The previous two articles handed you the parts: a growth engine and a shock absorber, plus the one free lunch for combining them. This article is the assembly manual, and it is short, because the honest version of portfolio construction has fewer steps than the industry implies.
Start from the shelf, not from scratch
You do not need to invent an allocation. Decades of practitioners already condensed the sensible options into a shelf of named recipes — this site keeps a library of eight, each with a century of measured behavior: growth, worst fall, longest underwater stretch.
Browse it the honest way. Don’t ask “which one returned the most?” — over a century, all-stocks wins that contest automatically, and it also spent thirteen years underwater twice. Ask instead: which row’s worst-case column can I actually live through? That single question eliminates most of the shelf for any given person, and whatever survives is, for you, roughly tied. The differences that remain are smaller than your ability to predict them.
Adjust for when you need the money
One correction to the shelf matters more than all the others: time.
Money you will spend within roughly five years does not belong in stocks at all — no allocation wizardry changes the fact that stocks can spend five years underwater. That money lives in cash and short bonds, full stop, next to your emergency fund.
Money with decades of runway can afford the stock-heavy end of the shelf, because the recoveries that terrify a retiree are merely discounts to a thirty-year accumulator with a monthly buy order. And a portfolio you are still contributing to is sturdier than the same portfolio in drawdown-and-withdrawal mode; the paycheck is, functionally, a bond.
A serviceable rule of thumb: hold your own age in bonds, then bend the number toward your actual nerve and actual timeline. It is a starting point with a century of common sense behind it, not a law.
Run the rehearsal
Before committing, take your candidate mix to the charts and rehearse history with it. Set your weights, then look at three things: the growth path (can you accept the destination?), the drawdowns chart (can you sit through the deepest valley, for that many years?), and the start-date sensitivity view (do you understand that your own decade might be a bad-luck column?). If this money must one day pay your bills, run the retirement rehearsal too: the withdrawal rates chart shows what your mix could historically sustain, and the spending trajectories show how differently the same rule treated lucky and unlucky retirees.
This rehearsal has one purpose: to relocate your surprise from the future, where it costs money, to the present, where it costs nothing. An investor who has already seen their portfolio lose a third on paper — even 1970s paper — is measurably harder to panic than one who was promised smooth compounding.
Write it down, then stop
The finish line is one paragraph in your one-page policy: the target weights, the sentence explaining why (“70/30 because I am 30 years from retirement and verified I can stomach a 35% drawdown”), and the worst historical year your mix would have had, written in your own handwriting where future-you will find it.
Then comes the counterintuitive part: stop constructing. The urge to keep tinkering (a new asset class this quarter, a five-point shift after every headline) is not diligence; it is the disposition to buy whatever just went up, wearing a spreadsheet costume. Your allocation needs rebalancing once a year and a genuine review when your life changes: new decade, new dependents, retirement in sight. Markets changing is not a life change. That is what the allocation was built for.