Whale alerts arrive at unpredictable times, describe large sums of money and often coincide with fast price moves. That combination is good at producing urgency and bad at producing understanding. A routine fixes this by deciding in advance what you will do with each alert: record it, ask a question, check the outcome later and review the pattern. Done consistently, a stream of alerts becomes a dataset you can learn from instead of a feed you react to.
This last part of the series turns the earlier guides into a workflow for SuperQuant’s whale alerts in Telegram. Part one explains what the alerts are built from; this guide is about what to do when they arrive.
Choose your filters before the first alert
Decide what you want to study before the alerts start, not after a dramatic one. SuperQuant’s guide to whale tracking lists the controls: after linking Telegram, send /whales on.
What arrives then depends on your access tier, and /whales tier lists what each one includes. According to the guide, Free accounts get the daily Whale Wrap, on by default, plus /whales top and /whales net; /whales alerts wrap off stops the Wrap. Holder (during the trial, or with a wallet that passes the access gate) adds real-time moves, the liquidation radar, positioning shifts and more. By default, the guide says, alerts start at changes of at least $1M by any of the top 100 tracked accounts, up to 24 whale alerts a day.
- Narrow the scope.
/whales coins BTC ETHlimits alerts to the coins you actually research./whales min 500kchanges the size floor, and/whales limit 10changes the daily cap. - Narrow the accounts.
/whales topshows the current list, which the guide says is ranked by 30-day profit, and/whales top 20limits alerts to the top 20. Style and hit-rate filters (/whales style,/whales minhit) need Holder and use the scorecard fields described in part four. The guide says a trader’s 24-hour hit rate is shown once 10 of their moves have been measured, which is still a small sample. - Know what each type means. Moves report opens, closes, flips and large resizes. The liquidation radar, as the guide describes it, watches positions of at least $5M and alerts when the mid price comes within 5% of the venue’s estimated liquidation price; part two explains how to read that distance. Positioning shifts, listed as
consensusin/whales alerts, report a flip in the tracked accounts’ net side, a surge in net positioning, or a crowd add, which the guide defines as six or more top-20 traders adding in the same direction within an hour; part three covers them. Optional types include entry zones on Holder and, on Pro, resting orders and against-you alerts, which flag top traders taking the other side of your own positions.
Fewer, better-scoped alerts are easier to review honestly than a constant stream.
Record the alert before the chart
When an alert arrives, copy its facts into your journal before opening a chart: the detection time, coin, side, size, the account’s rank and the alert type. The detection time is when the tracker saw the change; the guide says the top 100 are checked about once a minute, so the trade itself may be a little earlier. The order matters. Once you have seen what price did in the last few minutes, it becomes hard to write down what the alert said without a story attached.
Alerts are also time-sensitive by design. The guide says a real-time alert older than 30 minutes when it is due to be sent is dropped, with a two-hour limit for positioning shifts and against-you alerts. An alert you read hours later is a historical record, so log it as one. The guide also says the public whale channel posts with a one-hour delay, which makes it suitable for review and unsuitable for reaction.
Write the question first
Turn each alert into one question with a measurable answer, written before you look at what happened next. For example, using fictional alerts:
- “After a top-20 account opened a $12M SOL short, was SOL below the detection price 24 hours later?”
- “After a positioning flip to net long in ETH, did the 24-hour return beat ETH’s average 24-hour return for the month?”
- “After a liquidation radar alert with the mid 4 percent from the estimated liquidation price, did that distance shrink or widen within an hour?”
A question like this can be answered either way. “Is this a good trade?” cannot, which is why it does not belong in the journal.
For a timestamped record inside the bot, SuperQuant’s methodology page says /decide SYMBOL long, short or flat records a timestamped prospective decision against the last hourly close, places no trade and is graded after 24 hours. Because it measures from that hourly close, not the detection price, note both.
Check fixed horizons
Check every logged alert at the same horizons: one hour, 24 hours and seven days after the detection time, measured from the price at detection. SuperQuant’s guide says each top-trader move is measured at +1h, +24h and +7d from the price when it was detected, so your notes and the tracker’s scorecards use the same yardstick. Include alerts where nothing happened. A journal that only records the alerts that looked right will make any rule look better than it is.
The guide to keeping a crypto trading journal has a template for separating what you knew beforehand from what you observed afterwards. Whatever the bot stores about an alert, keep your own record: only you know what you expected before you saw the outcome.
Review weekly
Once a week, look at the log as a whole:
- How many alerts of each type did you record?
- For each question type, how often did the outcome go the way the alert seemed to suggest? Compare that with a base rate: how often the same outcome happened across all comparable windows for the same coin and week, alert or not.
- Which alerts led you to act, or want to act, and what happened in those cases compared with the rest?
- Which filters should change, and why?
The weekly trading review guide describes how to separate following a plan from getting a good outcome.
Guardrails
- Watch for urgency. A 2024 Psychology & Marketing paper reporting eight experiments found that externally evoked fear-of-missing-out (FOMO) appeals influence consumers’ investment decisions. It also found that FOMO appeals lead consumers to repeated investment decisions, even if prior losses have been incurred. Large-dollar alerts are designed to be noticed; your routine is there to slow the decision down. The guide to reviewing FOMO and revenge trades has a checklist for those moments.
- Treat outside “whale signals” with suspicion. Be wary of any paid group that promises returns from copying whales. The CFTC’s investor alert says that claims such as “guaranteed profit” are hallmarks of a fraud, and that fraudsters may try to create a false sense of urgency; the guide to crypto trading signal scams covers how such schemes work.
- Remember what the alert is. Every whale alert describes positions taken from public data. As the guide puts it, these are third-party positions, not advice, and nothing in the alert places an order. This is educational research context, not investment advice.
That completes the series. To see how the Telegram bot fits into a broader research process, read the guide to a Telegram signal-bot workflow.