Does a Smoother Investment Journey Mean a Better Financial Plan?

What we set out to show

Imagine asking a calculator to choose a portfolio whose value is expected to fluctuate as little as possible. That sounds appealing, particularly if large market swings make you uncomfortable. But you have only given it one instruction. You have not told it how much money you need, when you need it, or how much you can add along the way.

We wanted to make that distinction visible. We asked Portfolio Lab to choose a mix of US shares, US bonds and gold using one decade of market history, with the sole aim of reducing fluctuations. We then looked at how that mix performed over the following decade, alongside three other portfolios.

The result illustrates the trade-off: the chosen mix fluctuated less in the later decade, but it also grew much less than the alternatives shown here. We will walk through how we chose it, what happened afterwards, and why a calmer portfolio is only part of deciding whether a financial plan works.

First choose the mix, then see what happened next

We split the history into two parts. The first runs from January 2006 to December 2015. The calculator could use only those ten years to choose the percentages invested in each asset. Researchers often call this the “training period”: here, it simply means the history used to make the choice.

The calculator looked at how much the investments had moved and how they had moved together. Its instruction was to find the mix with the smallest estimated fluctuations, a method called “minimum variance”. It chose 4.8% US shares and 95.2% US aggregate bonds, with no gold. We call this the calculator’s mix below.

We then carried those same percentages into January 2016 to December 2025, the second decade. This is the “test period”: we measure what happened after the history used to choose the mix. The later returns did not influence that choice. Each month, the portfolio is brought back to its original percentages, a process called rebalancing.

Why separate the decades? A portfolio can look convincing when it is judged using the very history that helped select it. Looking at a later period asks a more useful question: how did that choice hold up afterwards? This is still a study designed with hindsight, rather than a decision actually recorded in 2015, so it cannot remove every advantage of looking back.

What does this mean for your own plan?

In the second decade, the calculator’s mix had smaller fluctuations but much lower growth than each of the three alternatives. That does not make it a mistake. It did what we asked it to prioritise. The important question is whether we asked it to solve the right problem for the person investing.

For someone who already has enough to meet a goal and places a high value on a calmer experience, accepting less growth may be a reasonable trade-off. For someone who still needs their savings to grow substantially, the same trade-off could mean contributing more, waiting longer or adjusting the goal. This example does not tell us which choice either person should make; it shows why their circumstances matter.

It also helps to separate fluctuations from losses. Volatility measures how widely returns vary, including movements upwards as well as downwards. It is not the same as losing money. The drawdown row answers a different question: how far did the portfolio fall from a previous high? Even the largely bond-based mix suffered a decline in this example, so a smoother journey did not mean a loss-free one.

Before choosing an allocation, bring the question back to your life: what amount are you trying to fund, when will you need it, and could you continue with the plan through a fall in value? The historical comparison gives you one way to examine the journey. It cannot establish whether you will reach your destination.

How we calculated it, and what to keep in mind

Each decade contains 120 consecutive monthly returns, all measured in US dollars. We represent the three investments with exchange-traded funds: SPY for US large-cap shares, AGG for US aggregate bonds, and GLD for gold. The share exposure is to the US, not to a global stock portfolio.

For readers who want the calculation detail, the optimiser uses sample covariance from the first decade only. Covariance describes how investments move together. We apply a fixed 5% shrinkage towards the diagonal, a small adjustment to those estimates intended to make the calculation more stable. This was an engineering choice, not a setting selected to make the later results look better. The calculation does not seek the highest historical average return. It cannot take short positions; the percentages total 100% and use the app solver’s one-decimal rounding.

All portfolios rebalance monthly to their stated percentages. We assume no money is added or withdrawn, and we do not add transaction costs, slippage or investor taxes. Fund expenses may already be reflected in the adjusted ETF returns. These assumptions make the comparison consistent, but an investor’s actual experience could differ.

Annual compound growth comes from the sequence of monthly portfolio values. Volatility is the sample standard deviation of monthly returns, multiplied by the square root of twelve to express it on an annual basis. Drawdown uses values at month ends, so it can miss a deeper fall during a month. The final value multiple compares the end and start of each decade separately.

We chose the assets and dates retrospectively and used the latest available adjusted price history. We did not reconstruct exactly what data would have been available at each historical date, and we did not splice in earlier history from before these funds existed. Keeping the later decade out of the allocation calculation limits one source of hindsight; it does not make this a forecast or proof that one approach is superior.

The portfolios also take different amounts of risk. This is a comparison of different investment mixes, not a contest between portfolios designed to have equal risk. The results are a worked example, not an allocation recommendation.

Underlying fund references: SPY, AGG, and GLD. Calculations use Portfolio Lab’s stored adjusted monthly ETF history, sourced from Yahoo Finance. You can download the dated results and source details.

A smoother portfolio is a choice, not a complete plan

We began with a simple question: does reducing portfolio fluctuations necessarily produce a better financial plan? In this example, the mix chosen to reduce fluctuations went on to provide a calmer experience, but considerably less growth. Neither result, on its own, tells us whether it was suitable for the person investing.

The useful lesson is to decide what you need your portfolio to do before asking a calculator to choose it. A good plan has to connect the investment journey with the amount you need and the time you have. Looking at both gives you a more useful starting point than choosing whichever result has the lowest risk number or the highest past return.

To explore that for yourself, look at how your own holdings performed in the past, then explore your future plan in Portfolio Lab. The future planning model is a separate analysis using forward-looking assumptions; those assumptions were not used to choose the historical mix in this article.