A risk plan and journal for a volatile asset
Lesson 24 · about 10 min
The risk management course gives you the general plan: fixed-fractional sizing, stops, R-multiples, portfolio heat, daily loss limits and a written one-page document. This lesson adapts it to an asset that moves three to five times as much as an equity index, trades all night, and sits on venues that can vanish. Nothing here replaces the general plan; it adds the crypto-specific lines.
Start from the volatility
BTC's typical daily range is 3% to 4%; large alts 5% to 8%; small alts anything. The general rule that a stop should sit outside normal noise means crypto stops are wide in percentage terms, and the sizing formula turns a wide stop into a small position automatically. Accept that. A trader who keeps equity-sized 1% stops on BTC gets stopped by noise all day; a trader who keeps equity-sized positions with crypto-sized stops blows the risk budget on the first loss.
A useful calibration: measure the average true range (ATR) over 14 days on your timeframe, and set stops at some multiple of it (1.5x to 2x is common for swing trades). Then size from the stop. If the resulting position feels too small to matter, the asset is too volatile for the account at that stop, not the other way round.
The crypto-specific lines
Add these to the one-page plan:
Per-trade risk. 0.5% to 1% of the trading account, as in the general plan. For alts, the lower end, because slippage on the stop is larger and less predictable.
Slippage allowance. Add expected entry and exit slippage to the stop distance before sizing (Module 3). State the allowance per asset class: perhaps 0.1% for BTC, 0.5% for large alts, 2% for small ones.
Depth cap. No alt position larger than 5% of the 2% bid depth on your exchange.
Correlation cap. Count all crypto positions as one bet. Total open crypto risk (sum of dollars at risk across all positions) capped at 3% of the account, because BTC and alts move together.
Leverage cap. Maximum leverage, if perps are allowed at all: 3x. Isolated margin only. Stop distance never more than a third of the liquidation distance. Funding included in trade cost for any hold over 24 hours.
Exchange exposure cap. No more than a stated fraction of total crypto (say 25%) on any single exchange; the rest in self-custody. Monthly withdrawal test on every exchange.
Time rules. No new positions between Friday 20:00 and Monday 08:00 local time; no new positions in the hour around major scheduled macro releases; a defined stop-trading time each day.
Event rules. No leveraged position held through a token's unlock cliff, a scheduled network upgrade, or a known exchange maintenance window.
Daily and weekly loss limits. Stop for the day at 2R lost; stop for the week at 5R lost. Crypto's continuous session makes it easy to keep going; the limit is the closing bell you do not otherwise have.
Drawdown rule. If the trading account is down 15% from its high, halve position size until it recovers to within 5% of the high. Three 75%+ bear markets in twelve years say you will need this line.
Key idea: Crypto's volatility, continuous hours and venue risk each get an explicit line in the plan: wider stops with smaller sizes, slippage and depth caps, a leverage ceiling, an exchange exposure cap, time and event rules, and loss limits that act as your closing bell.
The journal
A crypto journal needs everything a general trade log has (entry, stop, size, exit, R-multiple, setup, notes) plus fields that catch crypto-specific errors:
| Field | Why |
|---|---|
| Venue and pair | Slippage and fees differ by exchange; you will see patterns |
| Planned vs actual slippage | Tells you whether your allowances are honest |
| Funding paid/received | Makes the cost of carry visible per trade |
| Leverage used | Notional ÷ equity, computed, not the slider setting |
| Time of day (UTC) and weekday | Finds the sessions where you lose |
| Hours since last sleep | Blunt, and the most predictive field for many traders |
| Source of the idea | Your analysis, a feed, a group; grade each source over time |
| Tax lot noted | Confirms the trade is in the records system |
Review weekly: sort by R-multiple and by source. Most traders find within a month that one venue, one session or one idea source accounts for most of the losses, and that removing it is worth more than any new setup.
Sanity
The always-on market and the culture around it push toward constant activity, constant comparison and constant fear of missing out. A few rules that protect judgement:
- Trade from the plan, not from the feed. Mute accounts whose posts make you want to open a position.
- Set price alerts and close the chart. Watching worsens outcomes.
- Measure yourself in R over rolling 50-trade windows, never against what someone posted.
- Keep the trading account small enough that a total loss changes nothing about your life. Scale up only on evidence, per the general plan.
- Take days off. The market will be there; the drawdown from tired decisions will be too.
The plan is the product
Thin books, leverage, funding, unlocks, exchanges, scams, taxes and your own fatigue have all appeared in this course, and every one has a line in the plan above that limits it. Write the plan, keep the journal, and review both monthly. The traders still here after several cycles are, almost without exception, the ones who did.
Try it: Copy the crypto-specific lines into your one-page risk plan with numbers filled in for your account. Then set up the journal fields in whatever tool you use, and back-fill the last ten trades. Sort by source and by hours since sleep, and write one sentence on what you see.
Recap
- Size from a volatility-based stop; wide stops with small positions are correct for crypto, not a compromise.
- Add slippage allowances, a depth cap, a correlation cap on total crypto risk, and a low leverage ceiling with isolated margin.
- Cap exposure per exchange, run withdrawal tests, and set time, event and loss-limit rules to replace the missing closing bell.
- Journal venue, slippage, funding, real leverage, time of day, sleep and idea source; review weekly by R and by source.
- Trade from the plan, not the feed; keep the account small enough that a total loss is survivable, and scale up only on evidence.
See it drawn
Original diagrams for the ideas on this page. Illustrative, not real market data.