Risk Management

Position Sizing with Volatility: ATR Formula and Practical Rules

How to calculate position size using ATR, the Kelly criterion, VIX-based portfolio scaling, and inverse volatility allocation. Covers the core ATR formula with worked examples, ATR multiplier selection by strategy type, Van Tharp's percent volatility model, and specific rules for sizing options trades based on implied volatility.

February 27, 2026

Published February 27, 2026

Position sizing is the single decision that separates traders who survive drawdowns from those who blow up. The core idea: let the market’s current volatility dictate how many shares, contracts, or lots you trade. When volatility is high, you trade smaller. When it is low, you trade larger. This article covers the ATR-based position sizing formula, the Kelly criterion, VIX-based portfolio scaling, inverse volatility allocation, and specific rules for sizing options trades. Every method includes a worked example with real numbers.

1-2%
Standard Risk Per Trade
1.5-3x
ATR Stop Multiplier Range
Half-Kelly
Standard Practice for Real Trading
20-30%
Max Capital in Short Premium

Why Volatility Should Determine Your Position Size

Most retail traders pick a fixed dollar amount or a fixed number of shares per trade. That approach ignores the single variable that changes how far a stock can move against you: volatility.

A stock with a $1 average daily range and a stock with a $5 average daily range require different position sizes even if both are priced at $100. Trading 500 shares of each means $500 of daily heat on one and $2,500 on the other. Same share count, five times the risk.

Volatility-based position sizing normalizes risk across instruments and across market regimes. When the VIX spikes from 15 to 35, your position sizes shrink automatically. When volatility contracts, sizes grow. The result is a more consistent P&L distribution and fewer account-threatening drawdowns.

Key Takeaway
Volatility-based position sizing adjusts your exposure to match the market environment. You risk the same dollar amount per trade regardless of whether you are trading a low-volatility utility stock or a high-volatility biotech name.

The Core ATR Position Sizing Formula

Average True Range (ATR) is the foundation of most volatility-based sizing methods. ATR measures the average daily range of an instrument over a lookback period (typically 14 days), including gaps. It gives you a dollar-denominated measure of how much the instrument moves per day.

ATR Position Sizing Formula
Position Size (shares) = (Account Equity × Risk %) / (ATR × Multiplier × Price per Point)

Simplified for stocks (price per point = 1):

Position Size = (Account × Risk%) / (ATR × Multiplier)

The variables:

  • Account Equity: your total account value
  • Risk %: the percentage of your account you are willing to lose on this trade (typically 1-2%)
  • ATR: the 14-day Average True Range in dollar terms
  • Multiplier: how many ATRs away you place your stop loss (typically 1.5x to 3x)

The multiplier determines your stop distance. A 2x ATR stop means your stop is placed two average daily ranges from your entry. Wider stops require smaller position sizes. Tighter stops allow larger positions but get stopped out more frequently.

Step-by-Step Example: ATR Position Sizing

Suppose you have a $100,000 account and risk 1% per trade. You want to buy a stock trading at $50 with a 14-day ATR of $2.00. You use a 2x ATR multiplier for your stop.

Worked Example
Account: $100,000

Risk per trade: 1% = $1,000

Stock price: $50

14-day ATR: $2.00

ATR multiplier: 2x

Stop distance: $2.00 × 2 = $4.00

Position Size = $1,000 / $4.00 = 250 shares

Notional value: 250 × $50 = $12,500

Portfolio allocation: 12.5% of account

If this stock’s ATR doubled to $4.00 during a volatility spike, the same formula produces 125 shares instead of 250. Your dollar risk stays at $1,000 either way. That is the point: constant risk, variable position size.

ATR Multiplier by Strategy Type

Strategy ATR Multiplier Stop Distance Rationale
Day trading / scalping 0.5x – 1.0x Tight Short holding period, quick exits, high frequency
Intraday swing 1.0x – 1.5x Medium-tight Holds for hours, needs room for intraday noise
Swing trading (2-10 days) 1.5x – 2.5x Medium Multi-day holds, must survive overnight gaps
Position trading (weeks+) 2.5x – 3.0x Wide Longer timeframe, wider stop needed for trend following
Trend following (Turtle-style) 2.0x – 3.0x Wide Classic N-unit sizing, lets winners run

Day traders using the Bollinger Band squeeze strategy might use a 0.75x ATR stop. A trend follower running a Turtle-style breakout system uses 2x ATR (what the original Turtles called “2N”). The multiplier must match your strategy’s time horizon and win rate.

Van Tharp’s Percent Volatility Model

Van Tharp formalized volatility-based sizing in his position sizing research. His percent volatility model is a direct application of the ATR formula above, with one refinement: he frames the risk budget as the maximum acceptable daily portfolio fluctuation per position.

Van Tharp Percent Volatility Model
Position Size = (Account × Volatility%) / ATR

Where Volatility% = the max daily fluctuation you’ll accept per position

Example: $100,000 account, 1% daily volatility budget, ATR = $3.00

Position Size = ($100,000 × 0.01) / $3.00 = 333 shares

The difference from the ATR stop method is subtle but meaningful. Instead of sizing based on stop distance, you size based on daily expected movement. This means every position contributes roughly the same dollar volatility to your portfolio each day. A portfolio of ten positions, each sized at 1% volatility, produces an expected daily portfolio fluctuation of roughly 3.2% (1% × sqrt(10), assuming uncorrelated positions).

Tharp vs. ATR Stop Method
The ATR stop method asks: “How much do I lose if I’m wrong?” Tharp’s percent volatility model asks: “How much does this position move my portfolio each day?” Both use ATR. The stop method ties size to risk tolerance. The Tharp model ties size to portfolio volatility contribution. Professional fund managers typically use the Tharp approach because it normalizes day-to-day P&L swings across all holdings.

Kelly Criterion for Volatility-Adjusted Sizing

The Kelly criterion calculates the mathematically optimal fraction of your account to risk, given your win rate and payoff ratio. It was developed by John Kelly at Bell Labs in 1956 and adopted by traders including Ed Thorp.

Kelly Criterion Formula
f* = (p × b – q) / b

Where:

f* = fraction of account to risk

p = probability of winning

q = probability of losing (1 – p)

b = win/loss ratio (average win / average loss)

Example: 55% win rate, average win $600, average loss $400

b = 600 / 400 = 1.5

f* = (0.55 × 1.5 – 0.45) / 1.5

f* = (0.825 – 0.45) / 1.5

f* = 0.375 / 1.5 = 0.25 (25% of account)

Full Kelly is too aggressive for real trading. A 25% allocation per trade means a string of four losers wipes out your account. The estimation of win rate and payoff ratio also contains uncertainty. If your actual win rate is 50% instead of 55%, full Kelly produces catastrophic drawdowns.

Half-Kelly: The Standard Practice

Professional traders and fund managers use half-Kelly or less. Half-Kelly achieves roughly 75% of the growth rate of full Kelly while cutting the maximum drawdown nearly in half. The math favors fractional Kelly because the downside of over-betting is far worse than the downside of under-betting.

In the example above, half-Kelly would risk 12.5% of the account per trade. Many traders go further and use quarter-Kelly (6.25%), especially when trading strategies with uncertain edge estimates.

Risk Warning
The Kelly criterion assumes you know your exact win rate and payoff ratio. In practice, these are estimates from backtested data. Backtests overstate edge. Slippage, commissions, and regime changes erode performance. Always use fractional Kelly (half or less) and verify your edge with the Volatility Backtester before committing real capital.

VIX-Based Portfolio Position Sizing

The VIX provides a market-wide volatility reading that can scale your entire portfolio’s risk exposure. When VIX is low, volatility is compressed and positions can be larger. When VIX spikes, you reduce across the board.

The logic is straightforward: a VIX at 35 means the S&P 500’s expected daily move is 2.2%, roughly double the expected move at VIX 15 (0.95%). Your portfolio should reflect that difference.

VIX Level Market Regime Position Size Adjustment Max Positions Risk Per Trade
Below 15 Low volatility Full size (100%) 8-10 2% of account
15-20 Normal Full size (100%) 6-8 1.5% of account
20-25 Elevated Reduce by 15% 5-6 1% of account
25-35 High Reduce by 25% 3-5 0.75% of account
Above 35 Extreme Reduce by 50% 2-3 0.5% of account

These are guidelines, not hard rules. Some traders invert the VIX scaling for short volatility strategies, increasing premium-selling size when VIX is elevated because the edge on selling overpriced options is wider. But that approach requires deep understanding of tail risk and margin requirements.

VIX-Adjusted Position Size
Baseline position = ATR formula result at VIX 15-20

VIX adjustment factor:

If VIX > 25: multiply position by 0.75

If VIX > 35: multiply position by 0.50

Example: ATR formula gives 250 shares at VIX 18 (normal)

VIX spikes to 30: 250 × 0.75 = 187 shares

VIX spikes to 40: 250 × 0.50 = 125 shares

The VIX-based overlay works on top of your ATR sizing. The ATR already adjusts for the individual instrument’s volatility. The VIX overlay adjusts for the macro environment. Both layers matter. A low-volatility stock during a VIX spike can still gap 5% on correlated selling.

Inverse Volatility Position Sizing

Inverse volatility sizing allocates capital inversely proportional to each asset’s volatility. Assets with lower volatility get larger allocations. Assets with higher volatility get smaller allocations. This is the foundation of risk parity portfolio construction.

Inverse Volatility Weight Formula
Weight_i = (1 / Vol_i) / Sum(1 / Vol_j) for all assets j

Example: 3-asset portfolio

Asset A: 15% annualized volatility

Asset B: 30% annualized volatility

Asset C: 45% annualized volatility

Inverse vols: 1/0.15 = 6.67, 1/0.30 = 3.33, 1/0.45 = 2.22

Sum = 12.22

Weight A = 6.67 / 12.22 = 54.6%

Weight B = 3.33 / 12.22 = 27.3%

Weight C = 2.22 / 12.22 = 18.2%

Compare this to equal-weight (33.3% each). Under equal weighting, Asset C contributes three times the portfolio volatility of Asset A. Under inverse volatility weighting, each asset contributes roughly equal risk to the portfolio.

Inverse volatility is used by funds running risk parity strategies (Bridgewater’s All Weather being the most famous). For individual traders, it applies when allocating across multiple uncorrelated positions. If you trade both SPY and a biotech stock, inverse volatility sizing ensures neither position dominates your portfolio’s risk profile.

When to Recalculate Inverse Volatility Weights

Volatility changes. A quarterly rebalance captures regime shifts without over-trading. Monthly rebalancing is more responsive but incurs higher transaction costs. Weekly rebalancing is excessive for most portfolios. The Volatility Box scanner tracks real-time ATR and IV rank across 595 symbols, giving you updated volatility readings to recalculate weights when needed.

Position Sizing Methods Compared

Method Input Best For Weakness Complexity
Fixed dollar risk Account %, stop distance Beginners, simple systems Ignores volatility entirely Low
ATR-based (% risk) ATR, multiplier, account % Active traders, swing trading ATR is backward-looking Low-Medium
Van Tharp % volatility ATR, daily vol budget Portfolio managers, multi-asset Requires portfolio-level thinking Medium
Kelly criterion Win rate, payoff ratio Systematic traders with data Overestimates optimal size; use half-Kelly Medium
VIX-based scaling VIX level, base position Portfolio-wide risk management Blunt instrument; not symbol-specific Low
Inverse volatility Annualized vol per asset Multi-asset allocation, risk parity Assumes vol = risk; ignores correlation Medium

No single method covers every situation. Professional traders layer them. Use ATR-based sizing for individual trade entries. Apply VIX-based scaling as a portfolio overlay. Use inverse volatility for asset allocation across uncorrelated strategies.

How to Size Options Trades Based on Implied Volatility

Options sizing follows different rules than equity sizing because options have nonlinear payoffs, time decay, and leverage. A 100-share equivalent position in options can cost $500 or $5,000 depending on strike selection, expiration, and implied volatility.

Rule 1: Cap Risk Per Trade at 2-5% of Account

For defined-risk options strategies (vertical spreads, iron condors, butterflies), risk no more than 2-5% of your account per trade. The maximum loss is defined by the spread width minus the credit received.

Options Position Size (Defined Risk)
Max contracts = (Account × Risk%) / (Spread Width × 100 – Credit × 100)

Example: $50,000 account, 3% risk, $5-wide iron condor for $1.50 credit

Max risk per contract: ($5.00 – $1.50) × 100 = $350

Max contracts: ($50,000 × 0.03) / $350 = 4.28 → 4 contracts

Total risk: 4 × $350 = $1,400 (2.8% of account)

Rule 2: Scale Size Inversely with IV

When IV rank is elevated, individual options positions are more expensive but also carry more premium. The temptation is to sell more contracts when IV is high. Resist it. Elevated IV means larger expected moves and wider potential losses on short premium.

A practical approach: use your standard sizing formula at IV rank 30-50. Reduce contract count by 25% when IV rank exceeds 60. Reduce by 50% when IV rank exceeds 80. The wider spreads available at high IV partially offset the smaller contract count.

Rule 3: Portfolio-Level Short Premium Cap

Never deploy more than 20-30% of your account in short premium strategies at any one time. This includes short puts, short calls, strangles, and the short legs of spreads. The reason: correlated moves. When the VIX spikes, all your short premium positions move against you simultaneously. Keeping total exposure below 30% ensures you can survive a two-standard-deviation event without a margin call.

Risk Warning
Undefined-risk options trades (naked puts, naked calls, strangles) can lose multiples of the premium collected. Size these based on margin requirement, not premium received. A short strangle on a $200 stock might collect $5.00 in premium but require $20,000 in margin. Your risk is the margin figure, not the credit.

How Professional Traders Adjust Size for Volatility

Institutional desks and professional independent traders share several position sizing practices that differ from typical retail approaches.

1. They size to a volatility target, not a return target. A fund running a 10% annualized volatility target adjusts leverage daily based on realized and implied volatility. When vol drops, they increase exposure. When vol rises, they cut. The goal is a steady volatility output, not maximum return.

2. They use correlation-adjusted sizing. Two positions in the same sector are not independent bets. Professional traders reduce the combined size of correlated positions using portfolio-level VaR (Value at Risk). If you hold both AAPL and MSFT, the combined position should be smaller than two unrelated positions because they tend to move together.

3. They distinguish between entry size and maximum size. Initial entry might be one-third of the intended position. They scale in as the trade proves itself, adding at higher prices (for longs) or after the setup develops further. This “proof of concept” approach limits damage from immediate adverse moves.

4. They reduce size before known events. Earnings, FOMC decisions, CPI prints. Professional traders cut position size by 25-50% ahead of binary events where implied volatility understates the real risk of a gap move. The Conviction Score on Volatility Box incorporates upcoming catalysts into its signal quality rating.

Position Sizing for High vs Low Volatility Environments

The optimal position size shifts significantly between volatility regimes. Below is a framework for a $100,000 account trading equities.

Parameter Low Vol (VIX < 15) Normal Vol (VIX 15-20) High Vol (VIX 25-35) Extreme Vol (VIX > 35)
Risk per trade 1.5-2% 1-1.5% 0.5-1% 0.25-0.5%
ATR multiplier 1.5x 2.0x 2.5x 3.0x
Max open positions 8-10 6-8 3-5 1-3
Total portfolio risk 10-15% 8-10% 4-6% 2-3%
Typical position size 10-15% of account 8-12% of account 5-8% of account 3-5% of account

In low volatility, the ATR is small, so the ATR formula naturally produces larger positions. You can afford more simultaneous positions because each one contributes less daily volatility. In extreme volatility, the ATR is large, positions shrink, and you run fewer of them to prevent portfolio-level meltdowns.

Key Takeaway
The ATR formula automatically adjusts position size to volatility. But you should also manually adjust your risk percentage and number of positions based on the VIX regime. Both layers protect you: ATR handles instrument-level volatility, VIX scaling handles market-level volatility.

Putting It All Together: A Complete Sizing Workflow

Step 1: Check the macro regime. Look at the VIX. If VIX is above 25, reduce your base risk percentage from 1% to 0.75%. If above 35, reduce to 0.5%.

Step 2: Calculate ATR for the specific instrument. Use a 14-day ATR. For shorter-term trades, a 7-day ATR is more responsive. The Volatility Box scanner displays ATR data alongside its proprietary volatility levels.

Step 3: Choose your ATR multiplier. Match it to your strategy’s time horizon per the table above.

Step 4: Run the formula. Position Size = (Account × Risk%) / (ATR × Multiplier). This gives you your share count.

Step 5: Check portfolio exposure. Does this new position, combined with existing positions, exceed 20-30% of your account in any single sector? Does total portfolio risk exceed your target? If so, reduce the new position or close an existing one.

Step 6: Verify with the Volatility Backtester. Backtest the strategy with your proposed position size against historical data. Check maximum drawdown. If the worst historical drawdown at this size exceeds what you can tolerate, scale down.

Key Takeaways

  • Position size should be a function of volatility, not a fixed number of shares or dollars
  • Core formula: Position Size = (Account × Risk%) / (ATR × Multiplier)
  • ATR multiplier ranges from 0.5x (scalping) to 3x (trend following) depending on strategy
  • Kelly criterion gives the mathematically optimal bet size; use half-Kelly or less in practice
  • Reduce position size by 25% when VIX exceeds 25, by 50% when VIX exceeds 35
  • Inverse volatility weighting equalizes risk contribution across portfolio assets
  • Cap short premium exposure at 20-30% of account regardless of strategy
  • Options trades: risk 2-5% per defined-risk trade, scale down as IV rank rises

Backtest Your Position Sizing Before You Trade It

The Volatility Backtester lets you test any position sizing approach against historical volatility data across 595 symbols going back to 2008. See how ATR-based sizing, VIX-scaled positions, and different risk percentages perform through bull markets, bear markets, and volatility spikes. Every backtest includes drawdown analysis so you know the worst-case scenario before risking real capital.

Try the Volatility Backtester

Frequently Asked Questions

What is the volatility-based position sizing formula? +
The standard formula is: Position Size = (Account Equity × Risk Percentage) / (ATR × Multiplier). For a $100,000 account risking 1% with a $2.00 ATR and 2x multiplier, the position size is $1,000 / $4.00 = 250 shares. The ATR adjusts the denominator to current volatility, so position size automatically shrinks when volatility expands.
How do I use ATR for position sizing step by step? +
Step 1: Calculate or look up the 14-day ATR for your instrument in dollar terms. Step 2: Multiply ATR by your stop multiplier (1.5x to 3x depending on strategy). This is your stop distance. Step 3: Determine your dollar risk (account size × risk percentage, typically 1-2%). Step 4: Divide dollar risk by stop distance. The result is your position size in shares.
What is the Kelly criterion and should I use full Kelly? +
The Kelly criterion calculates the optimal bet size as f* = (p × b - q) / b, where p is win rate, q is loss rate, and b is the win/loss ratio. Full Kelly maximizes long-term growth but produces severe drawdowns. Professional traders use half-Kelly or quarter-Kelly because it captures most of the growth rate with substantially lower drawdown risk. Backtested win rates also tend to overstate real-world edge, making fractional Kelly more appropriate.
How should I adjust position size when VIX is high? +
A standard framework: maintain full position sizes when VIX is below 20. Reduce by 15% when VIX is 20-25. Reduce by 25% when VIX is 25-35. Reduce by 50% when VIX exceeds 35. Also reduce the number of simultaneous positions. In extreme volatility environments (VIX above 35), limit yourself to 1-3 positions with wider stops and smaller size.
What is inverse volatility position sizing? +
Inverse volatility sizing allocates more capital to lower-volatility assets and less to higher-volatility assets. The weight for each asset equals its inverse volatility divided by the sum of all inverse volatilities. A 15% volatility asset gets roughly three times the allocation of a 45% volatility asset. This approach equalizes each asset's risk contribution to the portfolio, similar to risk parity strategies used by institutional funds.
How do I size options trades based on implied volatility? +
For defined-risk strategies, divide your risk budget (2-5% of account) by the maximum loss per contract. For a $5-wide iron condor collecting $1.50 in credit, max loss is $350 per contract. A $50,000 account risking 3% can trade 4 contracts. When IV rank is above 60, reduce contract count by 25%. When above 80, reduce by 50%. Never deploy more than 20-30% of your account in short premium simultaneously.
What is the optimal position size in high volatility vs low volatility? +
In low volatility (VIX below 15), you can risk 1.5-2% per trade with 8-10 simultaneous positions. In high volatility (VIX 25-35), reduce to 0.5-1% risk per trade with 3-5 positions. The ATR-based formula handles most of this automatically: higher ATR produces smaller positions. Layer in VIX-based scaling and reduced position count for additional protection during volatile regimes.

Stop guessing. Start using data.

600+ symbols. Updated every 2 minutes. Backtested methodology since 2008.

Try the Scanner