What decades of evidence say about predicting stock markets: market timing, professional fund performance, the cost of missing the best days, diversification, and a sensible approach for UK investors.
Many people believe that with enough skill, data or an algorithm, you can predict when shares will rise and fall. The evidence suggests that consistently predicting short-term market moves is extremely difficult, even for professionals. Understanding why helps you avoid costly mistakes and focus on what you can control.
Why short-term prediction is so hard?
- Prices reflect information quickly: news, earnings and economic data are rapidly reflected in prices by millions of investors, making it hard to gain an edge.
- Markets react to surprises: by definition, surprises are hard to predict.
- Randomness: in the short term, price movements contain a lot of noise.
- Costs: trading costs, spreads and taxes eat into any edge.
What the evidence shows about professionals?
Studies such as S&P's regular SPIVA reports consistently find that, over long periods, most actively managed funds underperform their benchmark index after fees. Funds that outperform in one period often don't repeat it. That doesn't mean no one beats the market, but identifying who will in advance is very hard.
The cost of missing the best days
Many of the market's best days happen close to its worst, often during periods of high volatility. Investors who sell during falls and wait for things to "settle" risk missing the recovery. Analyses of major indices regularly show that missing a small number of the best days over a couple of decades can dramatically reduce returns.
Behavioural traps
- Overconfidence: believing we can predict better than we can.
- Recency bias: assuming recent trends will continue.
- Loss aversion: losses hurt more than gains please, leading to panic selling.
- Herding: following the crowd into fashionable investments at high prices.
- Confirmation bias: seeking information that supports what we already think.
What about long-term trends?
Over long periods, global shares have historically delivered returns above inflation and cash, reflecting company growth and dividends, but with significant falls along the way. Past performance isn't a guide to the future, and there have been long periods when markets went nowhere.
Beware predictions for sale
- "Trading signals", bots and courses promising market-beating returns;
- social media "finfluencers" promoting investments, who may be paid to promote them;
- schemes promising guaranteed high returns.
Check firms on the FCA register and use the FCA's ScamSmart tool. The FCA has warned about illegal promotion of financial products on social media.
A sensible approach
- Invest for the long term, typically five years or more, and keep an emergency fund in cash.
- Diversify across many companies, sectors and countries, for example using low-cost global index funds.
- Keep costs low: fees compound over time.
- Invest regularly ("pound-cost averaging") to reduce the risk of investing everything at a bad time.
- Use tax wrappers such as ISAs and pensions.
- Match risk to your goals and time horizon, and rebalance occasionally.
- Don't check too often: frequent checking encourages reacting to noise.
What you can control?
You can't control market returns, but you can control how much you save, the fees you pay, how diversified you are, the tax you pay, and your own behaviour during falls. Over a lifetime, these usually matter more than any forecast.
Lump sum or drip-feed?
Historically, investing a lump sum straight away has more often beaten drip-feeding, because markets tend to rise over time. But drip-feeding reduces regret if markets fall soon after. Either is reasonable for long-term investors.
Frequently asked questions
Is it a good time to invest?
No one knows reliably. Time in the market generally matters more than timing.
Should I sell when the market falls?
Selling during a fall locks in losses. Review your plan, not the headlines.
Sources and further reading
Facts checked on 20 October 2026. Spotted an error? Tell us and we will correct it.