Built-in Namespaces

Every built-in capability lives under a namespace. This is the reference for all of them. For language syntax (variables, control flow) see the Language Reference.

Bar series

Top-level price/volume series — a value per bar, indexable with [N].

Identifier	Meaning
close	Closing price
open	Opening price
high	High price
low	Low price
volume	Volume

gap_up = open > close[1]          # today's open above yesterday's close
range_today = high - low

ta — technical analysis

Function	Returns	Notes
ta.sma(series, length)	number	Simple moving average
ta.ema(series, length)	number	Exponential moving average
ta.wma(series, length)	number	Weighted moving average
ta.rsi(series, length)	number	Relative strength index
ta.atr(length)	number	Average true range
ta.adr(length)	number	Average daily range %
ta.macd(series, fast, slow, signal)	tuple (macd, signal, hist)	MACD
ta.bbands(series, length, mult)	tuple (upper, middle, lower)	Bollinger Bands
ta.highest(series, length)	number	Highest value over last N bars
ta.lowest(series, length)	number	Lowest value over last N bars
ta.crossover(a, b)	boolean	true only on the bar a crosses above b
ta.crossunder(a, b)	boolean	true only on the bar a crosses below b
ta.change(series, length=1)	number	series[0] - series[length]

fast = ta.sma(close, 50)
slow = ta.sma(close, 200)
golden_cross = ta.crossover(ta.sma(close, 50), ta.sma(close, 200))

Coming in v1.1

ta.stoch, ta.vwap, ta.twap, and ta.intra are not yet available. For a crossover on derived values (not bare series), use the manual previous-difference pattern shown in Examples.

strategy — trade actions

Function	Effect
strategy.enter_long(...)	Open a long position
strategy.enter_short(...)	Open a short position
strategy.close_long()	Close any open long
strategy.close_short()	Close any open short
strategy.close_all()	Close any open position

A position is opened or closed only by these calls. v1 allows one position per ticker — a second enter_long while already long does nothing. Closes always close the entire position.

Position sizing

Entries take exactly one of these sizing keywords:

Keyword	Meaning
size_pct=N	Allocate N% of current equity. This compounds — as profits grow, each position grows.
cash=N	Deploy a fixed dollar amount (shares = floor(N / price)). Does not compound.
qty=N	Buy a fixed number of shares. Does not compound.

strategy.enter_long(size_pct=100)   # all-in, compounding
strategy.enter_long(cash=10000)     # always $10,000
strategy.enter_long(qty=100)        # always 100 shares

Don't compound? Use cash= or qty=

size_pct is a percentage of your growing equity, so it snowballs. If you want every trade to be the same size, use cash= or qty= — never try to fake it with size_pct.

Fill timing

By default an order fills at the next bar's open (look-ahead safe). Override with one keyword:

Keyword	Fills at
(none)	Next bar's open (default)
at_open=true	This bar's open
at_close=true	This bar's close
at_next_close=true	Next bar's close

strategy.enter_long(cash=10000, at_close=true)   # fill at today's close

position — current position state

Field	Value when flat
position.size	0 (positive = long, negative = short)
position.entry_price	na
position.unrealized_pnl	na
position.unrealized_pnl_pct	na
position.bars_held	na

if position.size > 0 and position.unrealized_pnl_pct > 10:
    strategy.close_all()

input — tunable parameters

Inputs become editable fields in the UI, so users can tune your strategy without editing code.

Function	Returns
input.int(default, title)	integer
input.float(default, title)	float
input.bool(default, title)	boolean
input.string(default, title, options=[...])	string

length = input.int(14, "RSI Length")
threshold = input.float(30.0, "Oversold Level")

time — date & time

Calendar and clock fields for the current bar.

Field	Range
time.hour	0–23 (Eastern Time)
time.minute	0–59
time.dayofweek	0=Sunday, 1=Monday, 2=Tuesday … 5=Friday, 6=Saturday
time.dayofmonth	1–31
time.dayofyear	1–366
time.month	1–12
time.year	full year
time.unix	Unix timestamp (milliseconds)

# Trade only on Tuesdays
if position.size == 0 and time.dayofweek == 2:
    strategy.enter_long(cash=10000, at_close=true)

Day-of-week numbering

time.dayofweek is 0=Sunday, 1=Monday, 2=Tuesday, 3=Wednesday, 4=Thursday, 5=Friday, 6=Saturday. A common mistake is to treat 2 as Monday — it's Tuesday. On a daily timeframe, the value reflects the bar's trading date.

Pre-market & after-hours

On intraday timeframes, bars outside the regular session — pre-market (≈4:00–9:30 ET, time.hour < 9) and after-hours (≈16:00–20:00 ET, time.hour >= 16) — appear only when extended hours is enabled. Enable it from the header so the strategy can't silently no-op:

strategy("After-hours strategy", timeframe="5Min", extended_hours=true)

(Or toggle Extended Hours in the backtest settings.) With it off, only the regular session exists, so a pre/post-market strategy won't fire. Gate on time.hour to target a session. Note: extended-hours data must also exist for the ticker.

math — math helpers

Function	Returns
math.max(a, b, ...)	max
math.min(a, b, ...)	min
math.abs(x)	absolute value
math.sqrt(x)	square root
math.log(x)	natural log
math.exp(x)	e^x
math.floor(x)	floor
math.ceil(x)	ceiling
math.round(x)	nearest integer

Any argument that is na makes the result na.

external — other tickers

Read another symbol's price series from inside your strategy (read-only). Useful for market/regime filters.

Function	Returns
external.close("TICKER")	that ticker's close series
external.open / high / low / volume("TICKER")	other OHLCV series

The ticker must be a string literal.

# Only go long when SPY is above its 200-day average
spy_uptrend = external.close("SPY") > ta.sma(external.close("SPY"), 200)

if spy_uptrend and position.size == 0 and rsi < 30:
    strategy.enter_long(size_pct=10)

htf — higher-timeframe data

Read a higher timeframe for the current ticker or a named symbol — e.g. the weekly trend while trading daily bars.

Function	Returns
htf.close("1W")	this ticker's close on the given timeframe
htf.open / high / low / volume("1W")	other OHLCV fields
htf.close("SPY", "1W")	a named symbol's series at a given timeframe

The timeframe (and optional symbol) must be string literals, and the timeframe must be equal to or higher than the backtest's base timeframe.

# Daily strategy gated by the weekly trend
weekly_sma = ta.sma(htf.close("1W"), 30)

if close > weekly_sma and position.size == 0:
    strategy.enter_long(size_pct=10)

No look-ahead

A higher-timeframe bar only becomes visible once it has fully closed. While the current week is still forming, htf.close("1W") returns the last completed week's close — never a value from the future.

syminfo — current symbol

Field	Value
syminfo.ticker	the symbol the script is currently running on (a string)

A backtest runs the same script once per ticker. syminfo.ticker lets one strategy apply different rules to different symbols:

if syminfo.ticker == "QQQ":
    if position.size == 0 and time.dayofweek == 2:    # QQQ: buy Tuesday
        strategy.enter_long(cash=10000, at_close=true)
    if position.size > 0 and time.dayofweek == 3:     # sell Wednesday
        strategy.close_all(at_close=true)
elif syminfo.ticker == "AAPL":
    if position.size == 0 and time.dayofweek == 3:    # AAPL: buy Wednesday
        strategy.enter_long(cash=10000, at_close=true)
    if position.size > 0 and time.dayofweek == 5:     # sell Friday
        strategy.close_all(at_close=true)

Run this over a ["QQQ", "AAPL"] universe and each symbol follows its own branch.

Every branch must place real orders

syminfo.ticker only chooses which rules apply — it does not move money on its own. Each branch that should trade must call strategy.enter_long / close_all / etc. Tracking a "position" with your own true/false variable places no trades and produces no results.