The candle that closed your position was mostly wick, and forty minutes later the channel posted the win. The mechanic is ordinary rather than personal, and the useful half is that your stop and the channel's stop are often not the same order at all.

The candle is one minute long and mostly wick. It reaches a little past the level the post gave you, your position closes at a loss, and the wick is gone before you have finished reading the notification. Forty minutes later price is at the second target and the channel posts the screenshot with the green number on it.

The instinct is that something was done to you. The more useful reading is duller and more actionable: your order was sitting in a place that thousands of other orders were also sitting in, and a market that reaches into that place is behaving normally. What follows is the mechanic, and then the half of it that is yours to change. None of this is a claim that a better-placed stop makes money. It is a claim that being closed by structure you could have seen in advance is the avoidable part of the loss.

Why does everyone's stop sit in the same place?

Because the levels people choose are not spread evenly. They collect at round numbers and just past the last obvious high or low.

The clearest evidence for that is not from crypto. The Federal Reserve Bank of New York published a staff report in March 2001 built from the order book of a large foreign exchange dealing bank, Currency Orders and Exchange-Rate Dynamics, covering 9,667 conditional orders placed between 1 September 1999 and 11 April 2000 in dollar-yen, euro-dollar and dollar-sterling. If requested execution rates were spread evenly, each two-digit ending would hold about 1 percent of orders. Instead, roughly 8.7 percent of orders sat at rates ending in 00, with weaker clusters at 50, at other endings in 0, and at endings in 5.

The second finding in that report is the one a signal follower feels. Stop orders do not sit on the round number. They sit just past it. In the same data, 7.4 percent of stop-loss buy orders were placed at rates ending between 90 and 99, while 14.4 percent, almost twice as many, were placed at rates ending between 01 and 10. Stop-loss sell orders mirror the pattern on the other side, clustering just below the round number. Translated into the position you are actually holding: the order that closes a long tends to sit a little under the round number, and the order that closes a short a little above it.

That is a currency market from two decades ago, not a crypto pair, and it should be read as evidence about where people place orders rather than as a statistic about your exchange. The placement habit is the part that travels. A published signal makes that habit stronger, not weaker, because everyone who read the post is working from one number instead of from their own chart.

Is a cluster of stops a plot against your account?

A companion report from the same series, Stop-Loss Orders and Price Cascades in Currency Markets, describes what happens when a cluster is reached. Stop orders generate positive feedback: a triggered stop on a long is a sell, the sell pushes price lower, the lower price reaches the next stop. That report finds trends are unusually rapid at levels where stop orders are documented to cluster, and that the reaction to stop orders is both larger and longer-lived than the reaction to take-profit orders.

That is the honest version of the thing people call a stop hunt. It does not require anyone to know your account exists. A pool of resting orders is simply the one place on the chart where size can be filled without chasing price, and it is visible to anyone who reads a chart the way you do. Price reaching for it is ordinary behaviour, and it happens to accounts of every size on the same candle.

What that leaves you is a narrower and more honest question. The market moving against you is not in your control. Standing in the most crowded spot on the chart, in the exact place a broadcast told a few thousand people to stand, is.

Why is the channel not stopped out when you are?

Because in many cases the channel's stop and your stop are not the same kind of object at all.

Your stop is a resting instruction. It executes when its condition is met, at three in the morning, while you are asleep, without asking anyone. The channel's stop is frequently a line in a message. If the person behind the account watches the trade and closes by hand, a forty-second spike does not close them: by the time a human looks at the screen, price is back above the level and there is nothing left to react to. The same discretion can work against them too, and a late exit can be far worse than the published level. But on the specific event you are angry about, the wick, discretion is a shield that your resting order does not have.

There are quieter versions of the same gap. A channel scoring its own record from a chart of candle closes will not count a wick that closed back above the level, while your order was triggered by the print rather than by the close. A channel reading another venue's chart is looking at a different low. In each case the recap can be accurate on the author's screen while your account is flat and smaller.

This is why a results post and a fill are different pieces of evidence. Unless the record shows executions and timestamps, comparing your outcome with the channel's is comparing an order with an opinion. The mechanics of replaying a call one candle at a time are in replaying one signal on a chart.

Which price is your stop actually watching?

Here is the part most followers have never opened, and it decides whether the wick even reached you.

A perpetual futures position has more than one price attached to it. The last traded price is what changed hands most recently on that venue. The mark price is an estimate of fair value, built from an index of spot prices across several exchanges plus the basis between spot and futures. Kraken's explainer puts the reason plainly: "Because a large trader could theoretically manipulate the price of a contract on one platform, the Last Price is not used to trigger liquidations. Instead, that is dictated by the Mark Price" (Kraken Learn). Binance describes the same split, with mark price composed from a spread of major spot venues, and warns that "the last price might deviate dramatically and significantly from the mark price during extreme price movements" (Binance).

Your stop watches one of them, and you chose which, possibly without noticing. Bybit's take profit and stop loss settings offer a reference price of Last Traded Price, Mark Price or Index Price (Bybit Help Center), and the same three appear in its API as the triggerBy values LastPrice, IndexPrice and MarkPrice (Bybit API documentation). Binance offers the choice of Last Price or Mark Price as the trigger type on a stop order (Binance). All of that was as documented in August 2026.

The consequence is direct. A spike printing on one venue's book can take out a stop watching the last traded price while the mark price, averaged across several spot markets, never travels that far. The trade-off runs in both directions and neither setting is a safe default: a stop watching the last price can also fail to fire before the mark price reaches your liquidation level, which is the sequence Kraken warns about, and which is why the distance to liquidation deserves its own arithmetic before the trade rather than after it. That arithmetic is in liquidation explained. This is a setting to know about, not a setting we are telling you where to put.

What did the stop actually pay?

Being triggered and being filled are two events, and the gap between them is where the number on your statement comes from.

On Bybit, a take profit or stop loss on the entire position is triggered at the reference price and then closes the position with a market order. On Binance, a stop-market order, once activated, "will automatically execute the order at the best available price" in the book. On a one-minute spike the book on the far side is thin, because the spike is precisely the event that ate it, so best available can sit some distance past the level you typed. The alternative, a stop with a limit attached, protects the price by accepting the possibility that nothing fills at all and the position stays open. Which of those failures you are exposed to is set on the order ticket, and the differences are laid out in which order type turns a published entry into a position.

What is inside your control here?

Four things, and not one of them is a promise about the outcome of any trade.

The level itself. A number published to a channel is a public number, and copying it exactly puts your order in the pool that the post created. The research above says the crowd already leans just past the obvious round figure before any channel gets involved.

The trigger source. Last price, mark price or index price is a choice on the ticket, and it decides whether one venue's wick can reach you at all.

The order type. Market execution takes the fill it can get; a limit attached takes the price or nothing.

The size. This is the only lever that decides whether one wick is a bad morning or the end of the account, and it stays yours whatever level the channel published. It matters just as much when there is no published level to argue about, and what a missing stop hands to the exchange is covered in signals without a stop-loss.

What the win post proves, and what it does not

Both things in the story can be true. The target really did print. Your position really was closed before it got there. The screenshot is not evidence of a lie, and it is also not evidence that the trade was survivable as published, because what it shows is the trade the author held, not the trade the format handed you.

The question worth asking a channel is therefore narrow and answerable: at what price, at what time, and against which price source was the stop considered hit. A channel that closes by hand can say so. A channel that keeps fills can show them. Anything vaguer than that is a story about a candle, and you have the same candle.

The practical read

Stops gather in obvious places, and a market that reaches into an obvious place is doing something ordinary rather than something personal. That part is not yours to fix. What is yours is whether your order sits in the most crowded spot on the chart, which price feed is allowed to trigger it, which order type executes it, and how large the position was when it did.

None of that turns a losing call into a winning one, and none of it says anything about whether a given channel is worth following. That is a separate question with its own evidence, and the current ratings for the channels we can score are at Signal Providers.

Sources