Pine Script Strategy Position Sizing: Fixed, Cash, Percent of Equity, Qty and Margin

Position sizing is a test assumption, not a cosmetic setting

Start by separating a strategy default from an order-level override. The default_qty_type and default_qty_value arguments in strategy() apply to entry and order calls that omit qty. If a call supplies a non-na qty, that call does not use the default.

The distinction changes the path of the simulation. Fixed sizing generally sends the same unit count. Cash sizing converts an account-currency notional into units at the order price. Percent-of-equity sizing changes as both available equity and price change. A sizing change therefore deserves its own version label, export and comparison rather than being folded into one undifferentiated result.

The same value means three different things

The numerical value is not comparable until its unit is known.

Futures and CFDs can have a contract value that is not simply one chart price per contract. Check syminfo.pointvalue and the instrument specification before carrying a stock-style notional calculation across markets. The mini calculator below is deliberately simplified and is not a contract-specification engine.

Audit qty precedence before changing Properties

If changing Default order size in Properties leaves results unchanged, inspect every entry and order call for a hard-coded or calculated qty. A non-na value wins. To fall back to the default, omit the argument or deliberately return na when the default should apply.

TV17 covers exit reservations, partial exits and OCA behavior.

Keep only the capital, currency and margin links that change sizing

TV11 covers the full Strategy Properties fingerprint. Here, retain only the dependencies that enter quantity resolution: initial_capital seeds the first percent-of-equity allocation, currency defines the unit of a cash default, and margin_long/margin_short constrain whether the resulting notional can be maintained.

Cash sizing converts account currency into instrument quantity. Percent-of-equity sizing first creates a notional allocation from available equity. Margin does not select the entry qty, but rejected exposure or simulated liquidation can alter the quantity path, so the setting still belongs beside the run.

Hold entries constant and vary only the sizing method

This educational script uses a simple moving-average cross and deliberately omits qty. It is not evidence that the signal has an edge. The omission lets you change fixed, cash and percent-of-equity defaults in Properties without an order-level override silently defeating the test.

//@version=6
strategy(
    "Quantity interpretation audit",
    overlay = true,
    initial_capital = 100000,
    currency = currency.USD,
    default_qty_type = strategy.percent_of_equity,
    default_qty_value = 10,
    margin_long = 100,
    margin_short = 100
)

int fastLength = input.int(20, "Fast length", minval = 1)
int slowLength = input.int(50, "Slow length", minval = 2)
float fast = ta.sma(close, fastLength)
float slow = ta.sma(close, slowLength)

bool longCondition = ta.crossover(fast, slow)
bool flatCondition = ta.crossunder(fast, slow)

if longCondition and strategy.position_size == 0
    strategy.entry("Long", strategy.long)

if flatCondition and strategy.position_size > 0
    strategy.close("Long")

plot(fast, "Fast", color.aqua)
plot(slow, "Slow", color.orange)
plot(strategy.position_size, "Position size", display = display.data_window)
plot(strategy.equity, "Equity", display = display.data_window)

Keep code and signals fixed. Run fixed, cash and percent-of-equity variants one at a time. For each run, record type, value, capital, currency, margin, period, symbol, timeframe and costs, then save a separately named List of Trades export.

Reconcile an equity allocation with the first trades

The mini calculator divides an equity allocation by a simple order price and estimates collateral. It assumes one unit has a notional equal to price. It excludes futures point value, FX conversion, CFD contract size, quantity steps, open exposure and trading costs.

1. 01Capture Properties

   Record quantity type and value, capital, currency, margin, commission and slippage.

2. 02Reconcile the first three trades

   Compare entry price, transaction size, resulting position and equity available at entry.

3. 03Change one condition

   Keep signals fixed and change only the sizing type or the sizing value.

4. 04Export separately

   Name each Strategy Tester CSV/XLSX with the sizing assumption and version.

5. 05Compare the path

   Review return, drawdown, trade loss distribution, size concentration and margin events over the same period.

Bind one quantity ledger to one Strategy Tester export

Define the research unit as one run, one quantity-ledger row and one Strategy Tester export. Use a shared version ID so the source-side qty path and the realized transaction quantities can be reconciled later without relying on memory.

• Does every entry ID use the intended quantity path?
• Do filenames and notes identify fixed, cash and equity variants unambiguously?
• Did you recheck the unit meaning after changing symbol or timeframe?
• Are reversals, forced reductions and open trades separated from ordinary closed trades?
• Did you compare drawdown percentage and size concentration, not just net currency profit?

Minimum ledger fields are version ID, script hash, symbol, timeframe, entry ID, sizing type and value, override state, account and symbol currencies, point value, margin, and expected versus observed quantity for the first three trades. Mark any mismatch unresolved before publication.

Put the version ID and sizing method in the filename instead of overwriting fixed, cash and percent-of-equity runs. The ledger then separates entry-rule behavior from compounding and late-sample size concentration using the same evidence trail.