Premarket Break-and-Retest Trading: A Learning Guide to Scarface Trades’ $250/Day Strategy

Video: If I Had To Make $250/Day Trading, I’d ONLY Use This Strategy
Creator: Scarface Trades
Original Video: Watch the original video

The Big Idea

The method organizes a trading decision around a range established before the regular US stock-market session. Mark its high and low, wait for price to break outside that range, then look for a return to the broken boundary and evidence of continuation.

The mental model is location → breakout → retest → confirmation → defined risk. Location tells you where to pay attention. The breakout suggests direction. The retest offers a place to ask whether the market is accepting the new price area. Confirmation supplies a decision point; a stop defines where that decision has failed.

Scarface Trades demonstrates the setup on one-minute stock charts and describes expressing the directional view through weekly options. These are two separate layers: a chart setup on the underlying stock, and an option contract whose price responds to more than the stock alone.

The title frames the method around $250 a day. The useful lesson is the repeatable decision process. A daily dollar goal does not establish an edge, and it should not turn the absence of a valid setup into a reason to trade.

If You Remember Only One Thing

A broken premarket level is a hypothesis. The retest is where you look for evidence and define what would invalidate it.

60-Second Summary

The presenter marks the high and low between 4:00 a.m. and 9:30 a.m. Eastern.
The examples use one-minute charts and liquid momentum stocks.
A bullish setup breaks the high, retests it, and shows bullish continuation; a bearish setup mirrors that at the low.
The presenter waits for price-action confirmation rather than entering every breakout.
The illustrated stop is beyond the entry candle’s relevant extreme.
The examples use a prior intraday high or another chart level as a target, with an emphasis on at least roughly 2R of potential reward.
The option preference is a weekly expiration and a nearby out-of-the-money strike.
The supplied examples explain execution logic, but do not establish typical profitability. Our code below tests an explicit stock-price proxy.

Why the Premarket Range Gives the Setup Structure

Imagine price has spent the morning moving between $49 and $50. Those boundaries summarize where buying and selling have previously stopped the move. If price later closes above $50, the question changes: will the market now accept prices above an area that previously contained it?

A retest returns to that boundary. If price approaches $50 and then closes back above it, the pattern is consistent with resistance becoming support. That is an interpretation of price behavior, not direct evidence that a particular group of buyers is defending the level.

For this guide, the premarket range includes bars opening at 04:00 through 09:29 Eastern and freezes at 09:30. Use a named timezone such as America/New_York rather than a fixed UTC offset, because daylight saving time changes the offset.

The transcript extends the time window to other markets. Treat that as a proposed convention to investigate: US equity premarket activity, continuous crypto trading, and other instruments’ session structures are different populations. The baseline here stays with US equity charts.

Break, Retest, and Confirmation Are Different Events

Event	What you observe	What remains uncertain
Breakout	Price moves beyond the premarket boundary.	It may return immediately to the range.
Retest	Price comes back toward that boundary.	The boundary may fail rather than hold.
Confirmation	A completed candle closes in the intended direction.	The next move may still reverse.

In the Tesla example, the presenter describes an upside break followed by a bullish retest candle. In the AMD example, selling appears during the retest, so he waits for an additional bullish close. This shows that “retest” alone is not the complete discretionary entry rule.

Waiting has a tradeoff. It can remove some early failed signals, but it may also worsen the entry price or miss a move. Whether that improves net expectancy is a research question, not something a winning example resolves.

A Decision Sequence You Can Apply to Replay

Is the premarket range available and fixed? If not, skip the session.
Has price broken a boundary? If not, continue observing.
Has a later candle returned to that boundary? If not, do not relabel the breakout candle as its own retest.
Does the completed candle meet your written confirmation rule? If not, wait or invalidate the setup according to rules decided in advance.
Can you specify the stop, target, and realistic execution costs before entry? If not, the setup is incomplete.
Record the decision, including valid setups you skipped. Then observe the outcome.

An Original Worked Example: Risk Comes Before the Dollar Goal

Our hypothetical share example: the premarket high is $50.00. Price closes above it, and a later bullish retest candle closes at $50.30 with a low of $50.08. Suppose the intended stop is $50.05 and a previously known target is $50.80.

At an entry of $50.30, planned price risk is $0.25 per share and potential reward is $0.50: a 2R opportunity before costs. If the next available fill is $50.40 while those absolute levels stay fixed, risk becomes $0.35 and reward becomes $0.40, or about 1.14R. The setup has not changed, but the execution has.

With $40 of planned price risk and a $0.25 stop distance, the arithmetic gives 160 shares. That requires about $8,048 of stock exposure at $50.30. A small intended loss does not imply a small capital requirement. A worse fill, gap, or stop slippage can also increase the realized loss beyond the plan.

R means the initial planned risk unit. It lets you compare trades of different sizes. It does not guarantee that every loss will stop at exactly −1R.

Why Drawdowns Become Harder to Recover

The presenter correctly emphasizes that recovery percentages are larger than the preceding percentage losses. Our calculation is: required recovery = loss fraction ÷ (1 − loss fraction).

Account loss	Gain needed to recover
5%	5.26%
25%	33.33%
50%	100%

This arithmetic supports controlling losses. It does not establish that this particular chart setup has a low loss rate.

The Option Contract Is a Second Decision

The presenter prefers contracts expiring on the relevant Friday and a nearby out-of-the-money strike. For a call, out of the money means the strike is above the stock price; for a put, the strike is below it. “One strike away” depends on the strikes actually listed in the option chain.

The transcript’s Apple illustration is internally inconsistent: at a stock price of $117.18, if both $117 and $116 puts are listed, $117 is the nearest out-of-the-money put, although the narration selects $116. The displayed chain was not supplied. Use the actual chain to identify the nearest strike rather than treating those narrated numbers as a universal rule.

A standard equity option commonly represents 100 shares, but this does not mean it produces the same price exposure as owning 100 shares. Delta measures its immediate sensitivity to the underlying; that sensitivity changes. Time decay and changes in implied volatility can also alter the premium while the stock moves.

Likewise, the strike is a contract term, not a promised destination. Before expiration, a call can rise while the stock remains below its strike. At expiration, profitability also depends on the premium paid.

A stock-chart stop cannot be translated into an exact option-dollar loss without contract prices and execution assumptions. With an outright purchased option, the premium paid is the amount that can be lost if the contract becomes worthless; it is distinct from an intended earlier stop loss. Bid–ask spreads and liquidity affect the ability to exit near the observed quote.

What the Reported Option Returns Tell Us

The presenter reports approximately a 60% option move in the Tesla example and more than 100% in the AMD example. These are claims about selected trades, not a supplied distribution of results.

Our arithmetic confirms that $416 × 60% = $249.60 before costs. That does not show that $416 was an executable position size: contracts are discrete, the specific premium is absent, and entry/exit quotes are not provided. Nor does it imply that only $30–$50 was at risk. The source’s small-risk framing is not reconciled with actual option stop fills in the supplied transcript.

To evaluate repeatability, we would need dates, contract identifiers, entry and exit prices, spreads, fees, losing trades, skipped setups, and a defined sample period. The transcript also contains unusual or potentially corrupted stock-price numbers. They are not suitable inputs for reconstructing the examples.

What This Does and Does Not Establish

The transcript supplies a coherent discretionary workflow and two narrated winning examples. It also records the presenter’s statement that he took those trades live. Neither the live executions nor a complete account history were supplied for independent assessment.

It does not provide a reproducible win rate, net expectancy, drawdown, losing streak distribution, or evidence of consistent $250 daily earnings. The examples can teach what the presenter looks for; selected successes cannot tell us how often that opportunity fails.

Expectancy connects frequency and payoff: E = win probability × average win − loss probability × average loss. In our illustrative model, 40% wins at +2R and 60% losses at −1R yield +0.20R per trade before costs. Average costs of 0.25R would turn that example negative. These figures are invented for teaching, not measured strategy results.

Turning the Retest Into a Testable Baseline

Research question: under written rules, does a confirmed premarket break-and-retest on one-minute US equity charts produce positive net expectancy out of sample?

The source leaves “strong price action,” retest tolerance, setup expiry, and trading frequency partly discretionary. We define them below so two reviewers can identify the same signals.

Source rule	Our experimental definition	Implementation shortcut
Mark the premarket high and low.	04:00–09:29 Eastern; require an observed 04:00 bar.	Use available chart bars; missing bars cannot be reconstructed.
Break a boundary, then retest.	A crossing close arms a setup; a later candle’s relevant extreme must lie within five ticks of the boundary.	One active breakout; it expires after 20 bars or a close back through the boundary beyond tolerance.
Wait for strong directional price action.	Directional candle, close outside the range, body at least 50% of its range.	One qualifying confirmation candle; the AMD-style multi-candle judgment is omitted.
Use a candle-based stop and worthwhile reward.	Signal-close-to-extreme distance plus one tick, with a 2R target.	Translate both distances to the actual fill instead of retaining the absolute candle stop; omit chart-level targets.
Trade options on momentum names.	Test the underlying stock’s directional behavior.	One share per trade; at most one submitted entry per day, morning signals by default, no option pricing model.

The stop shortcut is intentional and material: a gap shifts the bracket away from the original candle extreme. This baseline asks about an entry pattern with a fill-relative payoff model. It is not a faithful reconstruction of the presenter’s candle-stop executions.

Pine Script Research Strategy

//@version=6
strategy("Premarket retest: research baseline", overlay=true,
     initial_capital=10000, pyramiding=0,
     default_qty_type=strategy.fixed, default_qty_value=1,
     commission_type=strategy.commission.percent, commission_value=0.01,
     slippage=2, margin_long=100, margin_short=100,
     process_orders_on_close=false, calc_on_order_fills=false,
     calc_on_every_tick=false, use_bar_magnifier=true)

// Use standard one-minute US equity candles with extended hours enabled.
if not timeframe.isminutes or timeframe.multiplier != 1
    runtime.error("Use a standard one-minute chart with extended hours.")

string tz = "America/New_York"
float rr = input.float(2.0, "Target R", minval=0.5, step=0.5)
int waitBars = input.int(20, "Retest expiry in bars", minval=1)
int tolTicks = input.int(5, "Retest tolerance in ticks", minval=0)
int bufferTicks = input.int(1, "Stop distance buffer in ticks", minval=0)
float bodyMin = input.float(0.5, "Minimum body / range", minval=0, maxval=1, step=0.1)
int cutoffHour = input.int(11, "Last signal hour ET", minval=9, maxval=15)
int cutoffMinute = input.int(0, "Last signal minute ET", minval=0, maxval=58)
bool allowLong = input.bool(true, "Longs")
bool allowShort = input.bool(true, "Shorts")

int etMinute = hour(time, tz) * 60 + minute(time, tz)
int dayKey = year(time, tz) * 10000 + month(time, tz) * 100 + dayofmonth(time, tz)
bool newDay = bar_index == 0 or dayKey != dayKey[1]
bool pre = not na(time(timeframe.period, "0400-0930:23456", tz))
bool regular = not na(time(timeframe.period, "0930-1600:23456", tz))
var float pmHigh = na
var float pmLow = na
var bool sawStart = false
var bool traded = false
var int direction = 0
var int breakBar = na
var int riskTicks = na

if newDay
    pmHigh := na
    pmLow := na
    sawStart := false
    traded := false
    direction := 0
    breakBar := na
    // Do not erase riskTicks if an exceptional overnight position exists.

if pre
    sawStart := sawStart or etMinute == 240
    pmHigh := na(pmHigh) ? high : math.max(pmHigh, high)
    pmLow := na(pmLow) ? low : math.min(pmLow, low)

// Only the available 04:00-09:29 bars define the range; it freezes at 09:30.
bool ready = sawStart and not na(pmHigh) and not na(pmLow)
int cutoff = math.min(cutoffHour * 60 + cutoffMinute, 958)
bool eligible = regular and etMinute < cutoff and ready and not traded and strategy.position_size == 0
float tol = tolTicks * syminfo.mintick
float span = high - low
bool strong = span > 0 and math.abs(close - open) / span >= bodyMin

if eligible
    // One active breakout. An opposite breakout can replace it after invalidation.
    if direction == 1 and (bar_index - breakBar > waitBars or close < pmHigh - tol)
        direction := 0
    if direction == -1 and (bar_index - breakBar > waitBars or close > pmLow + tol)
        direction := 0

    bool longRetest = direction == 1 and bar_index > breakBar and low <= pmHigh + tol and low >= pmHigh - tol and close > pmHigh and close > open and strong
    bool shortRetest = direction == -1 and bar_index > breakBar and high >= pmLow - tol and high <= pmLow + tol and close < pmLow and close < open and strong

    if longRetest or shortRetest
        // Our proxy: signal-to-extreme distance, translated to actual fill.
        float distance = longRetest ? close - low : high - close
        riskTicks := math.max(1, int(math.ceil(distance / syminfo.mintick)) + bufferTicks)
        int targetTicks = math.max(1, int(math.round(riskTicks * rr)))
        if longRetest
            strategy.entry("L", strategy.long)
            strategy.exit("LX", "L", loss=riskTicks, profit=targetTicks)
        else
            strategy.entry("S", strategy.short)
            strategy.exit("SX", "S", loss=riskTicks, profit=targetTicks)
        traded := true
        direction := 0
    else if direction == 0
        // A crossing close is necessary; wick-only breaks do not arm a setup.
        if allowLong and close > pmHigh and close[1] <= pmHigh
            direction := 1
            breakBar := bar_index
        else if allowShort and close < pmLow and close[1] >= pmLow
            direction := -1
            breakBar := bar_index

// Fixed distances persist; later candles cannot move the bracket.
if strategy.position_size > 0 and not na(riskTicks)
    strategy.exit("LX", "L", loss=riskTicks, profit=math.max(1, int(math.round(riskTicks * rr))))
if strategy.position_size < 0 and not na(riskTicks)
    strategy.exit("SX", "S", loss=riskTicks, profit=math.max(1, int(math.round(riskTicks * rr))))

// Full US sessions only. Holiday early closes need a separate calendar rule.
// At the close of the 15:59 bar, simulate a closing-tick liquidation.
// If that bar is missing, close at the first observed later bar instead.
bool flatten = (regular and etMinute == 959) or (not regular and etMinute >= 960)
if flatten
    strategy.cancel_all()
    strategy.close_all(comment="Session end", immediately=true)
    direction := 0

plot(regular and ready ? pmHigh : na, "Premarket high", color.green, style=plot.style_linebr)
plot(regular and ready ? pmLow : na, "Premarket low", color.red, style=plot.style_linebr)

This code tests one objective interpretation of the source’s concept. Any mechanical definitions introduced here are ours and are not necessarily the creator’s exact rules.

Use a standard one-minute US stock chart, enable extended-hours data, and use full regular sessions. Paste the complete script into TradingView’s Pine Editor and add it to the chart. The default commission is an illustrative 0.01% per filled order and slippage is two ticks; change them in Strategy Properties to match the market being researched.

What This Code Actually Tests

It freezes the available premarket range, arms a breakout only after a crossing close, and waits for a later directional retest candle meeting the tolerance and body-size rules. A close too far back across the boundary invalidates the pending setup. Expired or invalidated setups can be replaced by a new crossing; a still-valid setup is retained. This selection rule affects which retests enter the sample.

Only the first qualifying entry submitted each day is included, even if the emulator rejects it for insufficient capital. The code holds at most one position, uses one share, and closes at session end. It tests both directions, with independent switches.

Execution Assumptions That Change the Meaning of the Test

Signal and fill: the completed confirmation candle creates a market order. With the default settings, it becomes eligible on the next available tick, normally the next bar’s open. The signal close is not treated as the fill.
Bracket: relative tick-based stop and target instructions are submitted with the entry, then retained. TradingView derives their prices from the filled entry. Tick rounding makes the requested R ratio approximate; costs and exit slippage further change realized R.
Gaps: the fill-relative bracket avoids a target accidentally lying behind the entry, but does not reject a next-bar gap that makes the original chart setup unattractive. A gap through an active stop can produce a worse exit.
Intrabar sequence: Bar Magnifier is enabled. Where lower-timeframe history is unavailable, broker-emulator assumptions can determine the result if both bracket levels occur within one bar. Flag these trades as execution-sensitive.
Sizing: one share is an experimental normalization. It is neither a fixed $40 risk model nor a $250 income model. Stop distance, fees, and gaps determine its dollar outcome.
Session close: closing-tick liquidation is an explicit approximation. Missing 15:59 bars cause a later observed close; half-day sessions require a separate calendar rule and should be excluded from this baseline.

The tick-based bracket mechanism and its relationship to entry prices are documented in TradingView’s strategy FAQ (paid link). This documentation supports the execution mechanism, not the strategy’s profitability.

Validation status: This script has not been compiled in TradingView or run against historical data here. No performance results are claimed. Its order timing, state resets, fixed brackets, and source-to-proxy mapping have been reviewed.

What This Code Does Not Test

Weekly option premiums, delta, time decay, implied volatility, option spreads, or discrete contract sizing.
The presenter’s precise candle-extreme stop and intraday-high or psychological-level targets.
Discretionary multi-candle confirmation, live judgment, watchlist construction, or market-wide momentum.
The claim that gap-up days make these levels more reliable.
All setups in a day, early-close sessions, unavailable premarket history, or the feasibility of a daily income target.

Backtest Lab

Save a baseline before changing anything. Choose a fixed development period and a later untouched evaluation period. Use the same symbols, dates, chart session, quantity, and costs for each paired comparison. Record trades, eligible breakouts, confirmed retests, rejected entries, net expectancy, average win/loss, profit factor, maximum drawdown, and session-level results.

Experiment	Change only	Question and interpretation
Confirmation strength	Minimum body/range: 0.0, 0.5, 0.7	Does stronger confirmation change net expectancy? An improvement per trade with far fewer triggered setups may reflect selection, so also measure net R per eligible breakout.
Retest width	Tolerance: 1, 5, 10 ticks	Does the result depend on an unusually precise boundary? Similar behavior across adjacent values is more encouraging than one isolated peak. Tick widths differ economically across stocks.
Target distance	Target: 1R, 2R, 3R	Does the entry benefit persist across payoff choices? Keep stop-distance rules and costs fixed; measure expectancy rather than choosing the highest win rate.
Waiting window	Expiry: 5, 20, 40 bars	Do delayed retests differ? Longer expiry also changes which setups remain active, so record the resulting signal population.
Execution sensitivity	Slippage: 1, 2, 5 ticks	Does apparent profitability survive less favorable fills? Keep commission fixed for this comparison, then test commission separately.

A separate gap-day study: before looking at results, define an up-gap as the regular-session open at least 0.5% above the previous regular-session close. This threshold is ours. Group every eligible session into up-gap versus other days, retain identical entry and exit rules, and compare net expectancy, trade count, and failed-breakout frequency. Higher performance in a small subgroup suggests a hypothesis for another untouched period; it does not prove the presenter’s reliability claim. This grouping is not implemented in the baseline script.

For confirmation comparisons, count eligible breakouts even when no trade occurs. The baseline does not export those counts; record them in replay or add explicit counters before claiming results per setup. Define how you treat missing data, simultaneous stop/target touches, and early-close days before testing.

A broad stable region is more persuasive than one exceptional setting surrounded by poor results. Repeatedly revising parameters on the evaluation period turns it into development data. Keep a final period genuinely untouched.

Try It Yourself

In chart replay, review 20 consecutive full sessions without looking ahead. Mark the premarket boundaries, every crossing close, the pending setup’s status, and the first qualifying retest. Save the decision before revealing the next bar. This small exercise checks whether the written rules are reproducible; it is not enough to establish an edge.

Check Your Understanding

Why does the breakout candle not count as its own retest here? The experiment requires a later return after the breakout has been observed.
Why can a trade drawn as 2R become less attractive? A worse fill changes the distances to fixed chart levels.
Does controlling 100 shares mean a contract behaves like 100 shares? No. Contract size and delta exposure are different.
Does investing $416 in a premium mean only $40 is at risk? No. An intended stop loss and the premium that can be lost are separate quantities.
What does the script’s translated stop change? It preserves a selected risk distance from the fill while moving the absolute stop away from the original candle extreme.
What would two winning examples need before supporting a profitability claim? A complete, predefined sample with losses, costs, executable prices, and evaluation on unseen data.

Final Takeaway

The video teaches a compact way to organize attention: premarket boundaries, a breakout, a retest, and directional confirmation. Its deeper contribution is a decision process with an identifiable failure point.

The supplied material does not establish predictable daily earnings or typical option returns. The useful next step is to test explicit rules on the underlying, then investigate whether real option execution preserves any observed advantage. Keep the chart idea, execution model, and evidence distinct.

Risk Disclosure: This material is for educational and research purposes only and is not financial or investment advice. Trading can result in substantial losses; purchased options can lose their entire premium. Backtests are hypothetical, depend heavily on their assumptions, and cannot reproduce every aspect of live execution. Commissions, spreads, slippage, liquidity, gaps, order behavior, market regime changes, and discretionary decision-making can materially affect real-world results. Historical or hypothetical performance does not guarantee future results.

This article is an independent educational companion to the video and is not affiliated with, sponsored by, endorsed by, or authorized by Scarface Trades.