What is evidence-based investing?
Updated: 3 hours ago
Evidence-based investing is a way of deciding, not a list of products to buy. Anyone can claim the label, and plenty do, so here are four questions worth putting to anyone who uses it.
As a method, evidence-based investing means three things. You work out what the evidence actually supports. You act on it. And you change your mind when the evidence changes. The third part is the one that gets skipped.
Nobody owns the term. No regulator defines it, no body certifies who may use it, and nobody is penalised for using it loosely. So when a firm, an adviser or an article tells you its approach is evidence-based, the job of working out whether the phrase means anything falls to you.
The name came out of a conversation in 2015, when I was about to start a blog. I knew what I believed about investing and I had no idea what to call it, so I asked Larry Swedroe, who had been reading the academic research for longer than almost anyone. He suggested evidence-based investing, and it stuck.
I have since seen it written that I coined the phrase. I did not. It was in use before Larry and I spoke: two American advice firms had published a position paper five years earlier defining evidence-based investing as a method borrowed directly from evidence-based medicine. Who used it first, I have never been able to establish, and I have looked.
The beliefs came before the label, by four years. In 2011 I was filming a documentary for a Birmingham advice firm and I interviewed Jack Bogle, the founder of Vanguard, at his office in Pennsylvania. I went in with no settled views about investing and came out with most of the views I hold today. The name arrived afterwards, which is this whole argument in miniature: the method is the thing, and the label is just what we call it.
What counts as evidence
That definition is useless without a view on what counts as evidence, because everyone with an investment product to sell claims to have some. Four characteristics matter most.
Independent. Not produced by anyone who profits from the conclusion. Research can be sound and conflicted at once, but a commercial interest in the answer is a reason to look harder at how the question was framed.
Peer-reviewed. Work that other academics have examined and picked over deserves more weight. This is a weighting, not a gate. A great deal of useful finance research circulates as working papers for years, and some of the best never reaches a journal.
Based on robust data analysis. Long periods, large samples, more than one market. A finding that holds in US large-company stocks since 1990 and nowhere else is a hypothesis, not a conclusion.
Time-tested. Tested out of sample, and tested after publication. This is the guard against a result that only ever worked in the data where somebody found it.
Evidence of this kind usually comes from financial academics. Not every academic paper is reliable. But academic work should be the first place you look, and research from fund managers, brokers and platforms read with far more scepticism.
Evidence-based investing borrowed its name from medicine
The phrase wasn't invented in finance. Gordon Guyatt, a physician at McMaster University in Canada, gave evidence-based medicine its name in a 1991 editorial in ACP Journal Club.
Five years later David Sackett and four colleagues set out what it was, and what it was not, in the BMJ. They wrote the paper to correct misunderstandings that had already grown up as the term spread, which tells you how quickly a good label attracts bad uses. Their definition turned on the conscientious, explicit and judicious use of current best evidence in decisions about individual patients, combined with the clinician's own judgement and the patient's circumstances and preferences. As they put it, 'Evidence based medicine is not "cookbook" medicine'.
By 2014 three of the method's own supporters thought it had gone wrong. Trisha Greenhalgh, Jeremy Howick and Neal Maskrey argued in the BMJ that evidence-based medicine was a movement in crisis. Commercial interests were setting the research agenda, the quality mark had been 'misappropriated and distorted by vested interests', and the label was being worn as a badge by work that hadn't earned it. They weren't arguing for abandoning the method. They were arguing for going back to it.
Medicine also has something investing lacks. The Cochrane Collaboration synthesises evidence independently, publishes its methods and has rules about conflicts of interest. Investing has no equivalent. SPIVA counts how active funds fare against their benchmarks and Morningstar does useful work on costs, but nothing stands between the investor and the evidence and sorts it. So the filtering falls to you.
What the evidence supports
Start with cost, because it's the strongest finding in the field. Costs are certain and returns are not, and every pound of charges is a hurdle a fund has to clear before you're any better off. Morningstar's Russel Kinnel calls the expense ratio the most proven predictor of future fund returns there is: across US equity funds over the five years to 2015, the cheapest fifth were about three times as likely as the priciest to survive and beat their peers, and the link between fees and performance has strengthened since. In aggregate, active management adds no value after costs. Charley Ellis called active investing a loser's game in 1975 and the scorecards have been agreeing with him ever since. I'm not going to make that case again here.
Where it leads is index funds, held at low cost and spread broadly across countries and sectors. For a great many people a single global equity tracker is all they will ever need. I write about investing for a living and I'll still say that plainly, because the method is not a sales funnel. If the evidence pointed somewhere more complicated, I'd be writing that instead.
Where I go further than a tracker
My own portfolio isn't just a global tracker, and it would be dishonest to write this piece without saying so. About half of it sits in cap-weighted index funds. The rest is tilted towards value stocks, smaller companies and more profitable ones, through Dimensional funds held via an adviser and Avantis funds held directly. I've set out the full case for those funds, with the risks and my own holdings, separately.
Apply the first criterion to that tilt and it holds. The load-bearing research is independent academic work: Eugene Fama and Kenneth French on value and size, Robert Novy-Marx on profitability. The fund companies implement those findings. They didn't produce them.
I hold the position less firmly than I did in 2015, and this is where the method earns its keep. In 2016 David McLean and Jeffrey Pontiff examined 97 published predictors of stock returns in the Journal of Finance. Portfolio returns were 26 per cent lower outside the original sample period, and 58 per cent lower once the findings had appeared in print. The authors attribute the 32-point gap to investors reading the research and trading on it. The portfolios they measure are long-short research constructions rather than anything a private investor would hold, but the warning travels.
Same method, different confidence. The evidence on costs is about as strong as evidence in finance gets, so I state it flatly. The evidence on factor premiums is weaker and contested, so I state it less firmly, keep the cost of acting on it small and visible, and accept that I may be wrong for a decade. What would change my mind is a long run of out-of-sample data showing the premiums have gone, or a convincing case that what survived costs was never large enough to chase.
Behaviour matters more than fund selection
The most common misunderstanding of evidence-based investing is that it's an argument about which funds to own. What you do in a bad year probably matters more to your outcome than what you hold.
In rough order of importance. Do less rather than more: when Brad Barber and Terrance Odean tracked 66,465 American households between 1991 and 1996, the most active fifth earned 11.4 per cent a year while the market returned 17.9 per cent. The calmest investors in a falling market are usually not the bravest, just the ones holding something that asks them to make fewer decisions. Know your own biases, because the evidence on loss aversion, overconfidence and recency is strong and they don't disappear simply because you've read about them. Treat market commentary as noise until it tells you something about the next 20 years rather than the next 20 days. And stay invested, through the falls as well as the recoveries, for far longer than feels comfortable.
Regular rebalancing belongs on that list too, and the honest reason for it is behavioural rather than a claim about higher returns.
How far the label stretches
RA Capital Management, a Boston firm, calls itself an evidence-based investor and runs concentrated, actively managed positions in healthcare and life sciences. Their evidence is clinical and scientific, the proof needed to change a standard of care, so the word is doing genuinely different work there rather than being misused. It still shows how far the phrase will stretch.
In 2025 Palgrave Macmillan published an academic book called The Evidence-Based Investor, by two finance professors, Pascal Böni and Tim Kröncke. Same title as my blog, much the same framing of the method, and a portfolio built on a global market core with an optional alpha sleeve that can take in private debt, private equity and hedge funds. Shared method, different conclusions. That doesn't damage the argument. It is the argument.
Four questions worth asking
The label will not tell you any of this, so you need a test of your own. Four questions will do most of the work, whether you're reading an article, looking at a fund brochure or sitting across a desk from an adviser. In that last case they go alongside three blunter questions about how the adviser is paid.
Who produced this evidence, and do they profit from the conclusion? Has it been through academic peer review, and if not, why not? What is it based on: how long a period, how many markets, how large a sample? And has it been tested since it was published, rather than only in the data where it was found?
Anyone calling themselves evidence-based should be able to answer all four without changing the subject. If they can't, the label is doing the work the evidence should be doing.
Resources
Barber, B. M., & Odean, T. (2000). Trading is hazardous to your wealth: the common stock investment performance of individual investors. The Journal of Finance, 55(2), 773–806.
Ellis, C. D. (1975). The loser's game. Financial Analysts Journal, 31(4), 19–26.
Fama, E. F., & French, K. R. (1993). Common risk factors in the returns on stocks and bonds. Journal of Financial Economics, 33(1), 3–56.
Greenhalgh, T., Howick, J., & Maskrey, N. (2014). Evidence based medicine: a movement in crisis? BMJ, 348, g3725.
Guyatt, G. H. (1991). Evidence-based medicine [editorial]. ACP Journal Club, 114(suppl. 2), A-16.
Kinnel, R. (2016). Predictive power of fees: why mutual fund fees are so important. Morningstar Manager Research.
McLean, R. D., & Pontiff, J. (2016). Does academic research destroy stock return predictability? The Journal of Finance, 71(1), 5–32.
Novy-Marx, R. (2013). The other side of value: the gross profitability premium. Journal of Financial Economics, 108(1), 1–28.
Sackett, D. L., Rosenberg, W. M. C., Gray, J. A. M., Haynes, R. B., & Richardson, W. S. (1996). Evidence based medicine: what it is and what it isn't. BMJ, 312(7023), 71–72.
Advisers who can answer the four questions
Anyone can call themselves evidence-based, which is why the four questions above are worth putting to a professional before you hand over any money. Every adviser in TEBI's Find an Adviser directory has publicly committed to the approach this article describes, which makes the answers easier to check.
The TEBI Digest carries TEBI's latest articles alongside the research and writing worth reading elsewhere, with Robin Powell's commentary on why it matters. Readers who want the method applied to new evidence as it arrives can sign up here.
