PATH TO FI™ · BLOG

Core and Satellite: A Portfolio That Knows Its Job

October 6, 2026 · 7 min read · By Prakash Damodaran

Every portfolio has a job description, whether you wrote one or not. If you never wrote it, the market will assign one for you, and the market's version is unkind: your savings become a scratch ticket with fees.

The fix is to write the job description yourself, and it has two lines. Most of your money belongs in a boring core that compounds on autopilot. Your conviction bets live in small satellites around it, sized by a written checklist before you buy. The core does the compounding. The satellites do the learning, and occasionally the outperforming. Neither does the other's job.

What is the core?

Most of your money belongs in a boring core: broad, low-cost index funds bought automatically. The core is the part of your portfolio you could ignore for a year and still sleep, holding hundreds or thousands of companies at once. It does not try to beat the market. It tries to be the market, at almost no cost, with no decisions required from you.

That sounds like settling. It is not. It is the recognition that most professional stock pickers, people who do this forty hours a week with research teams, fail to beat the market over long periods. If the pros cannot reliably do it, your core should not depend on you doing it either.

The core has exactly one job: carry you to your FI number. That number is (your recurring annual spending divided by your chosen withdrawal rate) plus your planned one-time expenses, added on top, never multiplied. Everything stays in today's dollars. The withdrawal rate is your choice: 3% means 33 times spending, 4% means 25 times, 5% means 20 times, and 4% is a reasonable starting point for a horizon near thirty years.

Notice what the core does not need from you: brilliance, timing, or attention. It needs money, fed in on schedule, for a long time. Automation is what makes that possible. A common shape is 80 to 90 percent of the portfolio in the core, with the rest in satellites. The exact split matters less than the funding order: the core gets funded first, automatically, every contribution.

What lives in the satellite?

Satellites are concentrated positions in individual companies, sized by conviction. They exist because some investors, having done the work, develop real conviction in a specific business and want their capital to reflect it. The satellite is where that conviction is allowed to live, in the open, with a name and a size limit.

Satellites are optional. The core alone is a complete strategy. But for those who want them, a satellite starts small and earns its size. A typical satellite is 1 to 5 percent of the portfolio: enough that a win matters, small enough that a total loss is a tuition payment, not a catastrophe. The number that matters is not the percentage itself but the rule that produced it, because the rule is what survives your emotions.

Satellites are also where mistakes are supposed to happen. Everyone picks losers. In the core, a loser is invisible, one company among thousands. In a satellite, a loser is a lesson with a receipt. The engine's first job is to make sure every lesson is affordable.

Why never reverse the two?

Reversing the two looks like this: most of your money in a handful of exciting stocks, and a little index fund on the side "for safety." It feels diversified because two kinds of assets are present. It is not diversified. It is speculation with a fig leaf. Never reverse the two: speculation as the core is how accounts die.

The problem is arithmetic, not morality. A concentrated core has to be right about picking. A boring core only has to be right about showing up: saving, investing, waiting. One of those is a skill almost nobody sustains. The other is a habit almost anyone can sustain. Speculation as the core is how accounts die, because it asks the part of your portfolio that must survive to also be the part that takes the biggest risks.

Ask the reversed portfolio one question: if every satellite went to zero tomorrow, could you still reach your FI number on the core alone? If the answer is no, the satellites are not satellites. They are the core wearing a costume.

How do you size a satellite position?

Conviction is a feeling. Sizing is a decision. The engine's first job is sizing, not picking. Even a correct pick at the wrong size is a failure: too small and it changes nothing, too big and a normal drawdown breaks you.

So the sizing decision happens before the buy, in writing, when you are calm. The checklist turns "I really believe in this" into a number: a starter size, a maximum size, and the evidence that would earn an increase. Write it down, because the version of you who wrote it at noon is smarter than the version of you watching the price at midnight.

What does the checklist ask?

• What is the thesis, in two sentences? If you cannot state it plainly, you do not have conviction. You have a feeling about a ticker.
• What would prove me wrong? Name the evidence that kills the thesis before you buy. Write the exit condition when you still want to be in.
• What size, and why this size? A starter satellite is typically 1 to 2 percent of the portfolio. The maximum, earned by evidence over time, might be 5 percent. The number is less important than the reasoning written next to it.
• How long is the horizon? A satellite bought for a five-year thesis cannot be judged on a five-week chart. Write the horizon down so a bad month cannot rewrite it.
• What does zero cost me? If this position went to zero, the core still carries me to my FI number. If that sentence is not true, the size is wrong.

Run every satellite through all five before money moves. The checklist does not make your picks better. It makes your mistakes cheaper, which over a lifetime is the same thing.

How do the two parts work together over time?

The funding order never changes: the core gets fed first, automatically, every payday. Satellites get funded from what is left, and only after the core's contribution has already moved. This order is the whole discipline. It guarantees that your conviction never eats your compounding.

Winners get managed by the same checklist that sized them. A satellite that grows past its maximum size gets trimmed back to the checklist limit, with the profits moving into the core. Letting a winner become the core by accident is how a concentrated bet quietly becomes the portfolio. Winners get harvested, not crowned.

Losers get managed by the exit condition you wrote when you were calm. "What would prove me wrong" was answered before the buy, so the sell is an execution, not a debate. The tuition gets paid, the lesson gets logged, and the core keeps compounding like nothing happened. Because for the core, nothing did.

What if a satellite goes to zero?

Then you have paid tuition, and the core does not notice. That is the entire design. A satellite at 2 percent of the portfolio can fall 50 percent and cost you 1 percent of your net worth: a bad week for the market, a rounding error for your plan.

Compare that to the reversed portfolio, where a 50 percent fall in the biggest holding can delete years of savings. Same mistake, different architecture, opposite outcomes. The architecture is the strategy.

Start boring. The core goes in first, automatic and diversified, before the first satellite is ever bought. Earn the right to speculate by building the part that survives. Sizing first, picking second, speculation never as the core. That is the whole engine, and it fits on an index card.

← All articles