Two numbers turn the answer into dollars
The ladder tells you what qualifies. It does not tell you how much. That is handled by two settings that never change from day to day, plus one threshold that decides whether a trade is worth making at all.
A floor that never moves, and a cap that never lifts
The floor is 33.33%. A third of the book stays in the dominant major no matter how many mid caps are winning. That third is the part of the account that is not allowed to chase.
The satellite cap is 25%. SUI, XRP, DOGE and HYPE can never hold more than a quarter of the book each. SOL and BNB are treated as core and have no cap of their own beyond what is left over.
The remaining two thirds is split equally between whichever coins qualified. If a satellite's equal share would land above 25%, it is cut back to 25% and the leftover flows to any qualifying core coin. If there is no qualifying core coin to take it, the leftover goes back into the major. That is the whole calculation.
The floor is a minimum, not a maximum. On a day when nothing qualifies, the major holds the entire book rather than a third of it.
The same rule, on a $100,000 book
No mid cap is beating ETH, so there is nothing to split the book with. The whole account sits in the major: $100,000 of ETH. This is the single most common outcome inside crypto.
The floor takes $33,330. The other $66,670 splits evenly between the two winners, $33,335 each. Neither is a satellite, so the 25% cap does not apply and the book ends up in equal thirds.
An even split of the free two thirds would hand each of them 33.33%, but both are satellites and stop at 25%, so $25,000 each. The $16,670 left over has no qualifying core coin to move to, so it returns to the major and ETH ends the day at $50,000.
The cap is doing exactly what it exists for: the two riskiest names in the list cannot take over the account just because they had a good week.
The free two thirds divided four ways is 16.67% each, comfortably under the satellite cap, so nothing is trimmed. More winners means smaller positions, not a bigger bet. That is $33,330 in BTC and $16,667 in each of the four.
When crypto is out
The mid cap layer switches off entirely and the book holds one thing. There is no blending and no partial exit.
| Reading | Crypto | Gold | S&P 500 | Where $100,000 sits |
|---|---|---|---|---|
| Example 5 | Down | Up | Up | $100,000 gold |
| Example 6 | Down | Down | Up | $100,000 S&P 500 |
| Example 7 | Down | Down | Down | $100,000 cash, euro or dollar |
Gold is checked before the S&P, so example 5 holds gold even though shares are also trending up. The ladder never asks a lower question once a higher one has answered.
The rule that stops you trading for nothing
Prices move, so even on a day when the ladder says exactly what it said yesterday, your holdings have drifted away from the targets. Fixing that drift every single day would cost more in fees than it earns. So there is a threshold: a leg is only traded when it is at least 2% of the whole book away from its target.
| Leg | Target | You hold | Gap | Gap as % of book | Action |
|---|---|---|---|---|---|
| BTC | $36,000 | $35,500 | $500 | 0.46% | Hold |
| SOL | $36,000 | $37,000 | $1,000 | 0.93% | Hold |
| BNB | $36,000 | $35,500 | $500 | 0.46% | Hold |
Example 8, a $108,000 book with unchanged targets. Every gap is under 2%, so the day is a no-op. Most days look like this.
| Leg | Target | You hold | Gap | Gap as % of book | Action |
|---|---|---|---|---|---|
| BTC | $36,000 | $32,000 | $4,000 | 3.70% | Buy $4,000 |
| SOL | $36,000 | $44,000 | $8,000 | 7.41% | Sell $8,000 |
| BNB | $36,000 | $32,000 | $4,000 | 3.70% | Buy $4,000 |
Example 9, the same book after SOL rallied hard. All three gaps clear the threshold, so all three trade. Total traded is $16,000.
Example 9 is the part that feels wrong and is right. Selling the coin that is working to buy the two that are not is the mechanism that keeps one lucky position from becoming the entire account before it turns.
Trading is not free and the system trades often enough to notice
Every rebalance pays a fee on both sides, and the gold, S&P and euro legs are perpetual contracts that also charge or pay funding for as long as they are held. Over years those costs add up to a real share of the result. The 2% band exists to keep them from adding up faster than they need to.