9 and 21 EMA crossover strategy

On ES futures, the 9 and 21 EMA crossover lost $99,526 after costs and ran a $100,000 account down to its margin on March 3, 2020. The run stops there.

TrendIntraday5-minute barsBacktested Nov 6, 2026

9 and 21 EMA crossover backtest, the result

Contract

E-mini S&P 500

  • −$99.5K

    Net after costs

    After $105,276 in fees and slippage

  • 1 of 5

    Years that made money

    January 2016 to March 2020

  • −$106K

    Largest drawdown

    Peak to trough, 1 contract

  • −$27

    Average trade

    3,709 trades

9 and 21 EMA crossover on ES, net after costs by month, from January 2016 to March 2020: −$99,526; buying and holding one ES contract: +$236,212

Net after costs by month, January 2016 to March 2020.

9 and 21 EMA crossover rules

Buy when the 9-bar EMA crosses above the 21-bar EMA on 5-minute bars, and sell when it crosses back below.

A drawing of the rules on a made-up day, not a result.

  1. The 9 and 21 bar EMAs (alpha 2 / (n + 1)) run on every five-minute close of the Globex session.
  2. When the 9 crosses above the 21 the strategy goes long, and when it crosses below it goes short, turning round on the opposite cross.
  3. New positions are taken only on bars that close in the day session before 15:45 New York.
  4. Whatever is open is closed at 15:45 New York.
Decides on
5-minute bars
Holds
Nothing overnight
Parameters
warmup 21, open minute 570, exit minute 945, bar minutes 5

Results by year

It made money in 1 of 5 years on ES. Its best year was 2018, +$7,237, and its worst 2019, −$35,656. The account was closed out on March 3, 2020, so the years stop there.

Net after costs by year on ES, with trades, win rate, largest drawdown and buying and holding the same contract
YearNetTradesWin rateDrawdownBuy and hold
2016−$26,43487730%−$33,364+$11,497
2017−$34,02792625%−$34,596+$21,936
2018+$7,23789832%−$27,506−$8,589
2019−$35,65686230%−$39,852+$35,436
2020, Jan to Mar−$10,64614634%−$18,405+$27,761

9 and 21 EMA crossover backtest results

On ES it made $5,750 before costs. Fees and slippage took $105,276, more than all of it.

Net = before costs − fees − slippage
ES: $5,750 − $12,551 − $92,725 = −$99,526
Every figure of the 9 and 21 EMA crossover backtest, by contract
FigureESNQ
Net after costs−$99,526−$99,019
Before costs+$5,750−$12,335
Fees$12,551$12,519
Slippage$92,725$74,165
Trades3,7093,700
Trades a year824822
Win rate29%31%
Profit factor0.780.83
Average trade−$27−$27
Average trade, 95% interval−$39 to −$14−$43 to −$11
Sharpe ratio-2.00-1.51
Largest drawdown−$105,563−$103,036
Time in the market15%15%
Buying and holding the contract+$236,212+$401,004

The whole interval is under zero: after costs, the average trade lost money.

9 and 21 EMA crossover code and Pine Script

The strategy as the backtest ran it, and the Pine Script our compiler wrote from it for TradingView.

9 and 21 EMA crossover
ema_crossover.py
1from atlantive import BUY, EMA, NEW_YORK, SELL, Bars, Session, Strategy2 3 4class EmaCrossover(Strategy):5    """9 and 21 EMA crossover on five-minute bars, traded in the New York day session.6 7    Both averages run on every five-minute close of the Globex session, alpha 2 / (n + 1). When the8    9 crosses above the 21 the strategy goes long, and when it crosses below it goes short, turning9    the position round on the opposite cross. New positions are taken only on bars that close in the10    day session before 15:45, and whatever is open at 15:45 is closed. No cross is read before 2111    bars have closed on the contract.12    """13 14    data = "bars"15 16    class Params:17        warmup: int = 2118        open_minute: int = 57019        exit_minute: int = 94520        bar_minutes: int = 521        clip: int = 122 23    class State:24        clock: Session(zone=NEW_YORK)25        bars: Bars(interval_ns=300_000_000_000)26        fast: EMA(alpha=0.2)27        slow: EMA(alpha=0.09090909090909091)28        last_diff: float29        seen: int30 31    def on_event(self, view, ev, orders):32        self.clock.update(ev)33        if not self.bars.update(ev):34            return35        minute = self.clock.prev_minute36        day = self.clock.prev_day37        close = float(self.bars.prev_close)38        self.fast.update(close)39        self.slow.update(close)40        diff = self.fast.value - self.slow.value41        self.seen += 142        ends = minute + self.p.bar_minutes43        in_day = minute >= self.p.open_minute and ends < self.p.exit_minute44        crossed_up = self.seen > self.p.warmup and self.last_diff <= 0 and diff > 045        crossed_down = self.seen > self.p.warmup and self.last_diff >= 0 and diff < 046        self.last_diff = diff47        if view.position != 0 and not in_day:48            if view.position > 0:49                orders.ioc(side=SELL, qty=abs(view.position))50            else:51                orders.ioc(side=BUY, qty=abs(view.position))52            return53        if not in_day:54            return55        if crossed_up:56            if view.position < 0:57                orders.ioc(side=BUY, qty=abs(view.position) + self.p.clip)58            elif view.position == 0:59                orders.ioc(side=BUY, qty=self.p.clip)60        elif crossed_down:61            if view.position > 0:62                orders.ioc(side=SELL, qty=abs(view.position) + self.p.clip)63            elif view.position == 0:64                orders.ioc(side=SELL, qty=self.p.clip)
1//@version=62// EmaCrossover, exported from Atlantive.3// Chart simulation: decisions use completed bars; market orders fill at the next bar open.4// Native fills, fees, data gaps and session boundaries can differ. Re-test on the target symbol.5// Use standard time-based candles and the strategy's timeframe, not synthetic chart prices.6// Pine v6 trims old trade records at its order limit; the strategy keeps calculating.7// Margin is explicitly disabled to retain the native signal simulation's sizing.8strategy("EmaCrossover", overlay=true, initial_capital=100000.0, commission_type=strategy.commission.cash_per_contract, commission_value=1.68, default_qty_type=strategy.fixed, default_qty_value=1, calc_on_every_tick=false, process_orders_on_close=false, slippage=0, margin_long=0, margin_short=0)9 10if timeframe.in_seconds() != 30011    runtime.error("EmaCrossover was exported for the 5 minute timeframe; set the chart to 5.")12 13warmup = input.int(21, "warmup", minval=-9007199254740991, maxval=9007199254740991)14open_minute = input.int(570, "open_minute", minval=-9007199254740991, maxval=9007199254740991)15exit_minute = input.int(945, "exit_minute", minval=-9007199254740991, maxval=9007199254740991)16bar_minutes = input.int(5, "bar_minutes", minval=-9007199254740991, maxval=9007199254740991)17clip = input.int(1, "clip", minval=-9007199254740991, maxval=9007199254740991)18 19_q_safe_int(x) =>20    if math.abs(x) > 900719925474099121        runtime.error("an integer left the exact chart range; this run cannot be compared faithfully")22    int(x)23_q_int(x) => _q_safe_int(x < 0 ? math.ceil(x) : math.floor(x))24_q_iadd(a, b) => _q_safe_int(a + b)25 26_q_close = _q_int(math.round(close * 1000000000.0))27 28var float last_diff = 0.029var int seen = 030// clock: completed chart bar's opening timestamp (America/New_York).31_q_clock_prev_minute = hour(time, "America/New_York") * 60 + minute(time, "America/New_York")32_q_clock_prev_second = _q_clock_prev_minute * 60 + second(time, "America/New_York")33_q_clock_prev_weekday = (dayofweek(time, "America/New_York") + 5) % 734_q_clock_prev_day = int(timestamp("UTC", year(time, "America/New_York"), month(time, "America/New_York"), dayofmonth(time, "America/New_York"), 0, 0) / 86400000)35// Current clock: this chart bar's scheduled close, known without a future bar.36// A native next event after a data gap can carry a later time.37_q_clock_minute = hour(time_close, "America/New_York") * 60 + minute(time_close, "America/New_York")38_q_clock_second = _q_clock_minute * 60 + second(time_close, "America/New_York")39_q_clock_weekday = (dayofweek(time_close, "America/New_York") + 5) % 740_q_clock_day = int(timestamp("UTC", year(time_close, "America/New_York"), month(time_close, "America/New_York"), dayofmonth(time_close, "America/New_York"), 0, 0) / 86400000)41_q_clock_seen = 142var float _q_fast_value = 0.043var bool _q_fast_seeded = false44var int _q_fast_skipped = 045var int _q_fast_frozen = 046var float _q_slow_value = 0.047var bool _q_slow_seeded = false48var int _q_slow_skipped = 049var int _q_slow_frozen = 050int minute_l = na51int day = na52float close_l = na53float diff = na54int ends = na55bool in_day = false56bool crossed_up = false57bool crossed_down = false58 59bool _q_running = true60minute_l := _q_int(_q_clock_prev_minute)61day := _q_int(_q_clock_prev_day)62close_l := _q_close * 1.063_q_sample1 = close_l64if na(_q_sample1) or math.abs(_q_sample1) > 1.7976931348623157e30865    _q_fast_skipped := _q_fast_skipped + 166else67    if not _q_fast_seeded68        _q_fast_value := _q_sample169        _q_fast_seeded := true70    else71        _q_moved2 = _q_fast_value + 0.2 * (_q_sample1 - _q_fast_value)72        if _q_moved2 == _q_fast_value and _q_sample1 != _q_fast_value73            _q_fast_frozen := _q_fast_frozen + 174        _q_fast_value := _q_moved275_q_sample3 = close_l76if na(_q_sample3) or math.abs(_q_sample3) > 1.7976931348623157e30877    _q_slow_skipped := _q_slow_skipped + 178else79    if not _q_slow_seeded80        _q_slow_value := _q_sample381        _q_slow_seeded := true82    else83        _q_moved4 = _q_slow_value + 0.09090909090909091 * (_q_sample3 - _q_slow_value)84        if _q_moved4 == _q_slow_value and _q_sample3 != _q_slow_value85            _q_slow_frozen := _q_slow_frozen + 186        _q_slow_value := _q_moved487diff := _q_fast_value - _q_slow_value88seen := _q_int(_q_iadd(seen, 1))89ends := _q_int(_q_iadd(minute_l, bar_minutes))90in_day := minute_l >= open_minute and ends < exit_minute91crossed_up := seen > warmup and last_diff <= 0 and diff > 092crossed_down := seen > warmup and last_diff >= 0 and diff < 093last_diff := diff94if _q_int(strategy.position_size) != 0 and not in_day95    if _q_int(strategy.position_size) > 096        _q_qty5 = _q_int(math.abs(_q_int(strategy.position_size)))97        if _q_qty5 <= 098            runtime.error("market order quantity must be a positive whole unit")99        _q_pos5 = _q_int(strategy.position_size)100        if _q_pos5 < 0 or _q_pos5 > 0 and _q_qty5 < _q_pos5101            runtime.error("this order would add to or partially reduce a position, so its trade ledger would not be faithful")102        strategy.order("q1", strategy.short, _q_qty5)103    else104        _q_qty6 = _q_int(math.abs(_q_int(strategy.position_size)))105        if _q_qty6 <= 0106            runtime.error("market order quantity must be a positive whole unit")107        _q_pos6 = _q_int(strategy.position_size)108        if _q_pos6 > 0 or _q_pos6 < 0 and _q_qty6 < -_q_pos6109            runtime.error("this order would add to or partially reduce a position, so its trade ledger would not be faithful")110        strategy.order("q2", strategy.long, _q_qty6)111    _q_running := false112if _q_running113    if not (in_day)114        _q_running := false115if _q_running116    if crossed_up117        if _q_int(strategy.position_size) < 0118            _q_qty7 = _q_int(_q_iadd(math.abs(_q_int(strategy.position_size)), clip))119            if _q_qty7 <= 0120                runtime.error("market order quantity must be a positive whole unit")121            _q_pos7 = _q_int(strategy.position_size)122            if _q_pos7 > 0 or _q_pos7 < 0 and _q_qty7 < -_q_pos7123                runtime.error("this order would add to or partially reduce a position, so its trade ledger would not be faithful")124            strategy.order("q3", strategy.long, _q_qty7)125        else if _q_int(strategy.position_size) == 0126            _q_qty8 = _q_int(clip)127            if _q_qty8 <= 0128                runtime.error("market order quantity must be a positive whole unit")129            _q_pos8 = _q_int(strategy.position_size)130            if _q_pos8 > 0 or _q_pos8 < 0 and _q_qty8 < -_q_pos8131                runtime.error("this order would add to or partially reduce a position, so its trade ledger would not be faithful")132            strategy.order("q4", strategy.long, _q_qty8)133    else if crossed_down134        if _q_int(strategy.position_size) > 0135            _q_qty9 = _q_int(_q_iadd(math.abs(_q_int(strategy.position_size)), clip))136            if _q_qty9 <= 0137                runtime.error("market order quantity must be a positive whole unit")138            _q_pos9 = _q_int(strategy.position_size)139            if _q_pos9 < 0 or _q_pos9 > 0 and _q_qty9 < _q_pos9140                runtime.error("this order would add to or partially reduce a position, so its trade ledger would not be faithful")141            strategy.order("q5", strategy.short, _q_qty9)142        else if _q_int(strategy.position_size) == 0143            _q_qty10 = _q_int(clip)144            if _q_qty10 <= 0145                runtime.error("market order quantity must be a positive whole unit")146            _q_pos10 = _q_int(strategy.position_size)147            if _q_pos10 < 0 or _q_pos10 > 0 and _q_qty10 < _q_pos10148                runtime.error("this order would add to or partially reduce a position, so its trade ledger would not be faithful")149            strategy.order("q6", strategy.short, _q_qty10)

How it was tested

The published rules, never tuned on the data, with real costs on every fill. Every strategy here is tested the same way.

Market data
CME Globex 5-minute bars, Jan 2, 2016 to Jul 31, 2026
Rules
As published, with the published parameters, never tuned on this data
Size
1 contract on every trade
Contract
The most traded one, rolled as the market moved
Fees
The broker's $0.50 a side, plus CME's and the NFA's fees when the window opens
Slippage
1 tick a side on ES, 2 in NQ's regular session and 3 outside it, 1 on other markets, where it was not measured
Stops and entries
Tested on 1-second closes and filled at the next bar
Contracts traded
ES: 43 contracts, ESH6 to ESU6NQ: 43 contracts, NQH6 to NQU6
Costs paid
ES: $12,551 in fees, $92,725 in slippageNQ: $12,519 in fees, $74,165 in slippage
Warm-up
None: each contract reads only its own bars
Flagged days
ES: NoneNQ: None Days the data holds too few bars for. Flagged, not removed.

Could you pass
a prop firm challenge
with the 9 and 21 EMA crossover?

A backtest says whether a strategy makes money. Our AI runs its trades through the rules of every prop firm account, then tells you which one it would pass and at what size.

No card needed. Free credits every month.

  • Most money: Blue Guardian Futures Direct 50K, $5,084
  • Best payout odds: Funded Futures Family Straight to Funded 50K, 98%
  • Fastest pass: Phidias Express to Live 50K, 11 days
Blue Guardian Futures Direct 50K
Most money

$5,084

Median net across 2,000 simulated runs

Evaluation
None
First payout
97%
Fees paid
$1,521

Median net across 2,000 simulated runs

Best 5%$13,143Worst 5%−$412
Day 060120180
Example: ES opening range breakout

Questions about the 9 and 21 EMA crossover

What is the 9 and 21 EMA crossover?

Buy when the 9-bar EMA crosses above the 21-bar EMA on 5-minute bars, and sell when it crosses back below. Atlantive backtested it on CME futures from January 2016 to July 2026 after costs.

Is the 9 and 21 EMA crossover profitable on ES futures?

Not in this backtest: on ES futures from January 2016 to March 2020 it lost $99,526 after costs, over 3,709 trades, positive in 1 of 5 years; on NQ it lost $99,019. Buying and holding one ES contract over the same window made $236,212. The average trade was −$27, its 95% interval −$39 to −$14. These are hypothetical results of a backtest, not of trading.

What are the 9 and 21 EMA crossover's rules?

The 9 and 21 bar EMAs (alpha 2 / (n + 1)) run on every five-minute close of the Globex session. When the 9 crosses above the 21 the strategy goes long, and when it crosses below it goes short, turning round on the opposite cross. New positions are taken only on bars that close in the day session before 15:45 New York. Whatever is open is closed at 15:45 New York. Its parameters are the published ones, never tuned on the data it was tested on.

How was the 9 and 21 EMA crossover backtested?

On CME Globex data from January 2016 to July 2026, one contract, on 5-minute bars, rolled to the most traded contract as the market moved, with the broker's and CME's fees and the slippage measured on each market. Stops and entries were tested on 1-second closes and filled at the next bar. Nothing was fitted to the data.

Can I run the 9 and 21 EMA crossover in TradingView?

Yes. The page carries its Pine Script, written by Atlantive's compiler from the same code the backtest ran, for TradingView's strategy tester. A chart's fills and fees can differ from the backtest's, so test it again on the symbol you trade.

Can I change the 9 and 21 EMA crossover and test it again?

Yes. Rerun it in Atlantive: a new chat opens with the strategy's rules in the box, for you to read, change and send. The agent writes the code, backtests it on the market and window you name, and reports the result with its costs.

Rerun it in Atlantive. Change the market, the dates or a rule, then send it.

A new chat opens with this in the box

Backtest a 9/21 EMA crossover on five-minute bars on ES, 2016-01-02 to 2026-07-31, one contract on a $100,000 account with default costs: EMAs on the whole Globex session, go long on a cross up and short on a cross down, entries only on bars closing in the day session before 15:45, flat at 15:45.

Never sent for you

May we measure this visit?

With your consent, Google Analytics and PostHog count visits to these pages. You can separately allow Google Ads to measure visits from our ads. Visit measurement starts only after you choose. Cookie policy