Chart pattern scanners: how they work and what to look for

What a scanner does

A detector answers one question: does this chart, right now, contain a pattern that has just broken out? A scanner asks that question over and over, across a list of symbols and timeframes, and tells you when the answer changes. The detection logic is the same as on a single chart (see automated chart pattern detection); the scanner's job is breadth, speed and delivery.

Breadth: symbols × timeframes × venues

The workload multiplies rather than adds. For example, 50 symbols on 3 timeframes is 150 chart series. Add a second exchange listing the same 50 symbols and it becomes 300. Each series needs fresh candles before detection can run.

Breadth is only worth having if the extra series are ones you would trade. A scanner that covers thousands of symbols will surface many patterns on thin instruments, where the spread and slippage eat the move. A narrow, deliberate watchlist usually produces fewer and more tradable signals.

Timeframes deserve the same thought. Each timeframe is scanned on its own, so the same symbol can show a bullish flag on the 5-minute chart and a bearish wedge on the 4-hour chart. Many scanners use the higher timeframe as context, as in multi-timeframe analysis, rather than requiring every timeframe to agree.

Cadence and rate limits

Cadence is how often the scanner re-checks each series, and it matters most on short timeframes. With a 120-second cadence, up to two 1-minute bars can pass between checks, so a breakout on the 1-minute chart may be well under way, or already over, by the time it is seen. On a 4-hour chart the same delay hardly matters.

Faster cadence has a cost: data requests. Exchanges and brokers limit how many requests an API key or IP address can make in a given period. For example, refreshing 300 series every 10 seconds means 30 requests a second if each series needs its own request, which can quickly run into those limits. Scanners cope by batching, caching candles and spacing requests, and many tie faster cadences to higher plans. When you evaluate a scanner, ask for the actual cadence on your plan and timeframes rather than relying on the word "real-time".

Data quality

A scanner can be no better than its candles. Common problems:

• Missing or late bars, which shift swing points and distort lookback windows.
• Bad ticks: a single erroneous print that creates a fake high or low.
• Thin markets, where a handful of trades forms a "pattern" that can't be traded at the prices shown.
• Different venues, different candles. The same pair can print slightly different highs and lows on two exchanges, so a pattern on one may not exist on the other.

Good scanners say where their data comes from and apply minimum liquidity and spread conditions before reporting anything.

Alerts vs automation

Scanners deliver results in three broad ways:

1. Alerts only. You get a notification and decide everything else. You keep full control, but you need to be available, and the setup may have moved by the time you act.
2. Approval-based execution. The scanner prepares the order (entry, stop, target and size) and waits for one tap to approve or reject. It is fast, with a human in the loop.
3. Automatic execution. Signals go straight to the broker. This is the fastest option and the most dependent on good risk controls and on the scanner's rules being what you want.

There is no universally right choice. Many traders start with alerts or approval and automate only setups they have watched for a while.

Signal expiry

A breakout signal is a snapshot. Its entry, stop and target were calculated from prices at one moment, and as price moves away, the reward-to-risk ratio changes. For example, a long setup with entry 50.00, stop 49.00 and target 52.00 offers 2 to 1. If you only see it after price has run to 51.00, the same stop and target offer 1 to 2. A scanner that keeps old signals live invites you to chase. Signals should expire, and the expiry should be short relative to the timeframe.

What to look for in a scanner

• Coverage of your venues, symbols and timeframes.
• A stated scan cadence for your plan.
• Published detection rules, so you know what "bull flag" means in this tool.
• Market-quality filters for liquidity, spread and volume.
• Clear delivery options (in-app, push, messaging apps) and whether you can approve from them.
• Automatic expiry of stale signals.
• A paper mode to see the whole flow before any money is involved.

How the PatternX scanner works

PatternX has no fixed scanning universe. It scans each combination of your connected exchanges or brokers, the symbols you track and the timeframes you choose, from 1 minute to 1 day; new accounts start with 5-minute, 15-minute and 1-hour charts. For crypto, the symbol picker offers each exchange's top 20 pairs by 24-hour volume. For traditional markets it offers a curated list of 309 instruments (US stocks, ETFs, leveraged and inverse ETFs, Bitcoin and Ether ETFs, and FX pairs including gold and silver) through your own broker connection.

Each scan fetches 200 candles per exchange, symbol and timeframe and runs all 20 detectors. A breakout must be on the latest bar, which may still be forming. The next timeframe up is checked for agreement, which changes the score but never rejects a signal. The data side is described in market data.

Signals that pass the filters and your minimum confidence appear in the app and, depending on your notification settings, in Telegram, email and web push. In manual mode, the default, each signal waits for your approval, either in the app or with Telegram's approve and reject buttons, and expires after 15 minutes: three bars on a 5-minute chart. In automatic mode it goes straight through the risk checks to execution. See automation.

Limitations

A scanner only finds what it is pointed at: PatternX won't flag a pattern on a symbol you don't track. Even a 10-second cadence is periodic polling, not tick-by-tick, so very fast moves on 1-minute charts can be missed or caught late. And scanning more symbols doesn't make any single signal more reliable.