Beyond Moving Averages: Why Volatility Trends Matter More Than You Think

By Deb Shaw | on October 19, 2025 7 min read

Table of Contents

Beyond Moving Averages: Why Volatility Trends Matter More Than You Think

In the previous post, we applied a simple trend-following model to TiltFolio Balanced: a buy-and-hold portfolio of stocks, bonds, and gold. The rules were straightforward:

Invest only in assets trading above their 10-month moving average, and
Allocate fully to the one showing the strongest 10-month momentum.

Even this basic rule-based approach outperformed buy-and-hold by roughly 50%. That shouldn’t be surprising: trend-following systematically removes underperformers. But the question worth asking is why this works so well… and whether it can be improved.

This post explores exactly that. We’ll look at why long-term moving averages work, what their limitations are, and how a more advanced approach: one inspired by TiltFolio Adaptive: can capture the same essence with much greater precision.

Why Long-Term Moving Averages Work

If you’ve ever wondered why the 10-month moving average (or its close cousin, the 200-day moving average) appears so magical, you’re not alone. The answer lies somewhere between market noise and human behavior.

Short-term moving averages: anything under about six months: are notoriously noisy. They react to every squiggle in price, leading to frequent whipsaws: quick reversals that trigger false buy or sell signals. The costs of those whipsaws usually outweigh any benefit from avoiding downturns.

On the other extreme, very long-term averages (12+ months) are too sluggish. They only react after most of the damage or opportunity has already occurred.

The “sweet spot” lies between 6–12 months, where the average is long enough to filter noise but still responsive enough to sidestep major drawdowns. Countless backtests across dozens of asset classes confirm this range as the most robust middle ground.

The Hidden Variable: Volatility Clustering

Meb Faber famously demonstrated that when stock indices trade below their 10-month moving average, returns fall by over 60% while volatility rises by almost 50%. This isn’t coincidence: it’s the phenomenon of volatility clustering.

Volatility, in simple terms, tends to beget more volatility. Calm markets stay calm, and turbulent ones stay turbulent. The 10-month moving average, though primitive, acts as a crude but effective filter for volatility environments: flagging when the market is transitioning from calm to stormy waters.

That realization changed how I viewed trend-following entirely. Moving averages weren’t just price filters; they were indirect volatility filters. And if volatility clustering is the mechanism behind the edge, then the natural next question became: Can we do better by directly measuring volatility itself?

Searching for a Better Volatility Signal

Driven by that question, I began exploring whether it was possible to predict volatility: not just react to it after the fact.

My first stop was the obvious one: the VIX, the market’s implied volatility index. The VIX represents the market’s expectation for volatility over the next 30 days, inferred from option prices. It’s widely cited, and intuitively appealing: if the VIX is rising, volatility expectations are rising too.

But there’s a catch. The VIX’s history is relatively short compared to global asset data. Moreover, it’s heavily influenced by options market positioning: meaning it can sometimes reflect hedging demand rather than genuine fear.

So I went deeper. I studied option returns themselves: the profitability of selling puts or buying calls: as a way of inferring sentiment. I experimented with realized versus implied volatility spreads. None of these fully satisfied me. They were good, but did not have long-enough histories.

Then I stumbled on something different: what I now call the internals of the market.

The Market’s “Internals” as a Forward Signal

Instead of looking at volatility indexes, what if we could infer risk appetite from within the stock market itself?

The concept is simple: compare the performance of riskier stocks (small caps, cyclicals, high beta sectors) to safer ones (large caps, defensives, low volatility sectors). When riskier stocks outperform, it signals a healthy, low-volatility environment. When safer stocks lead, volatility is rising, and risk appetite is fading.

I liked this idea for two reasons:

Deep historical data. Unlike derivative-based measures, the relative performance of stock groups can be backtested over many decades, providing a much richer dataset.
Forward-looking behavior. Market internals capture what investors are currently doing with risk, not what volatility has been. Historical volatility is backward-looking: it lags. Internals are anticipatory.

When I tested it, the results were astonishing. The internal behavior of the equity market: this simple risk-on versus risk-off relationship: predicted not just future stock returns, but also the behavior of other asset classes.

”The Best Economist I Know Is the Inside of the Stock Market”

This conclusion echoed the words of Stanley Druckenmiller, one of the most successful investors alive, who once said:

“The best economist I know is the inside of the stock market.”

After studying the data myself, I came to the same conclusion. The relative strength between riskier and safer stocks effectively forecasts the macro environment: growth, inflation, and liquidity: before economic data or central bank decisions confirm it.

In other words, the stock market’s internal behavior functions as a leading indicator for all asset classes:

When internals show rising risk appetite, stocks and gold tend to outperform.
When internals weaken, suggesting rising volatility, bonds and commodities perform better.
And when no clear leadership exists, cash is often the safest bet.

This interplay between trend-following (price direction) and volatility (risk environment) became the intellectual foundation for TiltFolio Adaptive: our flagship, all-weather trend-following system.

From Moving Averages to Volatility Trends

By combining long-term trend signals with volatility dynamics, we can build systems that adapt fluidly to changing market regimes.

Moving averages tell us what is trending.
Volatility trends tell us when those trends are likely to persist.

In periods of falling volatility, risk-taking is rewarded, and assets like equities and gold flourish. In rising-volatility environments, defensive assets such as bonds or cash dominate. This combination produces smoother performance curves with smaller drawdowns: the holy grail of long-term investing.

From Intuition to System

I often joke with friends that TiltFolio Adaptive is “Druckenmiller-as-a-service”: a systematic way to express the kind of all-in, conviction-driven macro positioning that Druckenmiller himself is famous for, but through rules, not instinct.

Of course, the system itself is proprietary. But conceptually, it rests on two pillars that any investor can appreciate:

Follow the trend: focus only on assets in sustained uptrends.
Respect volatility: adjust exposure based on whether risk appetite is expanding or contracting.

The result is a dynamic portfolio that doesn’t try to predict news, policy moves, or economic cycles. It simply responds to what the market is already signaling beneath the surface.

The Takeaway

Moving averages remain one of the simplest and most effective tools for detecting long-term trends. But their real power comes not from the lines themselves: it comes from what they represent: changes in volatility and investor behavior.

By layering in a volatility trend filter derived from market internals, we can dramatically improve a traditional trend-following system’s responsiveness and stability.

In other words, beyond moving averages lies a deeper truth: Volatility trends drive everything. Learn to read them, and you move from reacting to the market: to anticipating it.

How TiltFolio Works Series

This post is part of the “How TiltFolio Works” series. Explore all posts in the series: