The Ultimate Moving Average Handbook: Bringing Science into the Art of Trend Following

Book Motivation

My friend, Valeriy, has produced some excellent trend following research articles over the years (much of which is on this blog!). Of course, trend-following isn’t new. Traders have been chasing trends for centuries, and moving averages are the workhorse tool behind it all. The basic idea is simple: markets are noisy, so smooth the data and try to extract the signal.

Sounds straightforward—until you realize there are hundreds of ways to do it. (we have an entire category of articles dedicated to this!)

Over time, we’ve seen an explosion of moving averages and trading rules. Each one claims to be faster, smoother, or somehow “better.” You’ll hear terms like zero-lag, super-smooth, or DSP-based thrown around as if they’re breakthroughs.

But here’s the problem: most of those claims don’t hold up under scrutiny.

The evidence is usually anecdotal. Maybe a nice-looking chart. Maybe a backtest with favorable assumptions (they always look good, right?). Rarely do we see a rigorous, apples-to-apples comparison grounded in data.

So we’re left with a fragmented field—lots of ideas, very little structure, and almost no consensus on what actually works.

This book is an attempt to fix that. The goal is simple: replace intuition, marketing, and hand-waving with a disciplined, quantitative framework. Instead of asking “which moving average looks best,” we ask a better question—what are the underlying trade-offs, and how do different methods actually stack up when you measure them properly?

What the book covers

The book combines two perspectives: 1) it is both a comprehensive guide to moving averages and 2) a new analytical framework for understanding them.

At the core is a simple idea that’s often overlooked: any trend-following rule should be judged on three dimensions:

• Responsiveness (how quickly it reacts)
• Smoothness (how much noise it filters out)
• Accuracy (how well it actually tracks the underlying trend)

Most practitioners focus on the first two, often assuming that smoother means more accurate. That’s a mistake. Smoothness controls noise, but accuracy is about how closely you track the underlying trend—and the two do not always move together

This book builds quantitative measures for all three—and then uses them to compare methods on a level playing field.

Once you do that, a few things become clear:

• Many “different” methods are just variations of the same underlying structure
• Claims of “zero lag” or “super smooth” always come with trade-offs
• There is no free lunch—every improvement in one dimension costs you in another

From there, the book shifts from analysis to design.

• How do you build better rules?
• How do you balance competing objectives?
• What does “optimal” even mean in a real-world setting?

The book also tackles the messy reality of implementation—how trading costs matter, why backtests can mislead, and how market dynamics change the game.

In short, the book is not just a reference—it is an attempt to turn a largely intuitive field into a coherent, quantitative discipline.

I highly recommend you check it out. Here is the amazon link.