Holy Grail strategy
On ES futures from January 2016 to July 2026, the Holy Grail lost $50,661 after costs and made money in 3 of 11 years.
TrendIntraday5-minute barsBacktested Nov 6, 2026
Holy Grail backtest, the result
Contract
E-mini S&P 500
-
−$50.7K
Net after costs
After $70,373 in fees and slippage
-
3 of 11
Years that made money
January 2016 to July 2026
-
−$93K
Largest drawdown
Peak to trough, 1 contract
-
−$21
Average trade
2,466 trades
Net after costs by month, January 2016 to July 2026.
Holy Grail rules
In a strong trend, wait for a pullback to the 20-bar EMA and enter as price turns back through the pullback bar's high.
A drawing of the rules on a made-up day, not a result.
- ADX(14) and the directional indicators are Wilder's, and the average is the 20-bar EMA, on every five-minute close of the Globex session.
- When the ADX is above 30, +DI is above -DI and a bar's low touches the EMA, a buy stop is set above the high of the bar before it, with a protective stop at the touching bar's low; shorts mirror it.
- The entry stop lives for the next five-minute bar; entries only in the day session.
- Whatever is open at the exit minute is closed: 15:45 New York on ES, 14:30 on CL.
- Decides on
- 5-minute bars
- Holds
- Nothing overnight
- Parameters
- threshold 30, warmup 28, open minute 570, exit minute 945
Results by year
It made money in 3 of 11 years on ES. Its best year was 2025, +$34,940, and its worst 2020, −$28,245.
| Year | Net | Trades | Win rate | Drawdown | Buy and hold |
|---|---|---|---|---|---|
| 2016 | −$9,229 | 228 | 19% | −$12,215 | +$11,497 |
| 2017 | −$7,102 | 245 | 20% | −$7,356 | +$21,936 |
| 2018 | +$9,521 | 244 | 25% | −$6,535 | −$8,589 |
| 2019 | −$15,204 | 266 | 19% | −$15,714 | +$35,436 |
| 2020 | −$28,245 | 257 | 20% | −$35,483 | +$27,761 |
| 2021 | −$9,250 | 243 | 23% | −$14,220 | +$51,361 |
| 2022 | −$11,880 | 239 | 22% | −$15,819 | −$45,939 |
| 2023 | −$10,110 | 217 | 24% | −$17,807 | +$38,435 |
| 2024 | +$6,188 | 218 | 24% | −$8,948 | +$42,947 |
| 2025 | +$34,940 | 200 | 31% | −$13,038 | +$37,010 |
| 2026, Jan to Jul | −$10,289 | 109 | 24% | −$19,920 | +$24,355 |
Holy Grail backtest results
On ES it made $19,713 before costs. Fees and slippage took $70,373, more than all of it.
- Net = before costs − fees − slippage
- ES: $19,713 − $8,723 − $61,650 = −$50,661
The interval holds zero: this test cannot tell the average trade from no edge.
Holy Grail code and Pine Script
The strategy as the backtest ran it, and the Pine Script our compiler wrote from it for TradingView.
1from atlantive import BUY, EMA, NEW_YORK, SELL, Bars, Session, Strategy2 3 4class HolyGrail(Strategy):5 """Raschke and Connors' Holy Grail on five-minute bars, traded in the contract's day session.6 7 ADX(14) and the directional indicators are Wilder's (alpha 1 / 14), the average is the 20-bar8 EMA, all on every five-minute close of the Globex session. When the ADX is above 30, +DI is above9 -DI and a bar's low touches the EMA, a buy stop is set one bar ahead above the high of the bar10 before it, with a protective stop at the touching bar's low; shorts mirror it (-DI above +DI, the11 high touching the EMA, a sell stop below the previous bar's low, the stop at the touching bar's12 high). The entry stop lives for the next five-minute bar. Both stops are tested on every13 one-second close, entries only in the day session, and whatever is open at its exit minute is14 closed: 15:45 New York on the equity index, 14:30 on crude oil. Nothing is read before 28 bars.15 The clock is one-second bars; the five-minute bars are built from them, by their start time.16 """17 18 data = "bars"19 20 class Params:21 threshold: float = 30.022 warmup: int = 2823 open_minute: int = 57024 exit_minute: int = 94525 clip: int = 126 27 class State:28 clock: Session(zone=NEW_YORK)29 bars: Bars(interval_ns=1_000_000_000)30 average: EMA(alpha=0.09523809523809523)31 true_range: EMA(alpha=0.07142857142857142)32 plus_dm: EMA(alpha=0.07142857142857142)33 minus_dm: EMA(alpha=0.07142857142857142)34 adx: EMA(alpha=0.07142857142857142)35 seen: int36 last_high: float37 last_low: float38 last_close: float39 armed: int40 trigger: float41 protect: float42 stop_level: float43 bar_key: int44 bar_high: float45 bar_low: float46 bar_close: float47 48 def on_event(self, view, ev, orders):49 self.clock.update(ev)50 if not self.bars.update(ev):51 return52 minute = self.clock.prev_minute53 second = self.clock.prev_second54 day = self.clock.prev_day55 high = float(self.bars.prev_high)56 low = float(self.bars.prev_low)57 close = float(self.bars.prev_close)58 key = (day * 1440 + minute) // 559 if key != self.bar_key:60 if self.bar_key != 0:61 self.armed = 062 h = self.bar_high63 lo = self.bar_low64 c = self.bar_close65 self.average.update(c)66 if self.seen > 0:67 up = h - self.last_high68 down = self.last_low - lo69 pdm = 0.070 mdm = 0.071 if up > down and up > 0:72 pdm = up73 if down > up and down > 0:74 mdm = down75 tr = max(h - lo, abs(h - self.last_close), abs(lo - self.last_close))76 self.true_range.update(tr)77 self.plus_dm.update(pdm)78 self.minus_dm.update(mdm)79 atr = self.true_range.value80 if atr > 0:81 pdi = 100.0 * self.plus_dm.value / atr82 mdi = 100.0 * self.minus_dm.value / atr83 total = pdi + mdi84 if total > 0:85 self.adx.update(100.0 * abs(pdi - mdi) / total)86 if self.seen >= self.p.warmup and self.adx.value > self.p.threshold:87 ema = self.average.value88 if pdi > mdi and lo <= ema:89 self.armed = 190 self.trigger = self.last_high91 self.protect = lo92 elif mdi > pdi and h >= ema:93 self.armed = -194 self.trigger = self.last_low95 self.protect = h96 self.seen += 197 self.last_high = h98 self.last_low = lo99 self.last_close = c100 self.bar_key = key101 self.bar_high = high102 self.bar_low = low103 else:104 self.bar_high = max(self.bar_high, high)105 self.bar_low = min(self.bar_low, low)106 self.bar_close = close107 late = second + 1 >= self.p.exit_minute * 60 or minute < self.p.open_minute108 if view.position > 0:109 if close <= self.stop_level or late:110 orders.ioc(side=SELL, qty=abs(view.position))111 elif view.position < 0:112 if close >= self.stop_level or late:113 orders.ioc(side=BUY, qty=abs(view.position))114 elif not late:115 if self.armed == 1 and close > self.trigger:116 orders.ioc(side=BUY, qty=self.p.clip)117 self.stop_level = self.protect118 self.armed = 0119 elif self.armed == -1 and close < self.trigger:120 orders.ioc(side=SELL, qty=self.p.clip)121 self.stop_level = self.protect122 self.armed = 0 1//@version=62// HolyGrail, 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("HolyGrail", 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() != 111 runtime.error("HolyGrail was exported for the 1 second timeframe; set the chart to 1S.")12 13threshold = input.float(30.0, "threshold")14warmup = input.int(28, "warmup", minval=-9007199254740991, maxval=9007199254740991)15open_minute = input.int(570, "open_minute", minval=-9007199254740991, maxval=9007199254740991)16exit_minute = input.int(945, "exit_minute", 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_q_imul(a, b) =>26 if a != 0 and math.abs(b) > 9007199254740991 / math.abs(a)27 runtime.error("an integer multiplication left the exact chart range; this run cannot be compared faithfully")28 _q_safe_int(a * b)29_q_ifloordiv(a, b) =>30 _q_r = a % b31 _q_q = (a - _q_r) / b32 _q_safe_int(_q_r != 0 and ((_q_r < 0) != (b < 0)) ? _q_q - 1 : _q_q)33 34_q_high = _q_int(math.round(high * 1000000000.0))35_q_low = _q_int(math.round(low * 1000000000.0))36_q_close = _q_int(math.round(close * 1000000000.0))37 38var int seen = 039var float last_high = 0.040var float last_low = 0.041var float last_close = 0.042var int armed = 043var float trigger = 0.044var float protect = 0.045var float stop_level = 0.046var int bar_key = 047var float bar_high = 0.048var float bar_low = 0.049var float bar_close = 0.050// clock: completed chart bar's opening timestamp (America/New_York).51_q_clock_prev_minute = hour(time, "America/New_York") * 60 + minute(time, "America/New_York")52_q_clock_prev_second = _q_clock_prev_minute * 60 + second(time, "America/New_York")53_q_clock_prev_weekday = (dayofweek(time, "America/New_York") + 5) % 754_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)55// Current clock: this chart bar's scheduled close, known without a future bar.56// A native next event after a data gap can carry a later time.57_q_clock_minute = hour(time_close, "America/New_York") * 60 + minute(time_close, "America/New_York")58_q_clock_second = _q_clock_minute * 60 + second(time_close, "America/New_York")59_q_clock_weekday = (dayofweek(time_close, "America/New_York") + 5) % 760_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)61_q_clock_seen = 162var float _q_average_value = 0.063var bool _q_average_seeded = false64var int _q_average_skipped = 065var int _q_average_frozen = 066var float _q_true_range_value = 0.067var bool _q_true_range_seeded = false68var int _q_true_range_skipped = 069var int _q_true_range_frozen = 070var float _q_plus_dm_value = 0.071var bool _q_plus_dm_seeded = false72var int _q_plus_dm_skipped = 073var int _q_plus_dm_frozen = 074var float _q_minus_dm_value = 0.075var bool _q_minus_dm_seeded = false76var int _q_minus_dm_skipped = 077var int _q_minus_dm_frozen = 078var float _q_adx_value = 0.079var bool _q_adx_seeded = false80var int _q_adx_skipped = 081var int _q_adx_frozen = 082int minute_l = na83int second_l = na84int day = na85float high_l = na86float low_l = na87float close_l = na88int key = na89float h = na90float lo = na91float c = na92float up = na93float down = na94float pdm = na95float mdm = na96float tr = na97float atr = na98float pdi = na99float mdi = na100float total = na101float ema = na102bool late = false103 104minute_l := _q_int(_q_clock_prev_minute)105second_l := _q_int(_q_clock_prev_second)106day := _q_int(_q_clock_prev_day)107high_l := _q_high * 1.0108low_l := _q_low * 1.0109close_l := _q_close * 1.0110key := _q_int(_q_ifloordiv(_q_iadd(_q_imul(day, 1440), minute_l), 5))111if key != bar_key112 if bar_key != 0113 armed := _q_int(0)114 h := bar_high115 lo := bar_low116 c := bar_close117 _q_sample1 = c118 if na(_q_sample1) or math.abs(_q_sample1) > 1.7976931348623157e308119 _q_average_skipped := _q_average_skipped + 1120 else121 if not _q_average_seeded122 _q_average_value := _q_sample1123 _q_average_seeded := true124 else125 _q_moved2 = _q_average_value + 0.09523809523809523 * (_q_sample1 - _q_average_value)126 if _q_moved2 == _q_average_value and _q_sample1 != _q_average_value127 _q_average_frozen := _q_average_frozen + 1128 _q_average_value := _q_moved2129 if seen > 0130 up := h - last_high131 down := last_low - lo132 pdm := 0.0133 mdm := 0.0134 if up > down and up > 0135 pdm := up136 if down > up and down > 0137 mdm := down138 tr := math.max(h - lo, math.max(math.abs(h - last_close), math.abs(lo - last_close)))139 _q_sample3 = tr140 if na(_q_sample3) or math.abs(_q_sample3) > 1.7976931348623157e308141 _q_true_range_skipped := _q_true_range_skipped + 1142 else143 if not _q_true_range_seeded144 _q_true_range_value := _q_sample3145 _q_true_range_seeded := true146 else147 _q_moved4 = _q_true_range_value + 0.07142857142857142 * (_q_sample3 - _q_true_range_value)148 if _q_moved4 == _q_true_range_value and _q_sample3 != _q_true_range_value149 _q_true_range_frozen := _q_true_range_frozen + 1150 _q_true_range_value := _q_moved4151 _q_sample5 = pdm152 if na(_q_sample5) or math.abs(_q_sample5) > 1.7976931348623157e308153 _q_plus_dm_skipped := _q_plus_dm_skipped + 1154 else155 if not _q_plus_dm_seeded156 _q_plus_dm_value := _q_sample5157 _q_plus_dm_seeded := true158 else159 _q_moved6 = _q_plus_dm_value + 0.07142857142857142 * (_q_sample5 - _q_plus_dm_value)160 if _q_moved6 == _q_plus_dm_value and _q_sample5 != _q_plus_dm_value161 _q_plus_dm_frozen := _q_plus_dm_frozen + 1162 _q_plus_dm_value := _q_moved6163 _q_sample7 = mdm164 if na(_q_sample7) or math.abs(_q_sample7) > 1.7976931348623157e308165 _q_minus_dm_skipped := _q_minus_dm_skipped + 1166 else167 if not _q_minus_dm_seeded168 _q_minus_dm_value := _q_sample7169 _q_minus_dm_seeded := true170 else171 _q_moved8 = _q_minus_dm_value + 0.07142857142857142 * (_q_sample7 - _q_minus_dm_value)172 if _q_moved8 == _q_minus_dm_value and _q_sample7 != _q_minus_dm_value173 _q_minus_dm_frozen := _q_minus_dm_frozen + 1174 _q_minus_dm_value := _q_moved8175 atr := _q_true_range_value176 if atr > 0177 pdi := (100.0 * _q_plus_dm_value * 1.0) / atr178 mdi := (100.0 * _q_minus_dm_value * 1.0) / atr179 total := pdi + mdi180 if total > 0181 _q_sample9 = (100.0 * math.abs(pdi - mdi) * 1.0) / total182 if na(_q_sample9) or math.abs(_q_sample9) > 1.7976931348623157e308183 _q_adx_skipped := _q_adx_skipped + 1184 else185 if not _q_adx_seeded186 _q_adx_value := _q_sample9187 _q_adx_seeded := true188 else189 _q_moved10 = _q_adx_value + 0.07142857142857142 * (_q_sample9 - _q_adx_value)190 if _q_moved10 == _q_adx_value and _q_sample9 != _q_adx_value191 _q_adx_frozen := _q_adx_frozen + 1192 _q_adx_value := _q_moved10193 if seen >= warmup and _q_adx_value > threshold194 ema := _q_average_value195 if pdi > mdi and lo <= ema196 armed := _q_int(1)197 trigger := last_high198 protect := lo199 else if mdi > pdi and h >= ema200 armed := _q_int(-1)201 trigger := last_low202 protect := h203 seen := _q_int(_q_iadd(seen, 1))204 last_high := h205 last_low := lo206 last_close := c207 bar_key := _q_int(key)208 bar_high := high_l209 bar_low := low_l210else211 bar_high := math.max(bar_high, high_l)212 bar_low := math.min(bar_low, low_l)213bar_close := close_l214late := _q_iadd(second_l, 1) >= _q_imul(exit_minute, 60) or minute_l < open_minute215if _q_int(strategy.position_size) > 0216 if close_l <= stop_level or late217 _q_qty11 = _q_int(math.abs(_q_int(strategy.position_size)))218 if _q_qty11 <= 0219 runtime.error("market order quantity must be a positive whole unit")220 _q_pos11 = _q_int(strategy.position_size)221 if _q_pos11 < 0 or _q_pos11 > 0 and _q_qty11 < _q_pos11222 runtime.error("this order would add to or partially reduce a position, so its trade ledger would not be faithful")223 strategy.order("q1", strategy.short, _q_qty11)224else if _q_int(strategy.position_size) < 0225 if close_l >= stop_level or late226 _q_qty12 = _q_int(math.abs(_q_int(strategy.position_size)))227 if _q_qty12 <= 0228 runtime.error("market order quantity must be a positive whole unit")229 _q_pos12 = _q_int(strategy.position_size)230 if _q_pos12 > 0 or _q_pos12 < 0 and _q_qty12 < -_q_pos12231 runtime.error("this order would add to or partially reduce a position, so its trade ledger would not be faithful")232 strategy.order("q2", strategy.long, _q_qty12)233else if not (late)234 if armed == 1 and close_l > trigger235 _q_qty13 = _q_int(clip)236 if _q_qty13 <= 0237 runtime.error("market order quantity must be a positive whole unit")238 _q_pos13 = _q_int(strategy.position_size)239 if _q_pos13 > 0 or _q_pos13 < 0 and _q_qty13 < -_q_pos13240 runtime.error("this order would add to or partially reduce a position, so its trade ledger would not be faithful")241 strategy.order("q3", strategy.long, _q_qty13)242 stop_level := protect243 armed := _q_int(0)244 else if armed == -1 and close_l < trigger245 _q_qty14 = _q_int(clip)246 if _q_qty14 <= 0247 runtime.error("market order quantity must be a positive whole unit")248 _q_pos14 = _q_int(strategy.position_size)249 if _q_pos14 < 0 or _q_pos14 > 0 and _q_qty14 < _q_pos14250 runtime.error("this order would add to or partially reduce a position, so its trade ledger would not be faithful")251 strategy.order("q4", strategy.short, _q_qty14)252 stop_level := protect253 armed := _q_int(0) 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 ESU6CL: 128 contracts, CLG6 to CLU6
- Costs paid
- ES: $8,723 in fees, $61,650 in slippageCL: $8,471 in fees, $42,100 in slippage
- Warm-up
- None: each contract reads only its own bars
- Flagged days
- ES: NoneCL: None Days the data holds too few bars for. Flagged, not removed.
Could you pass
a prop firm challenge
with the Holy Grail?
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
$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
Questions about the Holy Grail
What is the Holy Grail?
In a strong trend, wait for a pullback to the 20-bar EMA and enter as price turns back through the pullback bar's high. The rules follow Linda Bradford Raschke and Laurence A. Connors, Street Smarts: High Probability Short-Term Trading Strategies (1996). Atlantive backtested it on CME futures from January 2016 to July 2026 after costs.
Is the Holy Grail profitable on ES futures?
Not in this backtest: on ES futures from January 2016 to July 2026 it lost $50,661 after costs, over 2,466 trades, positive in 3 of 11 years; on CL it lost $48,761. Buying and holding one ES contract over the same window made $236,212. The average trade was −$21, its 95% interval −$52 to $15. These are hypothetical results of a backtest, not of trading.
What are the Holy Grail's rules?
ADX(14) and the directional indicators are Wilder's, and the average is the 20-bar EMA, on every five-minute close of the Globex session. When the ADX is above 30, +DI is above -DI and a bar's low touches the EMA, a buy stop is set above the high of the bar before it, with a protective stop at the touching bar's low; shorts mirror it. The entry stop lives for the next five-minute bar; entries only in the day session. Whatever is open at the exit minute is closed: 15:45 New York on ES, 14:30 on CL. The rules follow Linda Bradford Raschke and Laurence A. Connors, Street Smarts: High Probability Short-Term Trading Strategies (1996). Its parameters are the published ones, never tuned on the data it was tested on.
How was the Holy Grail 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 Holy Grail 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 Holy Grail 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.