Calendar-Year Return Bias: Why 'This Fund Returned 37% in 2023' Can Mislead

“2023 was a great year for this fund”

You have probably seen a fund described this way: strong calendar-year return, named in a headline, compared against other funds’ calendar-year returns in the same table. It feels natural, because we think in calendar years. But 1 January to 31 December is just one possible 365-day window among 365 that exist in any given year. A fund whose calendar-year return looks ordinary can have had an excellent 12 months measured from, say, 15 March to 14 March the following year.

The same fund, four calendar years

Here are Parag Parikh Flexi Cap’s real calendar-year returns, computed from actual NAV on or near 1 January and 31 December of each year:

2023 looks like the standout year at 37.64%, and 2022 the weak one at -6.29%. Both are real, correctly computed numbers. The issue is not that they are wrong, it is that they are only four data points, each anchored to one fixed boundary (1 January), out of the far larger set of 1-year windows the same fund actually lived through.

What rolling windows show that calendar years hide

In a separate check of this same fund using monthly-starting 1-year windows across its full history, the 1-year return ranged from -6.29% to 43.88%, with a median of 9.12%. The 2022 calendar year (-6.29%) sits close to the bottom of that full range, but it is not a special outlier created by the calendar boundary - a window starting a month or two earlier or later in that stretch would have shown a broadly similar result. The point is not that calendar years are inaccurate, it is that they are one arbitrarily chosen sample, and treating that sample as the year’s story discards the rest of the picture.

Why this bias creeps into comparisons

Calendar-year tables are common because they are easy to build and easy to read across many funds at once: every fund gets exactly one number per year, aligned to the same boundary. That convenience is also the trap. Two funds can have nearly identical 12-month rolling return distributions and still look very different in a calendar-year table, purely because of when within the year their best and worst stretches happened to fall relative to 1 January.

How to avoid being misled by it

Treat a calendar-year return as one sample from a wider rolling return distribution, not as a complete description of a fund’s year. If a fund’s calendar-year number looks exceptional, check whether its rolling 1-year median tells the same story; if it does not, the calendar year was likely a function of timing rather than a representative result. When comparing funds, prefer rolling return medians and the worst-case window over any single calendar-year figure.

Frequently asked questions

Is a calendar-year return ever a fair comparison?

It is fair between two funds for the exact same calendar year, since both are measured on the same two dates. It stops being fair the moment you use it to judge a fund’s overall skill or consistency, because it is only one window out of many.

Why do fund fact sheets lead with calendar-year or financial-year returns?

Because regulatory formats and common convention use fixed-year boundaries, and they are simple to read in a table. That convenience does not make the number more representative than a rolling return.

How is a rolling 1-year return different from a calendar-year return?

A calendar-year return is fixed to 1 January and 31 December. A rolling 1-year return recalculates the same 1-year window starting from every possible date in the fund’s history, so it captures many more real outcomes than the single calendar-year slice.