated, automated markets (index funds, ETF flows, algorithmic trading), both clusters can move in lockstep, rendering the divergence signal unreliable. The system works best in liquid single stocks and forex pairs where human discretion still influences price.

Finally, GMMA is often taught as a standalone system, but it is strongest as one ingredient in a broader framework. It answers “Is the trend intact and who believes in it?” It does not answer “Is the valuation stretched?” or “Are earnings accelerating?” A chart that shows GMMA perfectly aligned upward but the stock is trading at a 30-year valuation high is not a buy—it is a warning that the trend is priced to perfection and vulnerable to disappointment.

When to use GMMA and when to skip it

GMMA shines in directional, liquid markets with trending characteristics: individual large-cap stocks, major currency pairs, commodity futures. It fails in illiquid micro-caps, during earnings gaps and corporate actions, and in choppy consolidation ranges where no trend exists. Set it as a secondary filter: first ask if a trend is even plausible (fundamental, technical, sentiment), then use GMMA to identify alignment and entry windows within that trend.

Many traders adopt GMMA after reading one blog post and expect instant profits. They backtest on cherry-picked years of strong trends and forward-test on chop. The system is honest but not magical. Its real value is discipline: it forces you to distinguish trader sentiment from investor conviction and to wait for alignment before risking capital. In a market crowded with conflicting signals, that clarity is worth far more than any single indicator.

See also