A friend pulled up his portfolio on a call and walked me through it, fourteen positions, two AI tokens, a gaming coin, a restaking thing, an L2 I had never heard of, and a spread of small caps from a Telegram group. The combined BTC and ETH allocation was zero. When I asked about that, he told me BTC doesn't move enough, which is true in the same way that the foundation of a house doesn't move enough. What he had built was all satellite and no core, and portfolios like that tend to do the same thing in every drawdown, which is lose most of their value while their owner insists the narratives are still intact.
Core-satellite is an old idea from traditional asset management. Most of the money sits in a broad, boring base you barely touch, and a small sleeve of active bets orbits around it. In equities the core is usually an index fund. In crypto the honest equivalent is BTC and ETH, since they are the only assets in the space with a long record of surviving everything the market has thrown at them, liquidity deep enough that you can always exit, and a case for existing in five years that does not depend on a narrative staying warm. Everything else is a satellite, however strongly you feel about it this month.
Pick a ratio you can hold through a bad year
My default is 80/20. Eighty percent of the crypto allocation in core, twenty percent in satellites. If you are newer to this, or if you have ever panic-sold in a drawdown, make it 90/10 and do not feel bad about it. I know people who run 70/30 and sleep fine, but they have all been through at least two full cycles, and even then I suspect most of them overrate their edge in the satellite sleeve. The ratio is a promise you make to your future self in a calm moment, so pick one you can keep in a violent one.
Within the core I lean BTC-heavy, roughly two parts BTC to one part ETH, and I hold that view loosely. The exact split matters much less than the rule that the core is untouchable. It does not get sold to fund a hot new position. It does not get rotated into whichever sector is running. Its whole job is to sit there compounding the market's base rate while the satellites do whatever satellites do.
The test for core membership is simple. If an asset needs a bull market to survive, it is a satellite. If its investment case leans on the word narrative, it is a satellite. Solana holders will argue with me here, and maybe in a few years they will be right, but I would much rather promote an asset to the core late than early. The cost of being late is some missed upside. The cost of being early is finding out your foundation was another bet all along.
Size satellites so a zero is boring
Here is the sizing rule that makes the whole structure work. No single satellite should be big enough that a total loss costs more than 2 percent of the portfolio. Since altcoins can and do go to functional zero, the clean way to enforce that is a hard cap of 2 percent per position, measured at cost, and measured against the whole portfolio rather than the sleeve.
Run the arithmetic and a 20 percent sleeve gives you room for somewhere between six and ten positions at full size, fewer if you like keeping some dry powder in the sleeve, which I do. That is enough to cover a few themes you genuinely believe in without turning the sleeve into a museum of every narrative on the timeline. If you find yourself wanting position eleven, the honest move is to sell the weakest thesis you hold first, and it is remarkable how often that exercise talks you out of the new idea entirely.
The reason to size against zero instead of some polite 50 percent drawdown is that the historical record of altcoins is heavy with zeros. Teams dissolve, token unlocks dilute holders for years, narratives rotate and the liquidity leaves and never comes back. Historically, most altcoins that peak in one cycle never see those prices again, and you cannot know in advance which of your positions are the exceptions. The 2 percent cap means you never have to know. A position can go to nothing and the damage is a bad week and a mildly embarrassing story rather than a changed retirement date.
Sweep profits into the core, never sideways
The sweep rule is where the discipline actually lives, because sizing is easy on day one and hard after your first winner. Mine is simple. Once a month, on a calendar reminder, I check the sleeve. Any satellite that has grown past 4 percent of the portfolio gets trimmed back to 2, and every dollar of proceeds buys core. Any satellite whose thesis has broken gets sold entirely, and those proceeds buy core too. Profit flows one way, from satellites to core, and there is no branch of the flowchart where a win funds a new bet.
The failure mode this prevents is pyramiding, and it catches smart people constantly. A satellite triples and the gain feels like evidence of skill, so the profits roll into two new positions, and the sleeve quietly grows from 20 percent of the portfolio to 35 at exactly the point in the cycle when everything is expensive. Then the drawdown arrives, the oversized sleeve does the damage the structure existed to prevent, and the tripled position round-trips because nothing was ever actually sold. On paper everyone agrees this is obvious. In a live market, with a green portfolio, almost nobody executes it.
I will warn you that the sweep feels bad in real time, which is roughly how you know it is working. You are selling your best performer to buy the asset that does not move enough, and if the satellite keeps running after you trim, you will feel that too. Keep the frame straight, though. You still own a full-sized position in the winner. What you sold was the excess risk, and the core you bought with it is the part of the portfolio that will still be standing whenever the sleeve has a bad year.
The subtler failure mode runs the other direction. In a hot market the core starts to look like dead money, and people begin borrowing from it, just once, for a position that feels too obvious to size normally. That first raid is almost always followed by a second, and by the top of the cycle the 80/20 portfolio has become 50/50 without a single deliberate decision along the way. If you catch yourself calculating what the core could be earning in the hot sector of the month, that is the tell, and it is the moment to reread whatever you wrote down when you set the ratio.
The whole system on one page
- Pick the ratio, 80/20 by default, 90/10 if you are newer or nervous, and write it down somewhere you will see it.
- Core is BTC and ETH. It never funds satellites, whatever the market is doing.
- Every satellite gets a one-sentence thesis before you buy, and if you cannot write the sentence, you skip the position.
- Hard cap of 2 percent of the portfolio per satellite at cost, with no exceptions for conviction.
- Monthly sweep on a calendar reminder. Trim anything above 4 percent back to 2, sell anything with a broken thesis, and send all proceeds to core.
- Once a year, compare the sleeve's return to what the same money would have done sitting in core.
That last check is the humbling one. I track the sleeve against a plain BTC and ETH benchmark on a Blockcircle scorecard, and most years the honest reading is that the satellites earned their keep as tuition and entertainment more than as outperformance. Some years they win big, and that is the reason to keep them, but knowing the real number is what stops the sleeve from quietly taking over the portfolio.
Setting this up takes an afternoon. Maintaining it takes one reminder a month and one honest look a year. Pick the ratio, cap the satellites at 2 percent, sweep at 4, and let the boring part of the portfolio do the job you hired it for.