Bot Capital Allocation: How to Size and Split Funds

November 20, 202614 MIN2 views
Bot Capital Allocation: How to Size and Split Funds

Bot capital allocation is the set of rules that decides how much of your account a trading bot may use, how that amount is divided across bots, pairs and strategies, and how much any single trade may put at risk. A sound allocation is written down before the bot starts and enforced by the bot, not adjusted by feel after a bad day. Darkbot supports this approach with multiple simultaneous bots, paper trading, backtesting, risk management tools and automated rebalancing, so each rule can be set once and executed the same way every time.

Darkbot
Put Your Allocation Rules to Work
Darkbot supports multiple bots, backtesting, paper trading and automated rebalancing to help apply your allocation approach consistently.
  • ✓Multiple simultaneous bots
  • ✓Backtesting
  • ✓Paper trading
  • ✓Automated rebalancing
Start free trial

TL;DR

  • Allocate in three layers: total capital the bot may touch, capital per bot or strategy, and risk per trade. Set them in that order.
  • Keep the bot’s budget separate from money you may need, and keep it smaller than the account it trades on.
  • Size each trade by what you can lose if it fails (the distance to your stop), not by how much the account could buy.
  • Rebalancing returns a portfolio to its target weights. Decide in advance whether it runs on a schedule or on a drift threshold.
  • Test the allocation with backtesting and paper trading before live funds are committed.

What bot capital allocation actually means

A trading bot executes rules. Capital allocation is the first rule, and it sits above every entry and exit signal. It answers three separate questions:

  1. Exposure: how much of your total crypto holdings is the bot allowed to trade?
  2. Distribution: if you run more than one bot or strategy, how is that budget split?
  3. Trade size: how much capital, and how much loss, does a single position carry?

These are different decisions and mixing them is the most common source of confusion. A bot with a well-tuned strategy and an oversized position can still cause a loss the strategy was never designed to absorb. The reverse also holds: an average strategy with disciplined sizing keeps losses within a range you chose in advance.

The point of automation here is consistency. A bot applies the same sizing rule on the hundredth trade as on the first, at 3 a.m. as at noon. That removes one of the main failure modes of manual trading, which is changing size in response to the last result.

Layer one: how much capital the bot gets

Start with a number you can state in dollars, not a percentage of whatever the account happens to hold. Three practical constraints shape it.

Separate the budget from needed money. Funds for rent, tuition, taxes or an emergency reserve are not bot capital. Crypto prices can move sharply in a day, and a system that must be right to keep you solvent is under pressure no rule set can remove.

Keep the budget below the exchange balance. Darkbot connects to supported exchanges through API keys, and the bot trades within the account those keys reach. A bot can only commit what the account holds, so the cleanest way to cap exposure is to hold only the intended budget on the account the keys are linked to, or to use a dedicated sub-account where the exchange offers one.

Choose a budget you can lose without changing behavior. If a 20% drawdown on the allocation would make you switch the bot off, the allocation is too large for your tolerance. The right size is the one at which you would leave the rules alone through a drawdown.

Small accounts

For accounts in the $100 to $1,000 range, the allocation problem is different. Fixed costs matter more: exchange minimum order sizes, trading fees and spreads take a larger share of each trade. Fewer, larger-than-minimum positions across one or two pairs usually work better than many tiny positions across ten. Check the minimum order size on your exchange for each pair before choosing how many positions the budget supports.

A simple check: divide the bot budget by the number of positions the strategy can hold at once. If the result is near or below the exchange minimum, reduce the number of concurrent positions or the number of pairs, not the discipline of the sizing rule.

Layer two: splitting capital across bots and strategies

Running several bots does not diversify anything if they all hold the same risk. Two bots trading correlated large-cap pairs with similar logic behave largely as one position. When you divide capital, ask what each bot is exposed to and what could hurt it.

A workable split follows the function of each strategy:

Role Typical logic Allocation consideration
Core, rule-based Rebalancing or grid-style execution on liquid pairs Largest share of the bot budget, because the position sizes are small and defined
Trend or swing Entries on confirmed direction, defined stop Medium share, sized by stop distance
Experimental New logic still in paper trading or small live trials Smallest share, capped so that a total loss does not matter

The percentages are yours to set, and no table can set them for you. What matters is that each bot has its own cap, that the caps sum to no more than the bot budget, and that the experimental bot is sized so that failure is an acceptable outcome.

Correlation and concentration

Crypto assets frequently move in the same direction at the same time, so holding many coins does not by itself spread risk. Look at concentration by exposure: how much of the bot budget is tied to one asset, one exchange, one quote currency or one strategy type. Set a ceiling for each. A common structure is a maximum share of the budget per asset and a maximum share per bot, applied together.

If you also hold assets outside crypto, the same logic extends upward. Some investors allocate part of a wider portfolio to physical precious metals as a separate sleeve, a topic covered in this overview of reducing portfolio risk with precious metals. The bot budget then becomes one defined bucket within a larger plan, not the whole of it.

Multiple accounts

Traders who run several exchange accounts, for market making, arbitrage or hedging, add a fourth constraint: capital sitting on each venue. Balances on each exchange must cover that account’s positions with margin to spare, since a bot that runs out of funds on one side of a pair cannot complete the strategy. Track the allocation per account, not only the total.

Layer three: how much each trade puts at risk

Position sizing is where allocation becomes arithmetic. The most widely used framework is fixed fractional risk: choose the percentage of the bot budget you are willing to lose on one trade, then size the position so that hitting your stop produces exactly that loss.

The formula:

Position size = (budget × risk per trade) ÷ (entry price − stop price)

Worked example

Suppose the bot budget is $1,000 and the risk per trade is 1%, which is $10.

  • A trade enters at $100 with a stop at $95, a $5 distance per unit.
  • Position size = $10 ÷ $5 = 2 units, a $200 position.
  • If the stop triggers, the loss is about $10, or 1% of the budget, before fees and slippage.

Now widen the stop to $90, a $10 distance. The position size falls to 1 unit, a $100 position. The risk is unchanged at $10. Wider stops mean smaller positions, which is the cause-and-effect relationship that keeps risk constant across volatile and quiet conditions.

Two cautions apply. First, a stop is not a guarantee of the exit price: in fast markets, fills can occur beyond the stop level (slippage), so real losses can exceed the planned figure. Second, fees and spreads add to the cost of every round trip. Build both into the estimate, particularly on small accounts.

Choosing the percentage

Many traders keep per-trade risk to one or two percent of the account. That range gives a long sequence of consecutive losses room to occur without ending the strategy: at 1% risk, ten losses in a row cost roughly a tenth of the budget; at 5%, the same sequence costs far more than a third once compounding is included. Our guide to the 1-2% rule and a risk checklist walks through the same calculation with a checklist.

Drawdown comparison after ten consecutive losses

For micro-traders risking $5 to $50 per trade, the percentage approach and the dollar approach converge: $5 to $50 is the risk figure, and the budget should be large enough that this stays within one or two percent. If it does not, the budget is too small for that risk figure, and the risk figure should come down.

Portfolio-level limits

Per-trade risk is not enough when several bots can lose simultaneously. Add two portfolio-level limits:

  • Total open risk: the sum of the planned losses on all open positions, capped at a set percentage of the budget.
  • Drawdown limit: a maximum loss from the budget’s peak value at which the bots pause and you review the rules.

The drawdown limit is a decision made in advance, when you are calm, about what happens when you are not. Our overview of risk management in crypto trading covers how these limits fit with automated execution.

Rebalancing as an allocation rule

Allocation drifts. If one asset in a target mix rises faster than the others, it grows beyond its intended weight and the portfolio carries more risk in that asset than you planned. Rebalancing is the rule that restores the original weights. The U.S. Securities and Exchange Commission’s investor education office describes rebalancing as bringing a portfolio back to its target asset allocation, and the same logic applies to a crypto basket.

Two trigger types exist:

  • Time-based: rebalance on a fixed schedule, such as weekly or monthly. It is simple and predictable.
  • Threshold-based: rebalance when an asset’s weight departs from its target by a set margin, such as five percentage points. It responds to volatility, and it trades only when drift is meaningful.

Each rebalance generates trades, and trades cost fees and may create taxable events. A wider threshold or longer interval means fewer trades and more drift. The right setting balances those two costs for your account size. Darkbot’s rebalancing is automated, so the chosen rule is applied without manual intervention. For the mechanics, see our guides on what rebalancing is in crypto and portfolio balancing for automated trading.

Tax treatment of crypto trades depends on your jurisdiction. In the United States, consult a tax professional about how disposals during rebalancing are reported.

Testing an allocation before it goes live

An allocation is a hypothesis about how much loss you can tolerate and how your strategy behaves. Test it in stages.

  1. Backtest. Run the strategy on historical data with the exact sizing rule. Look at the maximum drawdown and the longest losing streak, not only the final return. Historical results do not predict future ones, but they show whether the sizing rule could survive conditions the market has already produced.
  2. Paper trade. Run the bot on live prices without real funds. This exposes problems backtests hide: order timing, partial fills, and how the strategy behaves in current conditions.
  3. Start small. Go live with a fraction of the intended budget. Increase it only after the bot has operated through varied conditions within your risk limits.
  4. Review on a schedule. Check allocation against results monthly or quarterly, and change one variable at a time so you can tell what caused any difference.

Both backtesting and paper trading are part of the Darkbot platform, so the whole sequence can be run in one place. Our strategy optimization guide covers how to adjust parameters without overfitting them to past data.

Common allocation mistakes

  • Sizing by account balance instead of risk. Buying the largest position the account allows makes the stop distance irrelevant to the loss.
  • Raising size after wins. This scales risk with recent luck, not with evidence.
  • Running identical bots on correlated assets. The count of bots rises while diversification does not.
  • No ceiling on total open risk. Each trade looks acceptable while the combined exposure is not.
  • Funding the bot from needed money. Pressure to recover losses leads to overriding the rules.
  • Skipping the test stages. A sizing rule that has not met a losing streak in testing will meet one live.

A starting framework by account size

The table shows structure, not recommendations. The figures are illustrations of the method for you to adjust to your own tolerance.

Account type Bot budget Risk per trade Concurrent bots
Nano-capital, $100 to $1,000 Amount you can afford to lose entirely 1% to 2% of budget, checked against exchange minimums One or two
Side-hustle or part-time Fixed monthly contribution, kept separate from living costs 1% of budget Two to three, different roles
Larger or multi-account Per-account caps plus an overall cap 0.5% to 1% per trade, total open risk capped Several, each with its own limit

Where Darkbot fits

Darkbot is an AI-based automation platform for crypto trading, not a signal provider. The AI component is used for consistency and probabilistic pattern evaluation within rules you set, and it makes no prediction and offers no assurance of profit. Allocation rules are the layer that sits above it, and they are yours.

The platform supports multiple simultaneous bots, strategy customization, risk management tools, portfolio management, automated rebalancing, backtesting and paper trading, with exchange access through API keys. There is a Free plan with limited features, a Standard plan at $12.50 per month and a Premium plan at $25.00 per month, and an Enterprise tier. A 14-day money-back guarantee applies. Current plan details are on the Darkbot pricing page.

Darkbot

Starting on the Free plan lets you set the budget, risk and rebalancing rules in paper trading first, then move to a paid plan when you need more bots or features. You can also read our guides on automated portfolio management and crypto portfolio management.

Our perspective

Most allocation problems we see described are not strategy problems. They are sizing problems: a reasonable rule set, run with a position that was too large for the stop, or a budget that was too large for the person.

That is why we treat allocation as the first decision and the signal logic as the second. A bot’s value is that it applies a rule the same way every time. If the rule is “risk one percent per trade, cap total open risk, rebalance at a set threshold,” the bot carries that out without fatigue or second-guessing.

Darkbot is built around that idea: systematic execution, structured risk control and repeatable logic, with a human setting the limits and keeping oversight. Multiple simultaneous bots let you give each strategy its own cap, and backtesting and paper trading let you examine an allocation before it holds real funds.

We would rather you start with a small, defined budget and a written rule than with a large one and an intention to be careful.

Grisha

FAQ

How much money should I give a trading bot?

Give it an amount you can lose entirely without changing your finances or your decisions. Keep it separate from money needed for living costs, taxes or emergencies. Start with a fraction of that amount, and raise it only after the bot has run within your risk limits across varied conditions.

What percentage of capital should a bot risk per trade?

Many traders risk one to two percent of the bot budget per trade, sizing the position from the distance to the stop. At that level a run of consecutive losses reduces the budget gradually. Smaller accounts must also check that the resulting position clears the exchange’s minimum order size.

Is it better to run one bot or several?

Several bots help only if they differ in logic, assets or timeframe, and each has its own cap. Several bots trading the same correlated assets behave like one larger position. Set a ceiling on total open risk across all of them.

How often should a bot rebalance a portfolio?

There is no universal interval. A time-based rule, such as monthly, is simple, while a drift threshold trades only when weights move materially. More frequent rebalancing keeps weights closer to target but generates more trades and fees, so match the setting to your account size.

Grisha Chasovskih
Written by

Founder & CEO, Darkbot

More articles

Start trading on Darkbot with ease

Come and explore our crypto trading platform by connecting your free account!

Start Free Trial

Free plan available • No credit card required

Contents

Free access for 7 days

Full-access to Darkbot Premium plan

Start now

Free plan available • No credit card required

Cryptocurrency trading involves substantial risk of loss. Past performance does not guarantee future results. Articles are for informational purposes only and do not constitute financial advice.