Sneaky Pivot: 4 Lines + 3 Candles [HR]
Sneaky Pivot is a rule-based version of the "sneaky pivot" day-trading method that a futures day trader named Doug presented for E-mini Dow futures (YM) on the 15-minute chart. It draws four daily levels and marks a three-candle reversal at those levels right after the New York open. A switch turns on a second rule set: a retelling by CryptoBosh that uses the same three candles at any time of day and was shown on Bitcoin. This is an independent formalization, not affiliated with either author.
ORIGINALITY AND USEFULNESS
This is not a mashup of existing indicators. Doug presented a discretionary routine ("strong, bold opening 15-minute candle", "go back to the left"); the script turns each vague step into an explicit rule with an input. The defaults are calibrated on his own YM trade of 12 May 2026 (index data plus that day's futures offset): entry 49,516 vs 49,521 he stated, stop 49,392 vs about 49,390, target 49,867 vs about 49,878. It does the error-prone manual work, gives each setup an exact entry, stop, target and reward-to-risk with an alert, and shows both rule sets on one chart.
HOW THE PARTS WORK TOGETHER
1. An NYSE calendar (holidays including Good Friday, Juneteenth from 2022 and special closures; early closes at 13:15 New York) decides which days trade and when each session ends.
2. That session defines yesterday as Doug draws it, 09:30-16:15 New York without overnight. This gives PDH and PDL.
3. A swing search over the whole loaded history (compact monotonic stack, no fixed lookback) finds LMAX / LMIN: the nearest swing high above PDH / swing low below PDL before yesterday (pivot, 3 bars on each side).
4. The three-candle rule runs only at these lines (shorts at PDH/LMAX, longs at PDL/LMIN), in Doug's version only at the open. A confirmed setup becomes a label and an alert.
THE THREE CANDLES (Doug's version, long; short is mirrored)
1. Candle 1: the first 15-minute candle after 09:30 New York; bearish; body at least 1.5 times the average of the last 20 bodies; low reaches PDL (0.05% tolerance; anchored at LMIN if it gets there). Circle "1".
2. Sneaky candle: the first bullish candle among the next two; it may make a lower low. On its close: triangle "2" and a label.
3. Entry: buy stop 1 tick above its high, valid for 2 candles. Stop: 0.01% below the pattern low. Target: PDH. Exit: 16:00 New York (13:15 on early closes).
Any-time variant (input "CryptoBosh version"): yesterday is the exchange daily candle; candle 1 is any 15-minute candle and closes back above the line; candle 2 is strictly the next one, makes no new low and closes above the line; the order lives 1 candle; stop 0.05% beyond the line; target 0.02% before PDH; no time exit or holiday filter.
INPUTS
Every rule above is an input; the values above are the defaults. Each version has its own settings block, so the switch changes everything at once. Also: Swing line (Swing pivot / First bar beyond the line), Target (Nearest opposite line / Far opposite line / Fixed R), Min reward/risk (0 = off).
HOW TO USE
1. Open CBOT_MINI:YM1!, 15 minutes, extended hours on (the tests used all bars). Any chart time zone works.
2. The top-right table shows the day's lines and where price is, and warns if the chart is not on 15 minutes. In Doug's version the day starts with the 09:30 candle, so before the open the table still shows the previous session's lines.
3. At 09:45 New York candle 1 closes; no "1" means no setup today.
4. If a "2" appears within two candles, its label gives entry, stop, target and R. The indicator does not track orders: if price hits the stop before the entry fills, skip the trade.
REPAINTING AND ALERTS
Confirmed signals do not repaint: lines use completed data, pivots are used only after confirmation, and a closed candle's markers and label never change. On the live, still-open candle a marker can appear and disappear, so act only on closed candles. Alerts with "Any alert() function call" fire once per bar close with side, line, entry, stop, target and R. For the fixed-text conditions "Sneaky Pivot: LONG" and "Sneaky Pivot: SHORT", choose "Once per bar close".
HONEST RESULTS (historical tests, not a promise)
Offline tests on 1-minute Dow Jones index data as a YM proxy, Sep 2013 - Sep 2026: 1-minute fills, $2.50 per side, 1-2 ticks of slippage; 15-minute moves correlate with real YM at 0.999. An independent re-implementation from the written rules matched all 549 trades of Doug's version (tick rounding off) and all 678 of the any-time variant.
- Doug's defaults: 549 trades (about 42 a year), +0.044 R per trade after costs; 95% interval -0.08 to +0.17 R, so not distinguishable from zero. Winners 39%, median -0.93 R. The five best trades made +31.5 R; without them, -7.4 R. Max drawdown -31 R, about -$11,400 per contract. With 4 ticks of entry slippage: -0.009 R.
- Random lines (200 runs): +0.047 R, the same as the real lines. 27,648 reasonable readings of the rules: half positive, median +0.0001 R.
- Any-time variant: 678 trades, -0.024 R. Bitcoin 2020-2026, after fees: Doug -0.11 R, any-time -0.28 R.
- Real futures, TradingView Deep Backtesting on CBOT_MINI:YM1!, strategy defaults (1 contract, $2.50 per side, 1 tick slippage). Doug since 2002: 880 trades, +$16,640 per contract, profit factor 1.10, 37% winners, max drawdown -$14,763; since 2013: 551 trades, +$19,130 (proxy, same costs: 549 trades, +$24,486). Any-time variant since 2002: 1,484 trades, -$49,645, profit factor 0.83; since 2013: 816 trades, -$34,740.
Simulations, not live trading. The indicator does not make the method profitable; it is a tool to watch and verify it.
LIMITATIONS
- Built and tested for YM on 15 minutes (plus Bitcoin, above); other timeframes may have no 09:30 candle.
- LMAX/LMIN come from loaded history only ("not in history" otherwise); loading older history can add lines to early days.
- On YM1! Doug's rules land close to the proxy, but the any-time variant is clearly worse: it trades around the clock, and the proxy lacks overnight quotes in 2015-2017.
Credit: the method was presented by a day trader named Doug; the any-time-of-day variant follows a retelling by CryptoBosh. The formalization, the calendar, the swing search and the tests are independent work.
For education and research only. Not financial advice. Historical results do not predict future results.
ORIGINALITY AND USEFULNESS
This is not a mashup of existing indicators. Doug presented a discretionary routine ("strong, bold opening 15-minute candle", "go back to the left"); the script turns each vague step into an explicit rule with an input. The defaults are calibrated on his own YM trade of 12 May 2026 (index data plus that day's futures offset): entry 49,516 vs 49,521 he stated, stop 49,392 vs about 49,390, target 49,867 vs about 49,878. It does the error-prone manual work, gives each setup an exact entry, stop, target and reward-to-risk with an alert, and shows both rule sets on one chart.
HOW THE PARTS WORK TOGETHER
1. An NYSE calendar (holidays including Good Friday, Juneteenth from 2022 and special closures; early closes at 13:15 New York) decides which days trade and when each session ends.
2. That session defines yesterday as Doug draws it, 09:30-16:15 New York without overnight. This gives PDH and PDL.
3. A swing search over the whole loaded history (compact monotonic stack, no fixed lookback) finds LMAX / LMIN: the nearest swing high above PDH / swing low below PDL before yesterday (pivot, 3 bars on each side).
4. The three-candle rule runs only at these lines (shorts at PDH/LMAX, longs at PDL/LMIN), in Doug's version only at the open. A confirmed setup becomes a label and an alert.
THE THREE CANDLES (Doug's version, long; short is mirrored)
1. Candle 1: the first 15-minute candle after 09:30 New York; bearish; body at least 1.5 times the average of the last 20 bodies; low reaches PDL (0.05% tolerance; anchored at LMIN if it gets there). Circle "1".
2. Sneaky candle: the first bullish candle among the next two; it may make a lower low. On its close: triangle "2" and a label.
3. Entry: buy stop 1 tick above its high, valid for 2 candles. Stop: 0.01% below the pattern low. Target: PDH. Exit: 16:00 New York (13:15 on early closes).
Any-time variant (input "CryptoBosh version"): yesterday is the exchange daily candle; candle 1 is any 15-minute candle and closes back above the line; candle 2 is strictly the next one, makes no new low and closes above the line; the order lives 1 candle; stop 0.05% beyond the line; target 0.02% before PDH; no time exit or holiday filter.
INPUTS
Every rule above is an input; the values above are the defaults. Each version has its own settings block, so the switch changes everything at once. Also: Swing line (Swing pivot / First bar beyond the line), Target (Nearest opposite line / Far opposite line / Fixed R), Min reward/risk (0 = off).
HOW TO USE
1. Open CBOT_MINI:YM1!, 15 minutes, extended hours on (the tests used all bars). Any chart time zone works.
2. The top-right table shows the day's lines and where price is, and warns if the chart is not on 15 minutes. In Doug's version the day starts with the 09:30 candle, so before the open the table still shows the previous session's lines.
3. At 09:45 New York candle 1 closes; no "1" means no setup today.
4. If a "2" appears within two candles, its label gives entry, stop, target and R. The indicator does not track orders: if price hits the stop before the entry fills, skip the trade.
REPAINTING AND ALERTS
Confirmed signals do not repaint: lines use completed data, pivots are used only after confirmation, and a closed candle's markers and label never change. On the live, still-open candle a marker can appear and disappear, so act only on closed candles. Alerts with "Any alert() function call" fire once per bar close with side, line, entry, stop, target and R. For the fixed-text conditions "Sneaky Pivot: LONG" and "Sneaky Pivot: SHORT", choose "Once per bar close".
HONEST RESULTS (historical tests, not a promise)
Offline tests on 1-minute Dow Jones index data as a YM proxy, Sep 2013 - Sep 2026: 1-minute fills, $2.50 per side, 1-2 ticks of slippage; 15-minute moves correlate with real YM at 0.999. An independent re-implementation from the written rules matched all 549 trades of Doug's version (tick rounding off) and all 678 of the any-time variant.
- Doug's defaults: 549 trades (about 42 a year), +0.044 R per trade after costs; 95% interval -0.08 to +0.17 R, so not distinguishable from zero. Winners 39%, median -0.93 R. The five best trades made +31.5 R; without them, -7.4 R. Max drawdown -31 R, about -$11,400 per contract. With 4 ticks of entry slippage: -0.009 R.
- Random lines (200 runs): +0.047 R, the same as the real lines. 27,648 reasonable readings of the rules: half positive, median +0.0001 R.
- Any-time variant: 678 trades, -0.024 R. Bitcoin 2020-2026, after fees: Doug -0.11 R, any-time -0.28 R.
- Real futures, TradingView Deep Backtesting on CBOT_MINI:YM1!, strategy defaults (1 contract, $2.50 per side, 1 tick slippage). Doug since 2002: 880 trades, +$16,640 per contract, profit factor 1.10, 37% winners, max drawdown -$14,763; since 2013: 551 trades, +$19,130 (proxy, same costs: 549 trades, +$24,486). Any-time variant since 2002: 1,484 trades, -$49,645, profit factor 0.83; since 2013: 816 trades, -$34,740.
Simulations, not live trading. The indicator does not make the method profitable; it is a tool to watch and verify it.
LIMITATIONS
- Built and tested for YM on 15 minutes (plus Bitcoin, above); other timeframes may have no 09:30 candle.
- LMAX/LMIN come from loaded history only ("not in history" otherwise); loading older history can add lines to early days.
- On YM1! Doug's rules land close to the proxy, but the any-time variant is clearly worse: it trades around the clock, and the proxy lacks overnight quotes in 2015-2017.
Credit: the method was presented by a day trader named Doug; the any-time-of-day variant follows a retelling by CryptoBosh. The formalization, the calendar, the swing search and the tests are independent work.
For education and research only. Not financial advice. Historical results do not predict future results.
Penafian
Maklumat dan penerbitan adalah tidak bertujuan, dan tidak membentuk, nasihat atau cadangan kewangan, pelaburan, dagangan atau jenis lain yang diberikan atau disahkan oleh TradingView. Baca lebih dalam Terma Penggunaan.
Penafian
Maklumat dan penerbitan adalah tidak bertujuan, dan tidak membentuk, nasihat atau cadangan kewangan, pelaburan, dagangan atau jenis lain yang diberikan atau disahkan oleh TradingView. Baca lebih dalam Terma Penggunaan.