KoinlyticsRebalancing back to a target allocation is a mechanical decision to sell whatever went up and buy whatever went down. That is the entire idea, and it is also exactly why it is not free: in a market that keeps trending in one direction, the discipline that protects you from over-concentration is the same discipline that caps your upside.
Two ways to trigger a rebalance
Calendar rebalancing resets the portfolio to target weights on a fixed schedule, monthly or quarterly being the most common. It is simple to automate and simple to explain, but it rebalances on a date whether or not the allocation has actually drifted much, which can mean trading (and paying fees and triggering tax events) to correct a drift of only one or two percentage points.
Threshold, or drift, rebalancing only trades once an asset's weight moves outside a band around its target, commonly 5 to 10 percentage points absolute, or a relative band such as 20% above or below the target weight. This trades less often in calm periods and reacts faster in volatile ones, since crypto can drift outside a 5-point band in days rather than months. The tradeoff is that it requires monitoring rather than a fixed date, and the band width itself is a judgment call: too narrow and you are trading constantly, too wide and the portfolio drifts a long way from target before anything happens.
The maths: rebalancing mechanically sells strength and buys weakness
Start with $10,000 split 50/50 between BTC and ETH, $5,000 each. Suppose BTC doubles and ETH is flat. The portfolio is now $10,000 of BTC and $5,000 of ETH, $15,000 total, with BTC at 66.7% and ETH at 33.3%. Rebalancing back to 50/50 means selling $2,500 of BTC and buying $2,500 of ETH, landing both legs at $7,500.
That trade locked in part of BTC's gain and added to ETH at a relatively lower price than BTC's, which is the mechanism by which rebalancing systematically sells strength and buys weakness. If the two assets subsequently mean-revert, meaning ETH's underperformance corrects and BTC's outperformance cools, the rebalanced portfolio ends up ahead of a static one, because it bought the underperformer while it was cheap relative to the other leg.
The honest counterargument: rebalancing loses money in a real trend
The entire benefit above depends on mean reversion between the two assets. If BTC keeps outperforming ETH for years rather than reverting, every rebalance was a sale of the winner and a purchase of the loser at a moment before either the winner kept winning or the loser kept losing. A static, never-rebalanced 50/50 portfolio would have ended up more BTC-heavy and, in that scenario, richer than the version that kept trimming BTC back to target.
This is not a hypothetical edge case. Multi-year, one-directional relative trends between major crypto assets have happened repeatedly. Rebalancing is a bet that relative performance between the assets you hold oscillates around some equilibrium rather than trending, and that bet is sometimes wrong for long stretches. Anyone recommending calendar rebalancing without naming this tradeoff is not describing the whole mechanism.
Two costs crypto investors pay that equity investors mostly do not
This is not tax advice, and what counts as a taxable event varies by jurisdiction and changes over time; check local rules before acting on anything below.
A rebalance inside a normal brokerage index fund is usually a free, same-day, no-fee internal trade, and inside a tax-advantaged account like an IRA or ISA it triggers no tax event at all. Neither of those is generally true in crypto.
- Gas and swap fees on every leg. A rebalance is at minimum one sell and one buy, each paying network gas (see gas fees) plus the swap fee or exchange spread on the trade itself, commonly in the tens of basis points on a DEX and comparable on a centralized exchange once spread is included. On a small portfolio rebalanced monthly, those costs compound: a $500 rebalance paying even 0.5% combined in gas and swap fee is $2.50 gone every time, before any tax.
- Every rebalance is typically a taxable disposal. In most jurisdictions that tax crypto as property or a capital asset, selling BTC to buy ETH is a disposal of the BTC, realizing a gain or loss at that moment, the same as if you had sold to fiat (see crypto tax basics). Continuing the earlier example, the $2,500 of BTC sold carried an approximate cost basis of $1,250, since that slice was originally $1,250 before BTC doubled, realizing roughly $1,250 of taxable gain on that single rebalance leg. At an illustrative 20% capital gains rate, that is $250 owed on a trade made purely to maintain a target weight, not because you chose to take profit.
Stack both costs across a year of monthly rebalancing and the combined drag from fees and realized tax can meaningfully outweigh the benefit rebalancing is supposed to provide, particularly on portfolios under roughly $10,000 to $20,000 where fixed gas costs are a larger fraction of each trade.
Practical steps
- Decide calendar or threshold rebalancing on purpose, and if using threshold, write down the band before you need it, not while staring at a drifted portfolio.
- Estimate combined gas and swap cost per rebalance before committing to a frequency; a wide band and infrequent rebalancing usually beats a tight schedule on a small portfolio.
- Track the cost basis on every leg of every rebalance, since each one is a separate disposal for tax purposes in most jurisdictions.
- Widen the acceptable band, or accept drift, during a period where you have a specific reason to expect one asset's outperformance to continue rather than revert.
Koinlytics shows realized profit and loss across wallets, which makes it easier to see the actual cost basis and gain on each rebalancing trade rather than reconstructing it from separate transaction histories.