Tax is not a percentage of every sale
For a US investor holding a digital asset as a capital asset, the IRS distinguishes gains and losses and the holding period. Its current guidance uses one year or less for short-term treatment and more than one year for long-term treatment. A crypto-to-crypto exchange can also be a disposition. This is US federal context, not a rule for every country or account.
A transparent hypothetical ledger
Assume sale proceeds of 5,000 and an adjusted basis allocated to the sold units of 3,000. Before any further adjustments, the gain is 2,000. An invented 25% applicable rate would produce 500 of tax on that gain, not 1,250 on the full proceeds. If separate execution costs were 20 and were not already included in the amounts used for tax, the cash budget would also need to account for those costs. Their tax treatment must be reconciled separately to avoid double counting.
Why a single rate is insufficient
Actual liability depends on applicable law, income, losses, account treatment and the tax lots disposed of. A marginal bracket is not an effective rate on all holdings. The 25% figure above is only an arithmetic input; it is not a statement of a current tax bracket or a personal liability estimate.
A sale at a loss does not create a gain merely because it was a rebalance. Receiving staking or other income can raise separate tax questions; using that income to buy an underweight asset does not erase the original income event.
Build the records before using a calculator
- Identify the jurisdiction and account type.
- Reconcile units, acquisition dates and adjusted basis to records.
- Identify the lots and the permitted identification method.
- Estimate proceeds, costs and gains or losses separately.
- Apply supported tax assumptions and record what remains uncertain.
An allocation calculator can compute trade sizes. Unless it explicitly models tax lots and the relevant rules, it does not optimize an after-tax outcome. A genuine comparison of holding and rebalancing would require the same starting lots and a consistent treatment of eventual liquidation; deferring a gain is not necessarily eliminating tax.
Sources and scope
References support the stated definitions or research context. The worked examples on this page use invented inputs and do not reproduce a market study. Source checks: 16 October 2026.
Continue with a practical tool
Revision published . Historical author labels are explained on the About page.
This is something I’ve been pointing out to clients for years — those rebalancing tools give you a false sense of precision because they’re optimizing for pre-tax allocations. A 37% haircut on short-term gains can completely negate the benefit of rebalancing back to target, especially if your drift is modest to begin with. Really glad someone is finally putting this in writing clearly.
Thank you for sharing that perspective from your client work — “false sense of precision” is exactly the right way to frame it, and the point about modest drift making the math even harder to justify is one I probably could have emphasized more strongly in the piece.