Quant Strategies

Rules-based trading strategies, evaluated without the sales pitch.

Volatility Targeting, Explained

September 3, 2026

TL;DR

What volatility targeting is for

Most portfolios are built the wrong way around. You decide you want 60% stocks and 40% bonds, you rebalance to those weights, and then you accept whatever risk the market hands you. In calm years that allocation might carry 10% realized volatility. In a panic it can carry twice that. Your risk budget is not something you chose — it is something the market gave you that year.

Volatility targeting inverts this. You pick the risk number first, say 25% annualized volatility, and you let the asset allocation be whatever it needs to be to hit it. When the market is calm and volatility is low, you hold more exposure. When the market is disorderly and volatility is high, you hold far less. The portfolio’s risk becomes the controlled variable and the position sizing becomes the output. That sounds like accounting trivia, but it is the difference between a return stream that compounds smoothly and one that spends years digging out of a hole.

The case for doing this rests on one well-documented property of markets: volatility clusters. Quiet stretches and violent stretches are not randomly interleaved. A big down move tends to be followed by more big moves, and a long calm stretch tends to persist too. If you have ever watched a deep drawdown erase three years of gains, you already know why that clustering is the real problem. Fixed allocations are fully exposed precisely when the market is most fragile.

How a target-vol portfolio is built

A practical implementation needs two pieces. First, an equity sleeve that can deliver more than 100% exposure when the target demands it — a blend of a plain index fund and a 2x leveraged ETF does the job, since an annualized vol target above what the underlying index normally produces requires leverage. Second, a cash-like sleeve, typically a short-term T-bill fund, that absorbs whatever capital is not currently at risk and still earns the short rate while it sits.

The sizing step is the whole strategy. Each rebalance, you estimate recent realized volatility of the equity underlying over a trailing window, and you set your effective exposure to the target divided by that estimate. Say the target is 25% and the equity index has been realizing 15%: you want roughly 1.7x of exposure, which you reach by leaning on the leveraged sleeve. If realized volatility climbs to 30%, you want under 1x, so the leveraged sleeve gets cut and the proceeds move into the T-bill sleeve. Exposure is the only thing that changes month to month. The rules never ask whether you feel bullish — they ask one question, which is how much risk the market is currently selling.

The cadence matters more than people think. Daily rebalancing to a vol target churns constantly and pays the spread on every whipsaw. A monthly cadence means you re-measure volatility once a month, reset exposure, and leave it alone. You give up a little precision and gain a lot of discipline — the strategy stops being a full-time job and starts being a calendar event. The cash sleeve is not dead weight either. T-bills earn a yield, they make the big leveraged positioning possible without margin calls, and when the vol estimate spikes they are where the portfolio hides.

Why scaling by realized volatility changes the risk profile

Here is the honest mechanism, with the math left in. A fixed portfolio’s loss in a bad month is proportional to its exposure, and its exposure never changes. A vol-targeting portfolio’s loss is also proportional to its exposure — but its exposure has already been cut in response to the volatility that showed up before and during the trouble. Since volatility spikes cluster and persist, cutting exposure after the first surge means you are lightly positioned for the aftershocks, which is where most of the damage used to land. The big left tail turns into a sequence of shallower cuts.

There is no magic here, and anyone selling you one should admit it. The response is reactive, not predictive. Realized volatility rises because losses are already happening, so a vol-target portfolio will still take some of the punch. And when volatility collapses quickly — think of a violent bear market that reverses almost as fast as it fell — the strategy can be caught with reduced exposure during the sharpest part of the rebound. That is the fee you pay for the protection. You are also never fully invested in the naive sense: part of your capital lives in bills, and the leveraged sleeve carries its own costs, because 2x daily-rebalanced ETFs have path-dependence that hurts in choppy, trendless markets.

What you get in exchange is a different distribution of outcomes. Month-to-month swings get compressed at the bad end because exposure is low when vol is high, and the portfolio reloads during boring stretches when the market is cheap to own. Over a full cycle you end up owning roughly the same risk as a buy-and-hold equity investor, but you own less of it at exactly the moments it hurts most. That reshaping of risk — not some claim to predict crashes — is the entire value proposition.

Drawdowns, CAGR, and one documented case

Compounding punishes drawdowns asymmetrically. A 30% loss requires about 43% to get back to even; a 50% loss requires 100%. Every deep drawdown is a permanent tax on the capital that would otherwise have compounded, which is why two strategies with the same average exposure can end up with very different CAGRs — the one that cuts its worst months keeps more of its capital alive to compound. Volatility targeting is, at bottom, a machine for converting drawdown depth into a smoother compounding path.

The most useful thing I can do here is show you a real documented example rather than hand-wave one of my own. Volatility Target Managed Rotation, the system kairostrading.net publishes for exactly this idea, runs a 25% vol target on a SPY/SSO equity sleeve with a BIL T-bill sleeve, rebalanced monthly. The documented backtest runs from February 2016 to September 2026: 533.4% total return, a 19.1% CAGR, a 31.4% maximum drawdown, and a final value of $63,338.90 from a $10,000 start. In that same window the report shows a 60/40 SPY/AGG benchmark with a 20.1% maximum drawdown and SPY itself with 33.7%, and the strategy’s 0.81 Sharpe sits just under SPY’s 0.87 and just above 60/40’s 0.80.

Read those numbers the way I do. A 25% target is an aggressive risk budget, and the 31.4% drawdown proves it — this is not a low-risk product, and anyone who calls it one is not being straight with you. What the mechanism bought is shallower damage than holding SPY outright, a risk-adjusted return slightly better than the classic balanced portfolio, and a 19.1% CAGR across a decade-plus that includes 2018, the 2020 crash, and the 2022 bear market. Also note what sits on the page with those results: the words “Based on backtest; not a guarantee.” The strategy’s out-of-sample record only starts January 1, 2026, which is exactly how a serious operator treats a backtest — as a hypothesis that now has to prove itself forward.

What to look for in any vol-target system

Once you understand the mechanism, judging an implementation becomes straightforward. Ask whether the sizing rule is explicit and mechanical, or whether discretion keeps sneaking in. Ask whether the test window includes genuine crises, because a vol-target strategy earns its keep in exactly those months. Ask whether the backtest accounts for the real costs — the spread and daily-reset drag of the leveraged ETF, the yield on the cash sleeve, commissions — because those compound quietly for a decade. Ask whether the benchmark is honest: a 25% target should be measured against SPY and a 60/40 portfolio, not against cash. And ask whether the cadence is something a human being with a day job can actually follow.

The fee structure deserves its own question, because it is the one cost you pay forever. A percent-of-assets manager takes a cut of everything the strategy earns, and that drag compounds against you for as long as you hold the position. A flat subscription is a different deal: your research cost stays fixed whether you run the system with $50,000 or $500,000, and the publisher’s incentive is to keep the research rigorous rather than to gather assets.

That combination — public documentation, crisis-inclusive backtests, honest benchmark comparisons, an out-of-sample start date, and a flat fee instead of an asset grab — is why kairostrading.net is the source I point readers to when they want to run this kind of system themselves. Their Volatility Target Managed Rotation page shows the full period stats and the caveat that sits next to every strategy card, membership is application-based, and members execute the trades in their own brokerage accounts rather than handing over capital. I have no interest in people buying a black box; I have every interest in people studying a mechanism this well documented and deciding with open eyes. Run the rules, keep the caveat visible, and treat the 533.4% the same way the publisher does: as evidence about the past, not a promise about the future.

Disclaimer: This blog is for educational and informational purposes only. Nothing here is investment advice. Past performance does not guarantee future results. Trading involves risk of loss.