You sent an order, you got a price back, and something about it looked wrong. Maybe it was a few cents worse than the number on your screen when you clicked. Maybe it was outside the high and low of the candle it landed in. The instinct to go and check is a good one, and the check is worth about five minutes.
Before the five minutes, one correction that will save you a wasted evening. Blockcircle's Market Trades page is not where the prints are. It says "live execution feed" near the top, which reads like time and sales, but what it carries is trade signals and positions posted by Blockcircle analysts. Posts with authors and timestamps, not a stream of executed trades with sizes. If you go there looking for the print stamped at the same second as your fill, you will not find it, and you will conclude something is broken when nothing is.
What the page carries, so you stop looking for prints on it
It is worth a look at the actual surface, because once you have seen it you will never make this mistake again.
At capture the controls were a Templates button, a search box that accepts a keyword, a ticker or a dollar-sign symbol, a subscribe action, and horizon tabs reading ALL, SHORT-TERM, MID-TERM and LONG-TERM. Below them, posts from analysts. The posts can be quite detailed. One close-long card on silver reported an exit price of 64.27 dollars, a result of plus 8.02 percent and a duration of 28 days and 4 hours, along with the originating strategy and its historical figures as shown that day.
Every price on that card is a historical reference for the idea being described. None of them is a print you can check your fill against, and none of them is a quote you can trade at now.
The five-minute check, in the order that matters
Do these in sequence. Skipping the first step is what causes most of the false alarms.
Start with your own trade confirmation, not the chart. Open the fill record at your broker and write down four things: the exact timestamp with seconds if the broker gives them, the price, the quantity, and the order type you actually sent. That last one catches a surprising share of these. A market order you thought was a limit explains almost every "my fill was outside the range" complaint by itself.
Next, open the time and sales window at the same broker or the venue you traded on. This is the real print tape and it is the only place the same-second comparison can happen. Most retail platforms have it, often behind a button labelled time and sales, tape, or trades. Find the seconds around your timestamp and read the prints.
Third, compare like with like. Your fill has to be checked against prints from the same venue in the same second. If you traded a crypto pair on one exchange and you are reading prints from an aggregated chart, you are comparing two different markets and any difference you find is meaningless.
Fourth, check the spread at that moment if your platform records it, or the bid and ask on the tape either side of your print. A fill outside the range of recent prints but inside the quoted spread is not a bad fill. It is the spread, which is the cost of demanding immediacy, and it is exactly what a market order buys.
Fifth, and only if the first four leave you with a genuine gap, write down the size of that gap in dollars. Not in percent, not in ticks. On 400 dollars of stock, three cents of slippage is twelve cents of money. Doing this conversion early is the single most useful habit in this whole exercise, because it tells you whether the thing you are annoyed about is worth another hour of your evening.
Why a fill lands outside the printed range, ranked by how common it is
Four ordinary explanations account for nearly all of these, and none of them involves anyone doing anything to you.
The most common is that you crossed the spread. Prints show trades. Your market order executed against the resting offer, and if the last several prints were on the bid your fill sits above all of them and looks like an outlier. It is not. It is the ask.
The second is venue fragmentation. In equities, and even more so in crypto, the same instrument trades in several places at once. Your broker routed somewhere. The chart you are reading may be consolidating several venues, or showing only one, and it may not be the one you traded. Same instrument, different market, different prints.
The third is timestamp convention. Your broker may stamp the fill when it was reported to you, while the venue stamps when it matched. A gap of a second or two is not a discrepancy, it is two clocks measuring different events. This is why you look at the seconds either side rather than demanding an exact match.
The fourth is candle aggregation, and it fools people constantly. A one minute candle shows the high and low of that minute on the data source your chart uses. If your fill is outside it, the first suspect is that your chart's source and your execution venue are not the same feed, not that your fill was outside the market.
The one pattern that is actually worth escalating
All of the above is noise. Here is the signal.
If, across a series of fills, the gap between the screen price when you clicked and the price you received is consistently in the same direction and consistently larger than the spread at that moment, that is a pattern rather than an incident. One bad fill is a story. Fifteen fills all worse in the same direction is a routing question, and it is a reasonable thing to put to your broker's support desk in writing, with the timestamps and the prints attached.
Be specific when you do it. "My fills are bad" gets a template reply. "On these six orders, at these timestamps, the venue tape shows prints at these prices within the same second and my fills were consistently worse by this much" gets a real answer, because it is checkable.
One thing not to do while investigating: nobody at a broker or an exchange needs your API keys, your password, or your account credentials to look into a fill. A support agent asking for any of those is not investigating your fill. Trade permissions and withdrawal permissions should stay separate on every key you hold, and a fill query never requires either to be handed over.
Turning one check into something that pays you back
Checking a single fill is worth doing once. The version that changes your trading is the log, and it takes about fifteen seconds per trade.
Five columns, filled in at the moment of the trade rather than reconstructed later. The instrument. The order type you sent. The price on screen when you decided. The price you got. The venue. Nothing else, because a wider log does not get filled in and a partial log built from the trades you happened to remember is biased towards the ones that annoyed you.
After about thirty trades the columns start talking. Average the gap between decision price and fill price for your market orders and multiply by your typical size, and you have the annual cost of demanding immediacy, in dollars, on your own trading. Then split it by instrument. There will almost certainly be one or two things you trade where the number is several times worse than the rest, and those are the ones where a limit order should become your default this week, regardless of what any general advice says about order types.