Best chart patterns for scalping crypto: 1m to 5m (2026)

How ChartScout helps you watch low-timeframe patterns

Low-timeframe monitoring becomes difficult when each additional pair needs frequent chart checks. A Pattern Watcher handles one exchange, pair, timeframe and pattern, sending detections through the platform, Telegram, Discord or email. Two patterns on the same pair require two watchers.

For discovery beyond that list, the Pattern Screener shows published futures detections with a chart image and exchange link. It sends no notifications. Coverage is limited to pairs with more than $1M in 24-hour volume on Binance, Bybit, KuCoin, MEXC and Hyperliquid; the current spread still needs your review.

Alerts show confidence and maturity alongside the detected structure. Forming patterns typically trigger around 75 to 80% maturity, before breakout completion. Neither score is a profit probability, and maturity does not measure elapsed time.

Both 1m and 3m watchers require Enterprise or Enterprise+. Only Enterprise+ includes the 1m screener. Basic covers none of these scalping timeframes and excludes cup and handle. The 7-day Pro trial requires a card and covers 5m, not 1m or 3m. Its screener has a 15-minute delay; watchers have no intentional delay. Check the plan limits.

The public Discord feed and @ChartScout_bot on X provide timestamped, curated examples. For your own one-week review, log bull and bear flags on one timeframe, recording invalidation, target, round-trip cost and outcome. ChartScout uses public market data without API keys or custody; you place any trade yourself.

The best chart patterns for scalping to learn first

Bull and bear flags give you a focused practice set on 1m, 3m or 5m: an impulse, a compact consolidation and boundaries you can use to define invalidation. Pennants share that consolidation idea, but we have no published pennant study.

Cup and handle reached its own target most often in our sample, at 33.36% to 36.18% across 1m to 5m. That does not make it the strongest scalp: smaller targets leave less room for costs.

Our three studies cover 39,408 setups across 59 Binance USDT-margined perpetual markets, not executed trades. These separate samples are descriptive, not predictions, and ignore entries and fees. Post-breakout rates cannot be assigned to an unfinished alert. Scalping is high risk. Our wider chart pattern guide provides broader context.

Why a 5% target misreads a 1m pattern

A fixed percentage threshold tests move size as well as structure. In our sample, moving from 1m to 1h changes the result dramatically.

Bull flags' 5% attainment rises from 10.69% to 39.04%, while full-pole attainment shifts from 22.26% to 21.12%. Bear flags move from 8.85% to 46.30% on the 5% test, versus 24.31% to 32.10% on their pole target. Cup and handle moves from 6.40% to 46.96%, versus 33.74% to 39.13% on its depth target.

The intermediate 3m and 5m rows show the same distinction: the percentage clearing a fixed 5% move rises more sharply than the percentage reaching the pattern target. Asking a small structure for 5% measures something different from asking it to complete its measured move.

Judge low-timeframe setups against their own target and its distance from your planned entry. These descriptive comparisons do not isolate timeframe as the cause, and reaching a target does not establish a strategy's return.

What makes a pattern fit a scalp

Favor compact structures with usable space between entry, invalidation and target. The pattern should complete in few candles, rather than require a prolonged sequence of touches.

Invalidation should sit near entry but outside ordinary candle noise. A tight stop is not useful if a routine wick crosses it without breaking the idea.

The target must be clearly larger than your round-trip cost. Measure from the entry you can obtain, not a breakout price already behind you.

The pair should have a tight spread at your order size. ChartScout's volume filter narrows coverage, but it cannot establish what your particular fill will cost.

Flags and pennants on 1m to 5m

Count candles before minutes. A hypothetical SOL/USDT flag spanning about 15 candles takes about 15 minutes on 1m, 45 minutes on 3m and 75 minutes on 5m. That is formation time, not a forecast holding period.

For a bull flag, structural invalidation sits below the flag low; for a bear flag, above its high. Allow for ordinary noise before deciding whether the stop distance remains usable. The pole target projects the preceding impulse's height from the breakout or breakdown, not a guaranteed move.

In our sample, full-target rates on 1m, 3m and 5m were 22.26%, 24.61% and 23.46% for bull flags, versus 24.31%, 24.36% and 25.53% for bear flags.

Pennants replace parallel flag boundaries with converging ones. Their similar shape does not justify borrowing flag percentages: we have not published a pennant study.

Cup and handle on low timeframes

The highest target-hit percentage does not mean the largest available move. In our sample, the median cup was 1.63% deep across timeframes, so its full-depth target was often smaller than 5%. Median favorable rises were 0.58% on 1m, 0.73% on 3m, 1.04% on 5m and 3.94% on 1h.

Almost every breakout returned near its neckline under the study's throwback definition: a low within the larger of 0.5 x ATR% or 0.3% of the neckline within 30 candles. In our sample, counts were 6,934/6,980 on 1m, 2,389/2,419 on 3m and 1,283/1,302 on 5m; median timing across timeframes was 1 candle.

That tolerance is substantial relative to a 0.58% median rise. The count measures a return near the neckline, not a failure rate. A stop at the neckline sits inside the area price revisited. Decide before entry where the structure is actually wrong, rather than treating every touch as failure. Our stop placement guide explains the structural approach.

Patterns to treat with care on fast charts

Treat head and shoulders, double tops and bottoms, and wedges or triangles with many touches as a separate study list. They can require more candles than a compact flag and are often labeled before the final swing or breakout exists.

We have not published outcome studies for those patterns. That is an evidence gap, not proof they fail on 1m. Ask anyone quoting a 1m success rate for the sample, completed-pattern definition and endpoint.

A forming ChartScout detection is a reason to inspect the structure, not call it confirmed. Keep your exchange chart for validation: ChartScout is not a candlestick charting terminal. Our fake breakout guide covers confirmation separately.

What costs do to a small move

A hypothetical 0.10% round-trip cost, not any exchange's fee, consumes the following shares of the median cup-and-handle rises in our sample:

The study measures from the open of the candle after the breakout close to the highest close before the first qualifying reversal, a close below the handle bottom, or the time cap. The resulting rise is a favorable excursion identified in hindsight, not a return earned by a trade. It does not establish how much of the move a real entry and exit would capture.

• Cost share (%) = round-trip cost (%) / planned gross move (%) x 100.
• Approximate move remaining after costs (%) = planned gross move (%) minus round-trip cost (%).

Include entry and exit fees, spread and slippage, plus funding if incurred. Use the remaining move from your planned entry, not the historical median. Pass on a setup when the remaining move cannot justify the stop distance and execution uncertainty. Our risk and reward math develops the calculation.

How to choose between 1m, 3m and 5m

Stay on 5m if a 1m target leaves too little room after costs, or you cannot review alerts promptly. Consider 3m when its target and your screen time fit. Use 1m only when both checks pass, not because it produces more activity.

Our bull flag study treats 1m as secondary and emphasizes 5m and above. No timeframe is best for everyone.

ChartScout's first-party measurements from October 1 to 9, 2026 track candle close to publication:

These are measurements, not guarantees or phone-delivery times. Exchange data, processing and your connection add time; your reaction adds more. Recheck remaining target room when an alert arrives. Choose the timeframe that survives your cost and attention checks.

Frequently asked questions