Skip to main content
Advising AlphaFREE REPORT
The behavior gap

Don’t be average.

The average investor consistently earns less than the very funds they own. Not because the market is rigged, and not because they picked bad funds. They earn less because of when they buy and when they sell.

The gap is small in any single year. Compounded over a working lifetime, it is the difference between a comfortable result and a disappointing one. This report shows the size of the gap across three independent studies, why it exists, and what disciplined investing looks like instead.


What the data says

Three studies, one finding.

Three of the most-cited bodies of research on investor returns all reach the same conclusion: the typical investor trails the funds and indexes they invest in. They differ on the size of the gap because they measure different windows and different investor populations.

1.2 pts / yr
GAP
Morningstar — Mind the Gap, 2025
Over the 10 years through December 2024, the average dollar in US funds and ETFs earned 7.0% a year versus the 8.2% those funds themselves returned. That 1.2-point gap is roughly 15% of the total return, lost to timing.
1.1 pts / yr
GAP
DALBAR — QAIB, 2025
Over the 20 years through December 2024, the average equity-fund investor earned 9.24% a year versus the S&P 500's 10.35%. A 1.11-point annual gap — modest yearly, large once compounded.
~3.6% / yr
GAP
J.P. Morgan — Guide to the Markets
Across all investor types — including those who hold bonds, sell in downturns, and chase performance — J.P. Morgan's Dalbar-based “average investor” earned roughly 3.6% a year over 20 years, far below a simple 60/40 portfolio's ~7.4%. The wider the behavior, the wider the gap.

The three studies cover different windows (10 vs 20 years) and different populations (equity-fund investors vs all investors), which is why the headline figures differ. Each is sourced below.


What it costs over time

$10,000, twenty years, three outcomes.

A one-point annual gap sounds trivial. Run it forward twenty years and it is anything but. Below, $10,000 compounded at each annualized rate — the average equity investor, the S&P 500, and our backtested Core 20 flagship — all measured as equity, over the same window.

$58,564Average equity investor9.24%/yr$71,688S&P 50010.35%/yr$117,080Core 2013.09%/yr · backtested

Illustrative. Each bar compounds a single annualized rate over 20 years from a $10,000 start; real returns arrive unevenly, not in a straight line. Average equity investor and S&P 500 rates from DALBAR QAIB (20 years through 2024). Core 20 is the backtested CAGR of our model portfolio over its own history, shown here on the same 20 year basis for comparison. Backtested results are hypothetical. Past performance does not guarantee future results.

Simply holding the index instead of mistiming it is worth about $13,124 here. The discipline of a conviction weighted process on top of that — what Core 20 is built to do — accounts for the rest.


Why it happens

The gap is behavioral, not informational.

The average investor does not trail the market because they lack information. They trail it because of three predictable behaviors that every study above traces back to timing:

  • Selling into fear. Money leaves funds after declines and returns after recoveries — buying high and selling low, one anxious decision at a time.
  • Chasing performance. Capital floods into whatever just worked, right as the easy gains are behind it. Morningstar found the most volatile, most-chased funds had the widest gaps.
  • Trading too much. The single clearest finding in the 2025 Morningstar study: the more investors traded, the less they made.

None of these require a market crash to cost you. They are the slow, ordinary leak that turns a market’s return into a smaller personal one.


The alternative

Invest like the process, not the mood.

The fix for a behavior gap is not a better forecast. It is a process that removes the moments where behavior does its damage. That is the entire design of the Advising Alpha model portfolios: a defined universe, conviction weights, scheduled rebalancing four times a year, and a written reason to hold every position. No selling because a headline was scary. No chasing because something else ran.

CORE 20 · OUR FLAGSHIP · BACKTESTED
Over its 25-year backtested history, Core 20 compounded at 13.09% a year versus the S&P 500’s benchmark in the same test — turning $10,000 into roughly $218,900 against about $92,400 for the index. The edge is the discipline, applied every quarter, for decades.

Core 20 figures are backtested and hypothetical. Past performance does not guarantee future results. See full methodology and disclosures below.

The Sunday Edge keeps you on process.

One free email a week — where the market sits in relation to 75 years of history, one stock, one behavioral principle. The antidote to the three behaviors above.

See what Pro unlocks →

SOURCES & DISCLOSURES

Morningstar, “Mind the Gap 2025” (US), investor return gap of 1.2 percentage points per year over the 10 years ended December 31, 2024. DALBAR, “Quantitative Analysis of Investor Behavior” (QAIB) 2025, 20 year average equity-fund investor return of 9.24% vs the S&P 500’s 10.35% through December 31, 2024. J.P. Morgan Asset Management, “Guide to the Markets,” average-investor return (Dalbar methodology, 20 year) versus a 60/40 portfolio. Figures are as reported by each source and cover the stated windows.

Core 20 performance is backtested and hypothetical. It reflects how the strategy would have behaved when applied to historical data, net of nothing, and was not achieved with real capital. Backtested results have inherent limitations and do not represent actual trading. Past performance, whether actual or hypothetical, does not guarantee future results. The $10,000 illustration compounds a single annualized rate over 20 years for each line; actual returns are not earned in a straight line.

Advising Alpha is a research publisher, not a registered investment adviser, and does not provide individualized investment advice. See our full disclaimer.