Portfolio performance data tends to be rooted in several big assumptions. Unless noted otherwise, returns for any given time period assume a single lump-sum investment with no additional investments, no withdrawals, no inflation and no taxation.

These assumptions have an indisputable value: They create a common starting point and simplify analysis and comparison. But there's a risk for advisors in relying too much on these assumptions - because they set up a disconnect with real-world performance, which is of course what matters ultimately to clients.

As an example, consider the impact of the sequence of returns on three different types of portfolios, as defined by the way the money will be invested or withdrawn. You might have an accumulation portfolio that assumes a single lump-sum deposit - a common theoretical assumption, but fairly uncommon in the real world. More common might be a distribution portfolio, in which money is being withdrawn systematically from a nest-egg balance, or an accumulation portfolio that assumes annual investments.

Here's why these differences matter: When you look at real-world performance - in this case, the sequence of returns - for these three different types of portfolios, you see an astounding range of outcomes.

Click to view related chart: Baseline Portfolio Returns

BASELINE PERFORMANCE

To study the sequence of returns, we started with the performance of a 12-asset, broadly diversified portfolio. The Baseline Portfolio Returns chart above shows the annual returns as they actually occurred over the past 15 years.

Let's examine how each of the three portfolio scenarios noted above play out if we alter the sequence of returns.

The Lump Sum Analysis chart below shows the ending balance of a portfolio assuming a single $15,000 investment on Jan. 1, 1999 - unusual in real life, although it might represent, say, a gift made to a toddler and intended for college costs.

The blue line represents the growth of $15,000 based on the actual year-to-year returns of the diversified portfolio.

The red line represents the growth of $15,000 if we change the sequence so that the lowest returns occurred first and the highest returns occurred at the end of the span. (Thus, the return of -24.62% occurred first, followed by -1.66%, followed by -1.01%, and so on. The last return in the 15-year period would have been 27.09%.) And the green line flips the order, assuming the largest returns occurred first and the lowest returns occurred last.

Regardless of the order of returns, the ending balance was identical 15 years later on Dec. 31, 2013; the sequence of returns made no difference in the final outcome.

The key word here is "final." As you can see, the account balances for the three different sequences of return were very different along the way - but they ultimately landed in exactly the same place.

Click to view related chart: Lump Sum &Distribution Portfolio Analysis

DISTRIBUTION PHASE

But such lump-sum portfolios aren't very representative of real-world client needs. Much more likely would be a retiree's portfolio that needs to tap a nest egg over time.

The Distribution Portfolio Analysis chart above looks at a retirement portfolio that started with $250,000 on Jan. 1, 1999. It turns out that the sequence of returns matters greatly here.

This analysis assumes a 5% initial withdrawal rate - putting the first of the 15 annual withdrawals at $12,500 - and a 3% annual cost-of-living-adjustment that increases the withdrawal amount each year. Over the 15-year period, the withdrawals add up to a total of $232,486.

As you can see, the ending outcome in a distribution portfolio is dramatically affected by the sequence of returns.

The blue line shows the annual account balance of distribution portfolio based on the actual year-to-year returns of the diversified portfolio; the ending portfolio balance was a bit more than $438,000.

The red line shows the annual ending balance assuming the lowest returns occurred first and the highest returns occurred at the end of the period. This would be a catastrophic sequence of returns for a distribution portfolio, with the ending account balance tumbling to less than $160,000.

And again, the green line assumes the largest returns occurred first and the lowest returns occurred last. This is the ideal scenario for a distribution portfolio; the ending balance in this scenario would top $573,000.

The key takeaway here: For a retirement portfolio, it's critically important to avoid negative returns in the early years of the distribution period.

ANNUAL CONTRIBUTIONS

The third scenario focuses on clients who are still accumulating wealth.

The Annual Investment Portfolio Analysis at right tracks a simple portfolio that starts from scratch with $1,000 invested annually at the start of each year, starting on Jan. 1, 1999. After 15 years, the final balance was substantially different depending on the sequence of returns.

Again, the blue line represents actual performance of the diversified portfolio; it shows an ending portfolio balance of just under $30,000. The red line shows the annual ending balances assuming the lowest returns occurred first - the ideal sequence of returns for a portfolio that is being added to each year - ending with a balance that's just a bit less than $50,000. And the green line shows annual balances if the largest returns occurred first - the least optimal scenario for this type of portfolio. You'll note that the ending account balance is a bit more than $20,000.

For a client's growing portfolio, the sequence of returns again has a significant effect. You ideally want to experience larger returns in the latter years, because those higher returns will affect a larger account value.

Click to view related chart: Annual Investment Portfolio Analysis

ADVISOR IMPACT

What does this all mean for advisors? After all, it's not possible to create an optimal sequence of returns.

Yet it is possible to tilt a portfolio toward a certain level of return, keeping in mind the type of portfolio it is.

For example, a retiree's portfolio is very sensitive to losses during the first five to 10 years after retirement, when the client begins withdrawing money. A retirement portfolio that is sustaining annual withdrawals should have two mantras: Avoid large losses and provide a modest return.

So an advisor might design that portfolio with a more conservative approach in the early years to avoid a bad sequence of returns. If appropriate, the portfolio could be tweaked toward a more aggressive stance after the retiree is safely out of the gate - say, 10 or so years into retirement.

The asset allocation (and therefore risk level) of a distribution portfolio can change and adapt over time, becoming more or less aggressive. But in those key early years, don't swing for the fences. Be more conservative to avoid an early meltdown.

Accumulation portfolios are a bit less sensitive to sequences of returns. Whether based on a lump-sum investment or annual investments, the appropriate risk level for the portfolio design is largely dictated by the age of the investor and the time the money will likely be needed. Younger clients, for instance, would generally want to maximize return potential from the start, so their portfolios would be largely equity based.

Annual investors would need to start tapering portfolio risk as they approach transition points such as retirement. Negative returns at the end of the investment horizon are more painful than at the start of the time horizon, because they will affect the larger balance that has accrued.

Across the board, advisors should reduce a portfolio's risk as the client approaches retirement and for the first several years after retirement. This logic suggests a roughly 10- to 15-year safety window (five years before retirement and five to 10 years after), during which the portfolio should be designed to avoid a large loss.

We can't control the sequence of returns, but we can control the risk level of a portfolio during periods of time in which it is particularly sensitive to losses.

Craig L. Israelsen, a Financial Planning contributing writer in Springville, Utah, is an executive in residence in the personal financial planning program in the Woodbury School of Business at Utah Valley University. He is also the developer of the 7Twelve portfolio.

This article has been updated to correct an error.

Read more: