Let us start with the hardest sentence in investing.
If a decision feels safe and obvious, you are usually late. Everybody else already made it, so the good price has gone.
The money is usually made doing the thing that feels wrong at the time. That is uncomfortable — and being uncomfortable is the whole point.
Hold on to that, because I spent months trying to prove it wrong.
I tested twenty-seven ways to be cleverer than a simple monthly plan.
These were not opinions over coffee. Each one was written as code and run over decades of real market prices.
Twenty-six of them died. I am going to show you the best ones dying, one at a time.
The one that lived is not a strategy at all. You will see it at the end.
My own gut instinct came dead last.
I liked waiting for a share to prove itself before I bought more of it. It feels responsible. It feels careful.
Over 281 weekly purchases it finished last of everything I tested. The least money at the end, and the highest price paid per share.
Waiting for proof is not free. You pay for it in the price.
Buying more when it falls sounds wise. Sometimes it is a trap.
The rule is simple: when the price drops, buy extra. On a company that recovers, it works beautifully.
PayPal fell 81% and never came back. The clever rule turned a $24,900 loss into $43,600.
It kept buying the thing that kept falling. A rule cannot tell a sale from a funeral.
So I built a safety brake. It cost me more than it saved.
The idea was sensible: stop buying while a price is clearly falling. On PayPal it worked and the loss shrank a lot.
But NVIDIA crashed too, and on the chart it looked exactly the same. The brake stopped me buying — and handed back $76,000 of profit.
On the day itself, a temporary fall and a permanent one look identical. That is the whole problem.
Then my best result turned out to be my own mistake.
One idea beat almost everything: buy the biggest company in each industry. About ten extra points a year.
It looked too good, so I checked the data. My files thought Google was worth $3.4 trillion back in 2005. It was really worth about $80 billion.
So “buy the biggest” was secretly “buy the ones that would go on to win.” I was reading tomorrow’s newspaper. I deleted the result.
With the numbers fixed, my system lost.
I put everything I had built up against the simplest thing there is: buying a little bit of the whole market, every month, forever.
My system only drew level when a big crash happened to land in the middle of the period. The rest of the time it lost — by 1.55 and 2.62 points a year.
That was a hard day. But a result you like is not the same as a result that is true.
Then I checked it against real people’s money.
Everything before this was a simulation. This is 73 real accounts, with real money in them.
I compared what those investors actually did against what a plain monthly plan would have done. Their choice of when to buy was worth −0.0%. Nothing.
They beat the simple plan 48 times out of 100 — the same as tossing a coin. All that waiting for the right moment paid them zero.
One thing worked. It was the fee.
Cutting the charge on each contribution from 5.5% to 1.5% left about 4% more money at the end.
That is bigger than every clever idea in four studies put together — and it is the only one that is certain. No forecast. No nerve. No luck.
The typical charge those real accounts were actually paying was 5.0%. This is the part you can change this month.
So here is the whole thing, in three lines.
Buy the same amount every month. Pay the lowest charge you can find. Do not stop when it hurts.
Of 400 twenty-year runs that kept buying, not one ended below the money paid in. But you will watch it fall — perhaps 60 sen back for every ringgit, for a few years.
That is the discomfort Robert Arnott named on the first card. It is the price of the return, and nobody can remove it for you.
See what the fee is costing you → See the other storiesIllustration only, not a recommendation to buy, sell or hold any specific investment. All figures are historical backtests on past data over a single favourable market path — they are not a forecast and not a promise of future returns. Past performance does not indicate future returns. Adezeno is a licensed Unit Trust Consultant with Eastspring Investments Berhad, FiMM No. F01029300.