Quick answer
Position sizing is deciding how many shares to buy based on how much money you’re willing to lose on the trade, not how much you want to invest. The formula: divide your dollar risk (a small percentage of your account) by the per-share distance between your entry price and your stop loss. Get this wrong and even a genuinely good strategy can wipe out an account through a normal losing streak. Get it right, and a mediocre win rate can still be profitable.
| Risk per trade | Capital remaining after 10 straight losses | Capital lost |
|---|---|---|
| 10% | 34.9% | 65.1% |
| 5% | 59.9% | 40.1% |
| 2% | 81.7% | 18.3% |
| 1% | 90.4% | 9.6% |
Most beginners spend their time picking what to buy and almost no time deciding how much of it to buy. That’s backwards. A trader with an average strategy and disciplined position sizing will outlast a trader with a great strategy and no sizing discipline, because the second trader only needs one bad stretch to be finished.
Why this matters more than picking winners
Nobody knows in advance which trade will win and which will lose. That’s not pessimism, it’s just how probability works even with a genuinely good strategy. A strategy that wins 60% of the time doesn’t produce six wins followed by four losses in a neat, predictable order. It produces streaks, sometimes six losses in a row before a single win shows up, purely from normal variance, the same reason flipping a fair coin ten times rarely lands exactly five heads and five tails.
That streak is the entire reason position sizing exists. It’s not there to make winning trades bigger. It’s there to make sure a normal losing streak, the kind that happens to every strategy eventually, doesn’t end the account before the strategy’s actual edge has a chance to play out over enough trades.
What a bad losing streak actually costs you
The table in the quick answer above isn’t a scare tactic, it’s just compounding math. Once a chunk of capital is lost, every subsequent trade is risking a percentage of a smaller and smaller base, but the required recovery works against you in the opposite direction.
| Loss taken | Gain required just to break even |
|---|---|
| 10% | 11.1% |
| 20% | 25% |
| 30% | 42.9% |
| 50% | 100% |
| 75% | 300% |
This asymmetry is the real danger of risking too much per trade. A 50% drawdown doesn’t need a 50% gain to recover, it needs a 100% gain, doubling the remaining capital, just to get back to even. Risking 10% per trade sounds modest until a normal losing streak turns into a hole that takes a genuinely extraordinary run of wins to climb out of. Risking 1-2% per trade, the same losing streak barely dents the account, and the strategy gets to keep operating long enough for its actual edge to show up.
The same sequence of trades, two different outcomes
Here’s what makes this concrete: the exact same sequence of wins and losses, from the exact same strategy, produces very different account outcomes depending only on position sizing. Ten trades, four winners and six losers in this order: loss, loss, win, loss, win, win, loss, loss, win, loss, each win paying 1.5 times the risk, starting from a $10,000 account.
| Trade | Result | Balance at 2% risk | Balance at 10% risk |
|---|---|---|---|
| 1 | Loss | $9,800 | $9,000 |
| 2 | Loss | $9,604 | $8,100 |
| 3 | Win | $9,892 | $9,315 |
| 4 | Loss | $9,694 | $8,384 |
| 5 | Win | $9,985 | $9,641 |
| 6 | Win | $10,285 | $11,087 |
| 7 | Loss | $10,079 | $9,978 |
| 8 | Loss | $9,877 | $8,981 |
| 9 | Win | $10,174 | $10,328 |
| 10 | Loss | $9,970 | $9,295 |
Same strategy, same ten outcomes, same order. The 2% account ends the sequence essentially flat, down $30. The 10% account ends down $705, more than 20 times the damage, from identical trades. Nothing about the strategy changed between the two columns. Only the sizing did.
Risk of ruin: why the math gets worse than intuition suggests
There’s a specific, well-known concept in risk management called risk of ruin, the probability that a losing streak, entirely possible even for a genuinely profitable strategy, does severe enough damage to end the account before the strategy’s real edge has time to play out.
To make this concrete rather than abstract, running a simple simulation, a strategy with a 50% win rate and a 1.5-to-1 reward-to-risk ratio (a reasonably realistic, not cherry-picked, profile), over 100 trades, checking how often a 50% drawdown occurs at different position sizing levels:
| Risk per trade | Chance of a 50% drawdown within 100 trades |
|---|---|
| 10% | ~16% |
| 5% | ~1.5% |
| 2% | effectively 0% |
| 1% | effectively 0% |
This is with a strategy that has a real, positive edge. The risk of a devastating drawdown isn’t coming from a bad strategy in this simulation, it’s coming entirely from sizing too aggressively relative to a perfectly reasonable win rate and payoff ratio. At 10% risk per trade, roughly 1 in 6 traders running this exact strategy would still see their account cut in half within their first 100 trades, purely from normal variance. At 2% or below, that risk essentially disappears, using the same underlying edge.
The position sizing formula
Once the amount you’re willing to risk in dollars is set, the formula for how many shares to buy is straightforward:
Shares to buy = (Account size × risk percentage) ÷ (entry price − stop-loss price)
The numerator is your dollar risk, a small, fixed percentage of total capital. The denominator is your per-share risk, the distance between where you’re buying and where you’ve decided you’re wrong. Dividing one by the other gives you the position size that keeps your dollar risk constant no matter how tight or wide that stop distance happens to be.
A worked example
Say the account is $10,000, and the rule is to risk 1% per trade, $100.
The stock trades at $45. Based on the chart, a sensible stop sits at $42, just below a recent support level, meaning $3 of risk per share.
Shares to buy = $100 ÷ $3 = 33 shares.
That 33-share position costs $1,485 to open (33 × $45), meaning roughly 15% of the account is invested in this one position, while only 1% of the account is actually at risk if the stop gets hit. That gap, invested amount versus amount at risk, is the entire point of the formula. Position size and risk size are not the same number, and confusing the two is one of the most common mistakes beginners make.
What happens when the stop is wider or tighter
The formula automatically adjusts position size based on how far away the stop needs to be, which is exactly the behavior you want.
| Stop distance | Shares bought (same $100 risk) | Position value at $45/share |
|---|---|---|
| $1 (tight stop) | 100 shares | $4,500 |
| $3 (moderate stop) | 33 shares | $1,485 |
| $6 (wide stop) | 16 shares | $720 |
| $10 (very wide stop) | 10 shares | $450 |
A tighter stop lets you buy more shares for the same dollar risk, since each share is risking less. A wider stop means fewer shares for the same risk. Beginners who skip this calculation tend to buy a round, comfortable-looking number of shares regardless of where the stop sits, which means their actual dollar risk swings wildly from trade to trade even though it feels consistent.
Capping total exposure, not just per-trade risk
Per-trade risk isn’t the only number worth capping. If five different positions are all open at once, each risking 2%, the account has 10% of its capital genuinely at risk simultaneously, even though no single trade looks aggressive in isolation.
A common guideline: cap total risk across every open position at somewhere around 5-6% of the account. If each trade risks 1%, that allows roughly five or six positions open at once before new ones need to wait for an existing position to close. If each trade risks 2%, that same total cap only allows two or three positions open concurrently.
This matters most during periods when several open positions move against you at the same time, which happens more often than beginners expect, since stocks in similar sectors tend to move together during a broad market pullback. Treating each position’s risk as fully independent from every other open position is a common and expensive miscalculation.
When the math tells you not to take the trade
Sometimes the formula produces an answer that doesn’t make sense in practice; a stock trading at $2 with a sensible stop at $1.50 only has $0.50 of risk per share, which at a $100 risk budget suggests 200 shares, a $400 position, technically fine, but the same calculation on a $150 stock with a $3 stop produces just 33 shares for a $4,950 position, more than a third of a $10,000 account in one name.
When the position size the formula produces would concentrate an outsized share of the account into a single stock, that’s the math telling you the trade doesn’t fit your account size at that risk level, not a reason to override the formula and buy more anyway. The honest options are: skip the trade, accept a smaller risk percentage for that specific trade, or wait for a setup with a tighter, more reasonable stop distance.
Common mistakes
Deciding position size before deciding the stop. The stop-loss distance has to come first, from the chart, not from a comfortable-feeling number of shares. Buying a round number of shares and then figuring out where to place a stop gets the entire formula backwards.
Confusing position size with position risk. A $5,000 position and a $50 position can carry identical dollar risk if their stop distances differ enough. Beginners who only look at how much money is invested, not how much is actually at risk, consistently misjudge how exposed they really are.
Increasing risk percentage after a losing streak to “win it back faster.” This is exactly backwards. A losing streak is the moment position sizing is supposed to protect the account, not the moment to abandon it.
Ignoring total exposure across multiple open positions. Five trades each individually risking a reasonable 1% can still add up to a genuinely risky 5% all moving against the account at once, especially during a broad market decline when unrelated stocks start behaving like related ones.
Treating the position sizing formula as optional for “high conviction” trades. The trades that feel most obviously correct are exactly the ones beginners tend to oversize, and exactly the ones that occasionally turn out wrong regardless of how confident the setup looked beforehand.
Frequently asked questions
What percentage should I risk per trade?
Most professional guidance clusters around 1-2% per trade for individual traders, with 0.5% being common for larger accounts or more conservative approaches. There’s no single correct number, but the consecutive-loss math above applies at whatever percentage is chosen, so it’s worth running the numbers at your own risk percentage before committing to it.
Does position sizing matter if I have a high win rate?
Yes, arguably more, not less. A high win rate can create false confidence that leads to oversized positions, and the eventual losing streak, which happens even to strategies with a strong track record, does exactly the same compounding damage regardless of how good the win rate looked beforehand.
How do I decide where to place the stop loss before calculating position size?
That’s a separate, chart-based decision, generally placed behind a real support or resistance level rather than an arbitrary dollar or percentage distance from entry. The position sizing formula takes that stop distance as a given input; it doesn’t tell you where the stop should go, only how many shares to buy once you know.
What if the position size formula suggests buying an oddly specific number of shares, like 33 or 17?
That’s normal and expected. Rounding down to the nearest whole share (or the nearest lot, if trading in blocks) is standard practice, and rounding down slightly reduces risk further rather than increasing it, so it’s the safer direction to round in.
Should position sizing change based on how confident I feel about a trade?
Varying size based on genuine differences in setup quality, tighter stops, clearer confirmation, is reasonable and common. Varying it based on a feeling of confidence untethered from the actual chart is how most oversized, account-damaging trades get justified in the moment.
Should I use the Kelly Criterion to size my trades?
Understanding it is worthwhile, using it at full strength usually isn’t. Full Kelly assumes your win rate and reward-to-risk ratio are known precisely, which they rarely are from a limited trading history, and the position sizes it recommends are typically far more aggressive than the 1-2% range that keeps a losing streak survivable. If you use it at all, use a small fraction of the calculated number, not the raw output.
Adjusting for volatility, not just stop distance
The formula above already adjusts for stop distance automatically, but it’s worth going one layer deeper: the stop distance itself should usually reflect how volatile the stock actually is, not a fixed dollar amount picked out of habit.
A stock that regularly swings 4-5% in a normal week needs a wider stop than one that typically moves 1-2%, or normal, meaningless noise will trigger the stop constantly regardless of whether the trade thesis was right. A common approach is basing the stop distance on a stock’s recent average trading range using a tool called average true range (ATR), a measure of typical price movement per period, rather than an arbitrary fixed percentage applied to every stock regardless of how it actually behaves.
The formula extends naturally: Shares to buy = (Account size × risk percentage) ÷ (ATR × multiplier), using the ATR-based distance in place of a manually chosen stop distance. The multiplier, commonly somewhere between 1.5 and 3, sets how many “typical moves” of breathing room the stop allows before it’s considered a real reversal rather than noise.
A side-by-side makes the effect concrete. Same $10,000 account, same 1% risk ($100), two stocks with very different volatility:
| Stock | ATR | Stop distance (2x ATR) | Shares bought | Position value |
|---|---|---|---|---|
| Calm, low-volatility stock | $1.50 | $3.00 | 33 shares | roughly $1,500 (at $45/share) |
| Volatile, high-beta stock | $4.00 | $8.00 | 12 shares | roughly $540 (at $45/share) |
Same dollar risk in both cases, $100. The volatile stock automatically gets a much smaller position, since each share carries more inherent risk. Skipping this step and buying the same share count regardless of volatility means the actual dollar risk on the volatile name is quietly two to three times larger than intended, even though the trade “looks” the same size on the screen.
The Kelly Criterion, and why most traders shouldn’t use it at full strength
No discussion of position sizing is complete without mentioning the Kelly Criterion, a formula developed in the 1950s to calculate the mathematically optimal fraction of capital to risk, given a known win rate and payoff ratio. It shows up constantly in more advanced trading material, and it’s worth understanding even though most individual traders shouldn’t actually use it at the number it produces.
The formula: Kelly % = W − [(1 − W) ÷ R], where W is the win rate as a decimal and R is the reward-to-risk ratio.
A worked example: a strategy with a 45% win rate and a 2-to-1 reward-to-risk ratio.
Kelly % = 0.45 − (0.55 ÷ 2) = 0.45 − 0.275 = 0.175, or 17.5% of capital per trade.
That number is almost certainly too aggressive to actually use. Full Kelly sizing assumes the win rate and payoff ratio are known with precision, which they never really are in live trading, they’re estimates from a limited sample of past trades. Even small estimation errors compound into wild swings in account value at full Kelly sizing. This is exactly why the consecutive-loss and risk-of-ruin math earlier in this article matters: 17.5% per trade sits close to the 10% row in those tables, the same territory that produced a 65% capital loss over 10 straight losses and a real, measurable chance of a 50% drawdown.
In practice, traders who use Kelly at all almost always use a fraction of it, commonly a quarter or half of the calculated number:
| Kelly fraction | Resulting risk per trade (from the 17.5% example above) |
|---|---|
| Full Kelly | 17.5% |
| Half Kelly | 8.8% |
| Quarter Kelly | 4.4% |
Even quarter Kelly, in this example, still sits above the 1-2% range most of this article has been built around. The honest takeaway: Kelly is a useful mental model for understanding that position size should scale with edge and payoff ratio, not a formula most individual traders should plug numbers into and follow literally. A fixed 1-2% risk per trade, chosen independently of any specific edge estimate, is a more forgiving starting point for anyone without a large, statistically reliable sample of their own trading results to calculate a trustworthy win rate and payoff ratio from in the first place.
Putting it together
None of this requires special software, a basic spreadsheet or even a calculator and the formula above is enough. What it requires is doing the calculation before every trade, not just the ones that feel risky, and treating a losing streak as confirmation the system is working as intended rather than a reason to abandon it.
Before your next trade, work out the actual dollar risk, the per-share stop distance, and the resulting share count, before looking at how much the position will cost to open. If those two numbers, position size and position risk, don’t match what you expected, that’s exactly the mismatch this formula exists to catch.
About the Author — Jamaluddin K.A.
Jamaluddin is the founder of The First Time Investor, a US and UK-focused personal finance and investing education site. He covers trading psychology, market mechanics, and the behavioral research behind why traders struggle with execution even when their strategy is sound.
Disclosure: This article is for educational purposes only and does not constitute financial advice. Trading involves significant risk including the potential loss of principal. Consult a licensed financial advisor before making investment decisions.