Idea in brief
An episodic pivot (EP) is a day when news sharply changes a company's prospects. The stock opens with a large gap up, and in the first minutes it trades volume it normally never sees. The concept was systematized by Pradeep Bonde and popularized by Qullamaggie. Catalysts vary: an earnings surprise with raised guidance, inclusion in the S&P 500 or NASDAQ-100, an agreement with a large company, clinical trial data, a major contract.
The main version of the card comes from three videos by Jack Corsellis. A gap above 10%, the stock breaks out of a multi-month base and trades above all daily averages. The first candle of the day is wide, opened near its low and closed near its high on huge volume. Buy through the high of the first 1-, 5-, 15- or 30-minute candle, stop one cent below the low of the day. The stop in percent should be around the stock's ADR% or less.
A person evaluates the catalyst. Platforms cannot classify news, so event selection is moved to a separate rule with proxies for testing. Variants: Ariel enters above the high of the previous 5-minute candle and allows a re-entry; Bonde buys at the open and treats volume of 9 million shares or more as the catalyst itself. Entering days after a failed gap is described in the delayed-gap-reaction card.
Why it might work
The authors explain it through institutional money. A big surprise breaks analysts' models, and funds need to reprice the stock and build a position. That buying stretches over weeks and quarters. Ariel says a true "earnings winner" can rise for 6-8 quarters. Volume many times above normal shows that retail is not the buyer. Bonde adds that the effect is stronger on a neglected stock that has gone nowhere for a long time: for the market the surprise is bigger.
A strong first candle shows demand right now, and the low of the day gives a stop known before entry. Bonde goes further: if a stock with a 10 million share float trades 100 million shares, Bonde does not argue with the market, even without knowing the reason.
These explanations have not been verified. All results from the videos are breakdowns of individual trades and the traders' own journeys; there are no backtests.
Rules
Event selection (a separate rule, proxies Finetiq)
// author: a list before the open. Jack: the strongest pre-market gaps.
// Bonde: after-hours gaps with a notable rise and volume, news, a list by 08:30 ET
Gap = Open / Close[1] - 1
GapOK = Gap >= 10% // author (Jack, EP rules slide)
// catalyst (author): an earnings surprise on profit and revenue, raised guidance,
// index inclusion, partnership, FDA or phase 2-3, a major contract.
// Jack also counts gaps on a hot theme and sympathy gaps as EPs, but warns
// about small caps: the risk to the low of the day there is 15-25%
// Finetiq: platforms do not classify news. Three proxies for testing:
// a) earnings date from an external list AND GapOK
// b) GapOK AND RelVol (below), no news
// c) manual event labelingEntry and stop: Jack Corsellis (author, thresholds Finetiq)
// intraday data for the gap day, regular session 09:30-16:00 ET
N = 1, 5, 15 or 30 minutes // author: Qullamaggie 1, 5 or 60 minutes, Jack 5, 15 and 30
Bar1 = the first N-minute candle after 09:30 ET
ADRpct = 100 * (Average(High / Low, 20) - 1) // on daily bars before the gap
RelVol = Volume(Bar1) >= 3 * Average(Volume, 30) // author: volume many times above the 30-bar average
// Finetiq: multiplier 3
Strong = Close(Bar1) - Low(Bar1) >= 0.8 * (High(Bar1) - Low(Bar1))
// author: open near the low, close near the high; for TSLA in the top 5-10%. Finetiq: 0.8
IF NOT Strong on 5 minutes: evaluate the first 15- and 30-minute candle // author
BaseHigh = Highest(High, 120)[1] // Finetiq: base up to 6 months
EntryLvl = High(Bar1) + 0.01
AboveBase = EntryLvl >= 0.95 * BaseHigh // author: entry near or above the base highs; Finetiq: 5%
AboveMAs = EntryLvl > EMA(Close,10)[1] AND EntryLvl > EMA(Close,21)[1] AND EntryLvl > SMA(Close,50)[1]
AND EntryLvl > SMA(Close,65)[1] AND EntryLvl > SMA(Close,200)[1]
// author: price above the 10, 21, 50, 65 and 200-day averages; Finetiq: type of the 65-day one
StopLvl = LowOfDay - 0.01 // author: one cent below the low of the day
RiskPct = (EntryLvl - StopLvl) / EntryLvl * 100
IF GapOK AND RelVol AND Strong AND AboveBase AND AboveMAs
AND RiskPct <= ADRpct // author: around ADR% or less, ideally 1/2-2/3
AND RiskPct < 10 // author: stop in single-digit percent
BUY STOP at EntryLvl
// author: if sellers knocked the price down from the first candle's high, enter less aggressively
IF High(Bar1) - Close(Bar1) >= 0.5 * (High(Bar1) - Low(Bar1)) THEN size × 0.5 // Finetiq
// Finetiq: after a stop-out, no new entries that dayTrade management (author, choice for the rest and threshold Finetiq)
// author: half the position at 1R, the stop on the rest stays in place (free-roll-exit)
SELL 1/2 LIMIT at EntryPrice + (EntryPrice - StopLvl)
// the rest (author): a close below EMA 10, EMA 21 or SMA 50 depending on the stock's character,
// or a stop below each prior day's low if the stock accelerates quickly
SELL STOP at Low[1] // Finetiq: starting choice for the rest
// author: when strongly extended from EMA 10, tighten the stop on part of the position to the low of the day
IF Close > EMA10 * (1 + 2 * ADRpct / 100) THEN Stop = MAX(Stop, Low) // Finetiq: 2 ADR% thresholdVariant B. Ariel: 5-minute candle breakout and re-entry
// event (author): a gap up on earnings with a surprise on profit and revenue,
// volume many times above normal: with a usual 4 million a day, 12, 15, 20 or 30 million
RelVolDay = volume since the open >= 3 × average volume for the same time of day over 50 days // Finetiq
// entry (author): after a 5-minute candle closes, buy above its high, stop at the low of the day
BUY STOP at High(last closed 5-minute candle) + 0.01
StopLvl = LowOfDay - 0.01
// author: if there was a morning flush, enter above the candle that ended it
// Finetiq: the level moves to each new closed candle while there is no position.
// Second reading for testing: only above the first candle of the day
// author: after a stop-out at the low of the day, price may later break the 5-minute range again
IF stopped out today AND no position THEN BUY STOP at High(first 5-minute candle) + 0.01 // Finetiq: one re-entry
SectorOK = the sector fund is rising: QQQ for technology, GDX or gold futures for gold miners
// author: checking the sector; Finetiq: "rising" = above its own open
Size = MIN(20% of account, 1% of account / (RiskPct / 100))
// author: a slow stock (3% a day) up to 20% of the account, a fast one (8% a day, 8-10% risk to the low of the day) about 10%, risk about 1% of the portfolio
// exit (author): after 3-5 days sell part into strength, hold the rest until a close below
// a rising 10-, 20- or 50-day SMAVariant C. Pradeep Bonde: volume as the catalyst, buying at the open
// EP 9 million (author): scan after the close
Scan = Volume > 9 000 000 AND Close >= 3 // author: does not trade below $3
Record = Volume > Highest(Volume, 250)[1] // author: "never traded this much"; Finetiq: 250 days
Early = Close[1] / Lowest(Low, 60)[1] - 1 <= 25% // author: early in the move, not at the top; Finetiq: threshold
// author: stronger if volume is comparable to the float or larger (float data needed)
// author: if such a breakout comes at the start of a swing, Bonde "just gets in". Finetiq: two readings
// a) intraday, once cumulative volume exceeds 9 million and price is above the prior day's high
// b) at the next day's open
// classic EP on earnings (author): a market-on-open order before the open or entry in the first 5-10 minutes
BUY AT OPEN
InitialStop = Open * (1 - ADRpct / 100) // Finetiq: the author did not name an initial stop
IF position in profit THEN Stop = MAX(Stop, EntryPrice) // author: breakeven
// author: as the move approaches the estimated target, the stop goes below each day's low
// author: day 1 is now often better closed as a day trade, otherwise the pullback eats 30-40% of the day's moveParameters
| Parameter | Value | Source |
|---|---|---|
| Gap | 10% or more | author |
| Catalyst | earnings, index, partner, FDA, contract | author, proxies Finetiq |
| First candle timeframe | 1, 5, 15, 30 minutes | author |
| First candle volume | 3 × the 30-bar average or more | author, multiplier Finetiq |
| First candle strength | close in the top 20% | Finetiq (TSLA: top 5-10%) |
| Base | entry at 95% or more of the 120-bar high | author, numbers Finetiq |
| Averages | above EMA 10, EMA 21, SMA 50, 65, 200 | author, type of the 65-day one Finetiq |
| Stop | one cent below the low of the day | author |
| Maximum stop | no more than ADR% and less than 10% | author |
| Quality zone | 1/2-2/3 ADR% | author |
| Partial exit | 1/2 at 1R | author |
| Exit for the rest | below the prior day's low | author, choice Finetiq |
| B: entry | above the previous 5-minute candle | author |
| B: re-entry | one, above the first candle | author, limit Finetiq |
| B: size | 10-20% of the account, risk about 1% | author |
| B: exit | part after 3-5 days, the rest on SMA 10/20/50 | author |
| C: volume | more than 9 million shares, a record | author, window Finetiq |
| C: price | $3 or more | author |
| C: entry | market-on-open or the first 5-10 minutes | author |
| C: initial stop | 1 ADR% from the open | Finetiq |
| C: management | breakeven, then below the low of the day | author |
What to test
- Catalyst proxy. Three runs: earnings dates with a gap of 10% or more, a gap with volume and no news, EP 9 million. If the no-news proxies work no worse, manual news evaluation is not needed. If they are clearly worse, the edge sits in reading the news, and the rules do not capture it.
- Entry timeframe. 1, 5, 15 and 30 minutes versus Bonde's market-on-open. Look at the ratio of RiskPct to ADR% and the share of events skipped because the stop was wider than ADR%.
- Gap and volume thresholds. Gap of 5%, 10%, 15%, 20%; volume 2, 3 and 5 ×. Bonde says a catalyst gap used to be 8-12%, and now it is 20% plus another 20% during the day. Compare results by year.
- Averages and base filter. With and without it. Jack requires price above all averages, Bonde looks for neglected stocks near their lows. These are different sets of events and are worth testing separately.
- Holding. Close at the end of day 1 (Bonde), half at 1R and the rest below the prior day's low, Ariel's variant with a partial sale after 3-5 days and SMA 20. Compare the average trade and time in position.
- Slippage. A stop order in the first minutes of a gap fills worse than calculated, and small caps have huge spreads. Assume 0.2-0.5% on entry and check what remains of the average trade.
- Survivorship bias. Gappers are often small companies that later leave the exchange. You need a history that includes delisted stocks.
Platform notes
TradingView (Pine Script)
- A strategy trades only the chart symbol. The morning gap scan across all stocks cannot be repeated in a backtest; you can only test known events one stock at a time.
- A stock's earnings history is available via
request.earnings. Other catalysts (index, FDA, contract) are not in the data. - The depth of 1- and 5-minute history is limited by the subscription plan. If extended hours are enabled on the chart, the open and the low of the day are taken from the pre-market, while the rules are written for the regular session.
- If the entry and the low of the day are both touched within one candle, their order is unknown. You need
use_bar_magnifier = trueor a lower timeframe.
MultiCharts and TradeStation (EasyLanguage)
- Events are set by an external list of dates; the scan across many stocks is done in Portfolio Maestro or Portfolio Trader. Check whether your data includes delisted stocks.
Timeis the bar's close time: the first 5-minute candle after 09:30 hasTime = 935, the 1-minute oneTime = 931.- On TradeStation intraday bars, full volume has historically been stored in
Ticks, whileVolumemay contain only up volume. The first-candle volume filter depends on this field. Buy next bar at X stoplives for one bar, so on the gap day the order must be sent on every bar. Accumulate the low of the day yourself from the first bar of the session.
MetaTrader 5 (MQL5)
- Brokers' stock CFDs usually cover a few hundred large companies. The small companies where most EPs happen are missing, and a market scan is impossible.
- CFDs have no real volume, only tick volume. The relative volume filters and the 9 million share threshold cannot be reproduced.
- A CFD gap depends on the broker's quoting hours. If the CFD trades in extended hours, the "open" and the gap are different.
- There is no market-on-open order. A market entry on the first tick of the session fills at a widened spread. Convert 09:30 ET to server time, accounting for daylight saving time.
Where the idea can break
- The edge, according to the authors themselves, is in evaluating the news. Bonde says outright that the method is not mechanical and requires analytical skills built over several earnings seasons. A no-news proxy may have no edge at all.
- The authors' results are breakdowns of selected trades and trading journeys. Ariel's figure was stated by the channel host. In reply, Ariel says that 2023-2024 were especially favorable years for a trend follower, but does not say how much of the result came from them.
- Gaps have become larger, and stops to the low of the day wider. The "stop no wider than ADR%" filter cuts out a noticeable share of events, especially in small companies.
- The best events run in the first minutes; Bonde says hesitation at entry is not forgiven. A test on 5-minute bars with perfect stop order execution is more optimistic than reality.
- You need pre-market data, minute history, earnings dates, float and data with delistings. For a retail trader this is expensive, and without them the test is incomplete.
- Both Chart Fanatics videos are published on a channel that advertises a prop firm; Jack Corsellis's videos are sponsored by MarketSmith and MarketSurge.