By Crypto Loop · Updated 2026-10-06T20:48:06.123Z
Loss first, size second
Position sizing is often described as a question of how much to buy or sell, but in practice it is a question of how much can be lost if the trade does not work. That framing matters because the entry price is only one part of the risk. A position can fail gradually, fail all at once, or fail in a way that bypasses the trader’s intended stop. If the sizing rule does not begin with the loss, it is easy to confuse exposure with control.
A practical risk budget starts by defining the maximum amount that can be lost on the idea under a normal execution assumption and under less favorable conditions. The first figure is the planned loss, usually measured from entry to stop. The second is the scenario loss, which allows for stop slippage, gap risk, and any extra costs that may appear when liquidity is thin. A position that looks acceptable under planned loss alone may be too large once scenario loss is considered.
This approach is more conservative than asking what percentage of capital feels comfortable, because comfort changes with market conditions and with recent wins or losses. A fixed comfort level can lead to oversized trades when volatility rises. A loss-based approach instead asks a narrower question: if the market moves against the idea, how much capital, and how quickly, is the trader willing to place at risk in this one setup and across the whole portfolio?
Build a risk budget before the trade exists
A risk budget is the amount of loss a trader can distribute across open and potential trades without exceeding a chosen limit. The budget can be framed at several levels. One level covers the single trade. Another covers related trades in the same asset, sector, theme, or direction. A broader level covers the entire portfolio under a market shock. The reason for separating these layers is that several small losses can combine into one large problem if they all depend on the same market condition.
For educational purposes, imagine a trader defines three internal limits. First, no single idea should lose more than a small fraction of account equity under planned execution. Second, several correlated ideas should not all fail at once for a combined amount that would be hard to recover. Third, the total drawdown from a plausible shock should remain within the trader’s tolerance for decision quality. These are not recommendations; they are examples of how risk budgets can be structured.
The key decision check is whether a new position fits within the remaining budget after existing exposure is counted. If the portfolio already contains positions that would likely weaken together, the new trade is not isolated even if it is in a different market. A risk budget that ignores overlap between ideas is only a partial budget. Good sizing asks not just how much the trade can lose, but how that loss interacts with the rest of the book.
Scenario loss is more informative than nominal stop distance
The nominal stop distance is the price difference between entry and the intended exit point. It is useful, but it is not the full loss estimate. In fast markets, the execution price may be worse than the stop price. Liquidity may be thin, the order book may move away, or trading may halt briefly. The resulting loss can exceed the nominal stop distance by a material amount. This is stop slippage, and it becomes more important as volatility and position size increase relative to market depth.
A scenario loss estimate tries to answer: if the stop is triggered in an unfavorable way, what is the likely realized loss? The estimate can include a small slippage buffer in liquid conditions and a larger buffer for assets or time windows where execution quality is less predictable. The point is not to forecast the exact fill, which is impossible, but to avoid pretending that a stop is a fixed price guarantee.
A practical check is to ask whether the trade remains acceptable if the stop is filled worse by a modest percentage or a few ticks, depending on the instrument. If a position only works when execution is perfect, it may be too fragile to justify the size. This is especially important around scheduled events, session opens, weekends, and thin liquidity periods, where the market can jump over the stop rather than trade through it.
Worked hypothetical arithmetic: from planned loss to scenario loss
Consider a hypothetical account with equity of 100,000 units. Suppose the trader sets an internal limit that the scenario loss on one idea should not exceed 1,000 units. Assume a trade idea has an entry at 50 and an intended stop at 48.50, so the nominal risk per unit is 1.50. Under a simple planned-loss calculation, the largest size would be 1,000 divided by 1.50, which equals about 666 units. That is only the first pass.
Now add a hypothetical slippage buffer. If the trader wants to allow for a worse exit at 48.20 rather than 48.50, the loss per unit becomes 1.80 instead of 1.50. Under the same 1,000-unit scenario-loss budget, the size falls to about 555 units. The difference is not trivial. It shows that the number of units is determined as much by execution risk as by the chart pattern or signal quality.
A further check is liquidity. If the average tradable depth is limited, a large order may move the market before the stop is even reached, increasing realized loss. In that case, the trader may decide that the scenario loss should be tested at a wider buffer or with a smaller size. The lesson of the arithmetic is simple: size is the quotient of risk budget and loss per unit, but loss per unit must include the possibility that the exit is not clean. Planning only with the intended stop can produce false confidence.
Correlation turns separate trades into one combined bet
Correlation matters because positions that appear distinct can still respond to the same driver. Two assets in the same risk regime may rise and fall together when liquidity contracts, when the broader market turns, or when a macro catalyst hits. A portfolio of many small positions can therefore behave like one concentrated position if the exposures are positively linked.
For position sizing, the question is not only how much each trade can lose in isolation, but how much the whole cluster can lose if the shared factor moves against it. If three trades each have a planned loss that appears tolerable on its own, the combined loss may become uncomfortable when all three stop out during the same move. This is especially relevant when trades are all long or all short in the same direction.
A useful decision check is to group positions by common sensitivity. These groups might share the same asset class, the same macro theme, the same liquidity profile, or the same event risk. Within each group, ask whether one adverse move could trigger several exits at once. If yes, the real risk budget should be assigned to the group, not only to the individual lines. The limit is not mathematical precision; it is avoiding the illusion that diversification exists where it does not.
Gap risk and why stops are not guarantees
Gap risk is the possibility that the market opens or trades through the stop level without offering a fill at the intended price. It can happen after news, overnight sessions, weekends, or any period when the market is not continuously available at the same depth. In such cases, a stop order may still execute, but the execution may be materially worse than expected. The loss is then governed by the gap, not by the stop line drawn on a chart.
This risk changes position sizing because a larger position amplifies the impact of a discontinuous move. A trader who sizes a position only by the visual distance to the stop may underestimate the true downside if the asset can gap. The more discontinuous the market, the more the trader should rely on scenario loss rather than nominal stop loss. That does not remove the risk; it simply makes the risk visible before entry.
The decision check here is to ask whether the trade is exposed to non-continuous pricing, scheduled announcements, or thin trading windows. If the answer is yes, the sizing assumption should be more conservative than for a continuously traded, deep market. In educational terms, a stop is a plan for action, not a promise of price. Position size should reflect that difference.
Common failure modes in sizing logic
One common failure mode is using account percentage alone without defining the underlying loss mechanics. Saying that a trade risks a small percentage is incomplete if the stop distance is unmeasured or if the market is prone to slippage. Another failure mode is ignoring the difference between a valid signal and a tradable signal. A setup can be technically acceptable and still be too risky to size because the execution conditions are poor.
A second failure mode is assuming that previous trades are independent. A series of positions entered for different reasons can still depend on the same market condition. When that condition changes, all of them can lose together. Another problem is adjusting size upward after a period of losses to compensate for frustration or to recover faster. That tends to increase the chance of compounding error rather than restoring control.
There is also the risk of stop placement that is too tight for the instrument’s normal volatility. In that case the trader may not be controlling risk at all; the position may simply be getting knocked out by routine noise. The trade then accumulates repeated small losses that are not informative. Good sizing and stop placement should be considered together. If the stop has to be unrealistically close to fit the chosen size, the better fix may be to reduce size or abandon the trade, not to force the same structure.
Practical decision checks before entry
A useful pre-trade checklist can be kept short. First, what is the planned loss if the stop is reached cleanly? Second, what is the scenario loss if exit is worse than intended? Third, how much of the current risk budget does the trade use on its own and within its correlated group? Fourth, is the market likely to gap, thin out, or become disorderly before the exit can be executed? Fifth, does the position remain acceptable if several other positions are hit at the same time?
These questions are designed to prevent a common mistake: sizing from conviction rather than from loss. Conviction can be useful in selecting ideas, but it is a weak basis for deciding exposure. The market does not reward confidence; it only reveals whether the risk was manageable after the fact. A trade that is compelling in theory may still be too large if the worst plausible fill would create an unacceptable drawdown.
A final check is whether the planned trade would still be acceptable after a partial portfolio shock. For example, if the account has already experienced a moderate drawdown, the same nominal position size may represent a much larger psychological burden and a higher chance of poor follow-through. Risk budgets should ideally adapt to both capital and decision quality. If the trader is likely to manage the trade badly after a loss, the economic risk is larger than the math suggests.
Limits, and a jurisdiction and risk caveat
Position sizing can reduce the scale of a mistake, but it cannot eliminate market risk, execution risk, or model error. Any stop can fail to protect at the intended level, especially in fast or illiquid conditions. Correlation can also change quickly, so a portfolio that looked diversified may not stay diversified when stress arrives. Because of those limits, sizing should be treated as a control mechanism, not as a guarantee of safety.
This material is educational and general in nature. Trading rules, margin treatment, tax consequences, product restrictions, and disclosure obligations vary by jurisdiction and by venue. Anyone applying these ideas should confirm the local legal and regulatory treatment of the instrument and execution method they use, and should assess whether the product is appropriate for their own circumstances. If the jurisdiction allows leveraged or derivative trading, that does not make the risk manageable by default; it only means the risk must be understood carefully.
The central message is straightforward: a position should be sized from the loss that can occur, not from the upside hoped for. Planned loss, scenario loss, correlation, gap risk, and stop slippage belong in the same decision. When those elements are measured before entry, the trade may still fail, but the failure is less likely to become a portfolio-level surprise.