Suppose you start with $10,000 split 60% BTC, 30% ETH and 10% USDC. One month later BTC is up 10% and ETH is up 45%. Without touching anything, your portfolio is now about 55% BTC, 36% ETH and 8% USDC. You did not decide to own more ETH; the market decided for you. Rebalancing is the rule you set in advance for when and how to pull the mix back to target. This article compares the two main rules, calendar and threshold, on the same illustrative price paths, and shows when each one helps and when it costs you.
Drift, calculated
Weight drift is simple arithmetic. Each holding’s new value is its starting value times its price change; its new weight is that value divided by the new total.
| Asset | Start value | Price change | Value after 1 month | New weight | Target |
|---|---|---|---|---|---|
| BTC | $6,000 | +10% | $6,600 | 55.2% | 60% |
| ETH | $3,000 | +45% | $4,350 | 36.4% | 30% |
| USDC | $1,000 | 0% | $1,000 | 8.4% | 10% |
| Total | $10,000 | $11,950 | 100% |
To get back to target you would need BTC at $7,170, ETH at $3,585 and USDC at $1,195. That means selling $765 of ETH, buying $570 of BTC and adding $195 to USDC. At a 0.1% trading fee on the two crypto legs ($1,335 traded), the direct fee is about $1.34. Spreads and slippage come on top, and on thin altcoin pairs they can be much larger than the fee.
The two rules
- Calendar rebalancing: on fixed dates (monthly, quarterly, yearly), reset to target regardless of how far weights have moved.
- Threshold (band) rebalancing: check regularly, but only trade when a weight moves outside a band, for example more than 5 percentage points from target.
- Hybrid: check on a schedule, and trade only if a band is breached at that check. This is the most common practical compromise.
For traditional multi-asset portfolios, Vanguard’s research on when and how to rebalance found annual rebalancing near optimal for many investors once trading costs are counted, with monthly or quarterly schedules too frequent. It also notes that pure threshold rebalancing needs close monitoring. Crypto is far more volatile than a stock-bond mix, so weights drift faster and bands get hit more often, but the underlying trade-off is the same: tighter control of risk versus more trades and costs.
Same portfolio, two price paths
Below are two illustrative three-month paths. Prices are given as multiples of the starting price. The calendar rule rebalances at the end of month 3 only. The threshold rule checks at each month end and rebalances when any weight is more than 5 points from target. Both pay 0.1% on crypto trades.
Path 1: rally, then give-back
| Month end | BTC | ETH | Weights if untouched (BTC/ETH/USDC) |
|---|---|---|---|
| 1 | 1.10× | 1.45× | 55.2 / 36.4 / 8.4 |
| 2 | 1.20× | 1.60× | 55.4 / 36.9 / 7.7 |
| 3 | 1.05× | 1.10× | 59.4 / 31.1 / 9.4 |
The threshold rule triggers at month 1 (ETH is 6.4 points over), sells $765 of ETH after its first jump, and is then within bands for the rest of the period. Results at month 3:
| Rule | Trades | Fees | Value at month 3 |
|---|---|---|---|
| Never rebalance | 0 | $0 | $10,600 |
| Calendar (quarterly) | 1 | $0.18 | $10,600 |
| Threshold (±5 points) | 1 | $1.34 | $10,758 |
Trimming ETH before it gave back most of its gain added about $158. That is the case people have in mind when they say rebalancing “sells high.”
Path 2: ETH keeps running
| Month end | BTC | ETH | Weights if untouched (BTC/ETH/USDC) |
|---|---|---|---|
| 1 | 1.10× | 1.45× | 55.2 / 36.4 / 8.4 |
| 2 | 1.15× | 1.90× | 50.7 / 41.9 / 7.4 |
| 3 | 1.20× | 2.30× | 47.7 / 45.7 / 6.6 |
| Rule | Trades | Fees | Value at month 3 |
|---|---|---|---|
| Never rebalance | 0 | $0 | $15,100 |
| Calendar (quarterly) | 1 | $4.23 | $15,096 |
| Threshold (±5 points) | 2 | $2.55 | $14,580 |
Here the threshold rule sold ETH twice on the way up and finished about $520 behind doing nothing. The untouched portfolio, meanwhile, ended with ETH at 45.7% instead of 30%: more return in this path, but a materially riskier portfolio than the one you chose.
What the two paths show
Rebalancing is a mean-reversion habit. It adds value when the asset you trimmed later falls back, and it costs you when that asset keeps trending. Neither rule wins in general; which one looks better depends on the path, and you do not know the path in advance. Our article on trend following vs mean reversion explains why the same trade-off appears in active trading. The real reason to rebalance is risk control: keeping the portfolio close to the allocation you decided you could live with.
Choosing your settings
- Band width: narrower bands mean more trades. In volatile assets, a 2-point band can trigger almost every week. Many investors use wider absolute bands (5 points) or relative bands (for example, 25% of the target weight, so a 30% target trades outside 22.5% to 37.5%).
- Check frequency: checking daily turns a threshold rule into something close to active trading. A monthly or quarterly check with bands is easier to follow consistently.
- Use cash flows first: directing new deposits into underweight assets can rebalance without selling anything.
- Partial rebalancing: trading back to the edge of the band, not all the way to target, reduces turnover.
Risks and costs people underestimate
- Taxes. In the U.S., the IRS treats virtual currency as property, and its virtual currency FAQ states that exchanging one virtual currency for another can produce a capital gain or loss. Selling ETH for BTC or USDC during a rebalance may therefore be a taxable event. Rules differ by country and situation; this is general information, not tax advice, so check with a qualified professional before choosing a frequency.
- Hidden trading costs. Fees, spreads, slippage and network withdrawal fees add up, especially on small portfolios and illiquid tokens. Track them as a line item, not as noise.
- Stablecoins are not cash. A stablecoin sleeve carries issuer, depeg and platform risk. Rebalancing into it reduces price exposure, not all exposure.
- Rebalancing into a collapse. If a token is failing (a hack, a broken peg, a delisting), a mechanical rule keeps buying it as its weight falls. Set an exception rule in advance, such as removing an asset from the target mix after a defined event, rather than improvising.
- Automated tools. Exchange “auto-invest” or rebalancing bots execute whatever rule you give them. Read exactly when they trade and which order type they use.
Run your own comparison
- Export monthly closes for each asset in your mix (most exchanges and charting sites allow CSV export).
- In a spreadsheet, track units held, value and weight per month for three versions: never rebalance, calendar, and threshold.
- Include a realistic cost per trade (fee plus an estimate for spread).
- Compare not just ending value but the highest weight any single asset reached and the largest peak-to-trough drop. Those are the risk numbers rebalancing is meant to control.
If you add leveraged or derivative positions to a portfolio, rebalancing logic changes because margin, not just weights, determines risk; see our guide to spot vs futures delta neutrality and to position sizing.
Financial disclaimer: This article is for educational purposes only and is not financial, investment, or trading advice. Trading involves substantial risk of loss. Example figures are illustrative.