How to read the breakdown of a screener alert
An alert breakdown is up to four short lines under the alert itself, each answering a question a trader would check by hand anyway: where in the indicators the move happened, whether one coin moved or the whole market did, whether there is a gap with the Korean exchange, and whether an exchange announcement landed close in time. The order of the lines is always the same.
In short
- An alert from any screener carries up to four breakdown lines in a fixed order: indicators, market, Korea, events.
- The breakdown describes the moment the alert fired: it is not a forecast, not a score for the alert, and not a side to trade.
- A line appears only when there is something to say. A missing line is normal, not a fault.
- In the dashboard card the same data sits in the “Breakdown” block, with more fields: a card has room, a message does not.
What is in the alert itself and what is in the breakdown
First comes what the screener itself found: a header with the pair and the window (“15 minute change”), then a block per venue where the condition fired, and inside it the metric change lines — price, open interest, turnover, funding rate. At the end comes the 24-hour range, and for the price, OI, turnover, funding, and custom screeners, the long-to-short ratio with its source exchange.
Only then come the breakdown lines. They are always in the same order, and none of them is mandatory:
| Line | The question it answers |
|---|---|
| Indicators | Where on the chart the move was caught |
| Market | Did the coin move, or did everything move at once |
| Upbit premium | How far the Korean price sits from the world price |
| Close in time | Was there an exchange announcement or a macro event nearby |
In the dashboard the same data sits in a “Breakdown” block with sections: Positions, Indicators, Market, Korea, Events. “Positions” holds what the screener found — funding, the change in OI and price, the setup score, the price a level was crossed at. A card stands for one episode of a move rather than one minute: the ×N badge says how many times the condition fired.
Indicators: where the move was caught
The line looks like this: “Indicators: 1h RSI 71, 12% above the daily SMA200”. There are two numbers, and neither is accidental: the 1h RSI answers “what now” — how stretched the move is inside the hour — while the distance to the daily SMA200 answers “against what backdrop”: whether this is happening above the long-term average or below it.
The line appears only when both numbers are there: 71 in a rising market and 71 on a bounce inside a decline are different situations behind the same number.
Where the candles come from
The source is Binance, and Bybit when the needed candles are not there. It can be a spot chart even when the alert came from a futures contract, which is why the card names the venue next to the indicators together with the market: “Binance · spot”. Candles are computed for the 500 highest-turnover pairs: for other coins this line will not appear at all.What the card adds
In the card, the daily RSI, the distance to the hourly SMA50, and volume versus average join those two numbers. A snapshot older than two candles of its own timeframe is not used: around somebody else’s signal, stale numbers are worse than none.
Market: did the coin move, or did everything
The line looks like this: “Market: 23% of coins moved the same way in 15 min”. It is a share of the market and nothing more: we deliberately draw no boundary between a coin’s “own” move and a market-wide one, and never name one as a number. The window is 15 minutes, the same as the rest of the breakdown. How the counting works:
- A coin counts as moved once it passes 1% over that window: less than that is quote noise.
- Coins are counted, not pairs. SOL against USDT and SOL against USDC are one vote; otherwise a coin with ten pairs would weigh as ten.
- Tokenised stocks are excluded. They open on the exchange schedule and gap all at once — that is the start of a session.
- A sample below 50 coins means no line. That means the collection history has not warmed up, not that the market is calm.
The line has two forms. When the direction is known it says “the same way”; when it did not resolve, it reads differently: “Market: 31% of coins moved notably in 15 min”. That is the same value counted both ways: saying “the same way” without knowing the way would be untrue. For the price screener the direction comes from the scenario that fired; for the others, from the sign of the price change over the same window, and a move under 0.2% does not count as a direction.
The line exists for the price, open interest, 24-hour turnover, funding, and custom screeners. The rest do not have it: their data carries no direction, and for liquidations the price change is stored as a magnitude — breadth read from it would come out “up” even during a downward cascade. The card shows the sample size next to the share.
How the screener helps
Everything described above is what a trader checks by hand: open the chart, look at the RSI, compare with the rest of the market, check whether there was an announcement. For one alert that is a few minutes; for a dozen in a day it is not. The breakdown comes with alerts from every screener and needs no setup: the only thing to configure is the screener itself — which move is worth an alert.
Upbit premium: the gap with the Korean market
The line looks like this: “Upbit premium: +6.31%, was +4.21% 15 min ago”. The premium is how far the Korean price of a coin sits from the world price in that same minute. The second number is the previous value, not “it grew by that much”: two numbers are clear without jargon.
Both prices come from the same minute, and that is not pedantry: compare a three-minute-old Korean price with our current one and the difference comes from the gap between collection times, not from a premium. So a price older than a minute is not used in the calculation, and data older than three minutes is not used at all. For the same reason, won is converted to dollars using the rate from the same Upbit response.
The line appears when the premium is noticeable: from a percent, or when it has shifted by a percentage point over the window. On liquid coins a small gap is always there, and printing it in every message would train people to skip the line. Korea trades about 287 won markets against roughly 4,800 tracked pairs, so most coins will never have this line. A premium, meanwhile, is a gap between two markets and nothing more: it does not predict price and does not mean the gap has to close.
Close in time: an announcement near the alert
The line looks like this: “Close in time: listing on Upbit, 4 min ago”. We read the announcement sections of Bybit, OKX, KuCoin, Bitget, Binance, Upbit and pick up listings, delistings, futures listings, and maintenance windows, and separately macro events on the official schedule: US inflation and employment data, and Fed rate decisions.
The wording is literal on purpose. “Close in time” is all that is claimed: not “because of”, not “caused by”. We measured the coincidence in time, not a causal link, and the move after an announcement can be anything, including nothing.
- An exchange announcement counts as close for a day after publication, and only after it: the market plays a listing out over hours, not minutes.
- A macro event counts for half an hour either side of the scheduled time, so the line can also say “in 20 min”: the date is known in advance, and the market moves in anticipation too. No venue is named here — no exchange announces it.
Why a line can be missing
An empty line reports nothing and takes up space — so a missing line is normal, not a fault.
The data was not there
Indicators are missing when candles are not computed for the coin, the snapshot is stale, or the second number was absent. The market line is missing when the sample did not reach 50 coins. The premium is missing when the coin is not on Upbit, the data is stale, or the gap is too small. The event is missing when nothing was announced inside the window — the most common case.Some lines never exist
The technical signals screener has no indicator line, and funding arbitrage has no premium: its alert is about two exchanges at once. In the card the rule is softer: it shows fields whenever they exist, so a half-percent premium will be in the card and not in the text.What the breakdown does not say
The breakdown carries no side to trade and no probability. None of the numbers means “buy” or “sell”, and there is no score for the alert either: the numbers describe a moment, and what to do with it is the trader’s call.
None of the four lines claims price went somewhere because of anything: indicators describe a point, the market line describes reach, the premium describes a gap between venues, and the event describes proximity in time. Signs, not causes.
“What happened after” is a median, not a forecast. On a screener page, next to the counters, sits the median price move 24 hours after an alert and the number of alerts it was computed from. It is a measurement of the past and promises nothing about the future. There is no share of successful trades there and there cannot be: the side of a trade is not stored anywhere, and a price rise after an alert would mean the opposite for a short.
The breakdown saves the trip to a chart, but it does not replace checking: the decision is still made from the chart, the order book, and your own entry rules, and the position size is calculated separately — from the arithmetic of risk.
Alert breakdown FAQ
What is the alert breakdown in JustScreener?
Up to four lines under an alert from any screener: the indicators at the moment it fired, the share of the market that moved the same way, the premium on the Korean exchange Upbit, and an announcement close in time. The order is always the same, and each line appears only when the data is there.
Why is there no indicator line in my alert?
The line needs two numbers at once — the 1h RSI and the distance to the daily SMA200. It will be absent if candles are not computed for the coin (they are computed for the 500 highest-turnover pairs), if the snapshot is older than two candles of its timeframe, or if the alert came from the technical signals screener.
What does “23% of coins moved the same way in 15 min” mean?
Over the last 15 minutes, 23% of the coins in the sample moved at least 1% in the same direction as the coin in the alert. It is the reach of the move, not its strength. Coins are counted rather than pairs, and a sample below 50 coins is not shown at all.
What does the Upbit premium mean?
How far a coin’s price in Korea sits from the world price in the same minute. The line appears from a percent of gap, or when the premium has shifted by a percentage point over 15 minutes. It is a gap between two markets and nothing more: it does not predict price.
Does “close in time: listing” mean the listing moved the price?
No. We measured the coincidence in time, not a causal link: an exchange announcement counts as close for a day after publication, a macro event for half an hour either side of its scheduled time. We did not measure a causal link, and the move after an announcement can be anything, including nothing.
Does the breakdown say whether to buy or sell?
No. The breakdown carries no side to trade, no score for the alert, and no probability. The median 24-hour move shown on a screener page is a measurement of past alerts, not a forecast.
How to weed out false hits is in the article on filtering out false hits; how to combine several conditions into one is in the piece on combining screener conditions; and what the metrics themselves mean is covered in the articles on open interest and the funding rate.
Other Knowledge Base Articles
Basic theory
About Our Screeners
Screener's settings