What Cand Jump Actually Is
Cand Jump is a small utility that detects when price candles on a chart break above or below a defined threshold and fire a notification or trigger. It's mostly used by traders who watch multi-timeframe setups and don't want to sit glued to a screen. The core idea is simple: watch candle close data, compare it against rules you set, and log or alert when something matches. Nothing dramatic about it. You can find it on GitHub under the standard public repos. Download the latest release zip from the releases tab. Extract it to wherever you keep your tools. Open a terminal or command prompt in that folder and run npm install if it's Node-based, or just double-click the executable if it's a standalone build. Some forks include a config file called config.json or settings.yml right in the root directory. That's where all your parameters live. Don't skip reading it before launching anything. Most Cand Jump implementations pull candle data from a WebSocket or REST endpoint on your exchange or data provider, then run a loop that checks each new candle close against your conditions. The conditions are typically things like "close price exceeded the previous high by X percent," "volume spiked above the 20-candle average," or "price jumped the opposite direction of the trend within Y minutes." Once a condition triggers, it writes to a log, sends a webhook, or prints to console depending on your config.
I spent a few hours last year trying to get one of the earlier builds working with a Binance stream because I was tracking specific volatility expansions on the 5-minute chart. The problem was that the default polling interval was set to 3 seconds, which meant on slower connections or during high-volume periods, candles would get processed late and the alert would fire after the move was already over. The fix was straightforward: I changed the interval to 1 second and switched from REST-based candle fetching to the WebSocket feed, which cut the latency from roughly 4-5 seconds down to about 800 milliseconds. That made a real difference when I was scalping.
Configuring It Without Breaking Anything
The config file is where most people mess this up. Here's what matters. The symbol field accepts the trading pair, usually in the format like BTC/USDT. The timeframe should match the exchange's supported intervals: 1m, 5m, 15m, 1h, etc. Setting a timeframe the exchange doesn't support will cause silent failures where no candles come through and you'll wonder why nothing is triggering. Make sure the timeframe is valid first. The threshold parameter controls how much of a price movement qualifies as a jump. I've seen people set this to 0.001 and wonder why they're getting flooded with alerts. On a volatile pair like SOL/USDT, a 0.1 percent jump happens dozens of times per hour. Start with something like 0.5 or 1.0 for a 5-minute chart and dial it down only if you're getting too few signals. There's no universal correct number. Another thing that catches people out: the lookback period. This determines how many previous candles the tool examines for comparison. If you set lookback too low, say 5, you'll get a lot of false positives because a single green candle can look like a jump compared to five flat ones. A lookback of 20 to 50 is usually more representative. I keep mine at 30 on the 5-minute charts I trade.
Get the Full Details
Common Pitfalls I've Hit
The biggest issue I ran into was stale data. When the exchange API has a brief downtime or the WebSocket reconnects, Cand Jump doesn't always handle the gap gracefully. It might resume processing from the last known candle and skip everything that happened during the outage, which means you miss a jump that occurred 10 minutes earlier. The workaround is to add a max_latency check in the config if your version supports it, or simply refresh the session manually once an hour during active trading. Some people build a wrapper script that pings the exchange health endpoint and restarts the tool if the connection drops, but that's extra work you may or may not want to do. A second issue is multiple timeframe confusion. If you set Cand Jump to monitor both the 1-minute and the 5-minute chart on the same symbol simultaneously, you'll get duplicate alerts because the 1-minute will trigger first and the 5-minute will fire later as that candle closes. I disable the shorter timeframe and just let the tool run on 5m and 15m. Cleaner output, fewer wasted notifications.
Advanced Usage: Webhooks and Alerts
If you're serious about this, configure the webhook output to send to a service like Discord or Telegram. The JSON payload typically includes the symbol, the candle close price, the threshold that was breached, and the timestamp. I set up a Discord bot that filters messages by keyword so I only get alerts for specific pairs. This saves you from checking your phone every three minutes during slow market hours. There's also a lesser-known feature in some builds where you can export the signal data to a CSV file for backtesting. This is useful if you want to verify whether your threshold settings are actually producing meaningful signals over a longer period. Run it against historical data, export the results, and check how many of the triggered jumps actually led to favorable price movement afterward. Most people skip this step and jump straight into live trading, which is where the learning curve gets steep.
Is Cand Jump Worth Using?
It depends on what you're doing. If you're a discretionary trader who watches charts actively, it's a minor convenience at best. You'll get alerted, but interpreting whether a jump is meaningful still requires your own judgment. If you're building an automated system or running multiple strategies across several pairs, the consistency of machine-driven detection is more valuable than the alert itself. The tool is lightweight, mostly reliable, and free, but it's not going to replace your analysis. It's a filter, nothing more. And filters work best when you know exactly what you're filtering for.
