Why Did My Stock Order Get a Bad Fill?

A stock order usually gets a bad fill because it was larger than the shares available at the price you saw. The first shares execute at the best quote, and the rest move to higher offers on a buy or lower bids on a sell, so the broker reports a weighted-average price that is worse than the quote on your screen. Quotes that change before the order arrives, orders ahead of yours, and liquidity your screen did not show can push the result either way. A disappointing average is therefore not, by itself, evidence of a broker error.

What a bad fill is, and what it does not prove

Before deciding whether anything went wrong, separate three numbers. The reference price is what you looked at before sending the order, usually the ask for a buy or the bid for a sell. The execution prices are the individual partial fills your broker reports. The average fill is the total dollar value of those executions divided by the shares filled, and it may be a price no individual trade actually printed.

The last trade is the wrong benchmark: it is historical and says nothing about how many shares were offered when your order arrived. A buy that fills above the initial ask is not automatically a mistake either. Fidelity's order-placement FAQ states that when a buy order is larger than the size available at the ask, part of it may execute above the ask, and that in fast markets orders ahead of yours can deplete the shares at the bid or ask before your trade reaches the market.

The reverse also happens. Internalization, price improvement, and non-displayed or reserve interest on exchanges can fill shares inside the spread or add size that never appeared on your screen, so an execution can beat the visible book.

Why a liquid stock can still be too thin for your order

Average daily volume describes activity accumulated over hours. It does not tell you how many shares you can buy right now within five cents of the ask, or how many bids will remain if you have to exit during a sharp drop. A stock can trade millions of shares a day and still show only a few hundred shares at the inside quote at a given second. Executable liquidity is also side-specific and size-specific: a deep bid does nothing for a buyer, and the same book that easily absorbs 100 shares may not absorb 5,000.

Spread and depth are separate measurements. A one-cent spread can sit on top of 100 shares at the ask and sparse offers above it, while a wider spread can carry thousands of shares on each side.

The depth screen has limits of its own. Check the unit first: Fidelity states that equity bid and ask sizes on its platform should be multiplied by 100, so a size of 5 means 500 shares. A "Level 2" window may cover only certain venues or a fixed number of levels, while US equity trading is spread across many exchanges and off-exchange venues. FINRA's explanation of the online trade lifecycle notes that the brokerage firm either executes the order itself or sends it to another venue, so your fills may come from several places rather than one ladder.

Odd lots add another wrinkle. Since April 27, 2026, the best odd-lot bid and offer from each exchange, plus a consolidated best odd-lot quote, have been part of required SIP core data. The UTP SIP odd-lot FAQ explains that these quotes are not protected under Regulation NMS and do not set the round-lot NBBO; full odd-lot depth has been deferred to 2028. A platform may therefore show small-size prices better than the NBBO that cannot absorb a meaningful order.

Session matters too. FINRA Rule 2265 requires brokers to warn that extended hours can bring lower liquidity, wider spreads, and unlinked markets quoting different prices, so a regular-session depth estimate does not carry over.

A 1,200-share example: how depth changed the trade

The following scenario is hypothetical. At 10:18 a.m. ET, ABC shows a bid of $31.98 for 800 shares, an ask of $32.00 for 150 shares, and a last trade of $31.99. A trader plans to buy 1,200 shares with a stop at $31.80 and a target at $32.40. Using the $32.00 ask, the plan looks clean: $0.20 of risk per share, $240 of total risk, $0.40 of reward per share, and a 2-to-1 reward-to-risk ratio.

The offers behind the ask are thinner than the plan assumes: 250 shares at $32.03, 400 at $32.08, and 700 at $32.15. An immediately executable buy with no price limit takes 150 shares at $32.00, 250 at $32.03, 400 at $32.08, and the last 400 at $32.15. The total cost is $38,499.50, and the average fill is about $32.0829. Depth alone added roughly $0.083 per share, or $99.50 across the order, to the entry.

The real damage is to the trade's arithmetic. Measured from the actual average, risk to the $31.80 stop becomes roughly $339.50, which is 41.5% more than planned, and the reward-to-risk ratio falls from 2-to-1 to about 1.1-to-1. The chart, the stop, and the target did not change. The executable trade did.

The exit can widen the gap again. If the position later has to be sold into bids of 300 shares at $31.80, 400 at $31.74, and 500 at $31.65, the average exit is $31.7175 and the loss is about $438.50, or 82.7% more than the $240 the trader thought was at risk. This is a capacity test, not a forecast of a stop order: entry liquidity alone does not validate a position size.

Sizing the order from depth before you send it

A depth screen is a scenario model, not a promise, but it can still reject an order that is obviously too large. Start from the trade plan rather than from the book. For a long position with target T, stop S, and a minimum acceptable reward-to-risk ratio R, the highest average entry that still works is (T + R × S) ÷ (1 + R). With a 1.8-to-1 minimum in the ABC example, that ceiling is ($32.40 + 1.8 × $31.80) ÷ 2.8, or about $32.0143.

Now walk up the offers until either the planned size is reached or the running average crosses that ceiling. At $32.00 there are 150 shares; adding shares at $32.03 raises the average quickly, and the ceiling is breached after roughly 286 shares. A 400-share purchase would already average about $32.0188. On the visible book, the setup supports a position of under 300 shares, not 1,200. A practical pre-trade routine fits into five checks:

1. Add up displayed shares at each acceptable price, in the correct unit, until the planned quantity is reached.
2. Estimate the weighted-average fill for the full quantity.
3. Recalculate dollar risk from that estimated average to the stop, not from the best quote.
4. Compare the estimate with the maximum entry the trade plan allows, and cut size until it passes.
5. Run a harsher version of the same test on the side you would use to exit.

A marketable limit order enforces the ceiling. In the example, a buy limit at $32.03 prevents any fill above that price, but only 400 displayed shares are available at or below it, so the remaining 800 could stay unfilled. The limit protects the price, not completion, and a partial fill is an inventory result rather than a problem to fix by chasing the rest. Splitting the order reduces the size exposed at any one moment, but it can also mean missing the move or paying a different average as the book refills. Skipping the trade is a legitimate outcome when depth cannot support even a reduced size or the book changes faster than it can be evaluated. At the open, the close, or a halt reopening, the continuous ladder does not describe how the auction will clear, and hidden liquidity may improve a fill, but a trade should pass its risk rules without depending on it. Professional desks read execution the same way: whether size was cut when the book thinned is one of the checks described in hi2morrow's guide to how a stock prop risk manager evaluates a trader.

How to reconstruct a bad fill and when to ask the broker

Start with the broker's order record, not a chart candle. Confirm the instructions actually in effect, including order type, limit price, session, and any modification. Then collect the submission time and every execution with its time, quantity, price, and venue, and recompute the average yourself. Benchmark a buy against the offer and a sell against the bid at the moment the order arrived, and reconstruct how many shares were accessible at better prices on the same session and feed. That separates two different causes: the order consuming several price levels, and the quote moving before the order arrived.

A fill is normally explainable when the executions reconcile to the reported average and the order was larger than the shares available at better prices. A review is warranted when the record shows the wrong instructions or quantity, an execution is missing, the average does not reconcile, a fill violates your limit, or the result stays materially out of line with the market at the time.

The regulatory standard is best execution, not the first displayed price, and it is separate from Rule 611, the trade-through rule that the SEC proposed in June 2026 to rescind; the proposal has not been adopted as of October 2026, and hi2morrow's guide to whether the SEC is repealing Rule 611 explains what the rule requires today. FINRA Rule 5310 requires firms to use reasonable diligence so that the resulting price is as favorable as possible under prevailing conditions, and its listed factors expressly include the character of the market and the size and type of the transaction. The broker should be able to explain how your order was handled, so ask in writing for the receipt time, each execution's time, price, size, and venue, and the order instructions in effect. FINRA's guidance on trade confirmations recommends checking confirmations when they arrive, raising discrepancies with the firm in writing, and filing a complaint if a genuine problem is not resolved.

There is also a newer public reference point. Under the amended SEC Rule 605, in force since August 1, 2026, brokers with 100,000 or more customer accounts, not only market centers, must publish monthly execution-quality reports, including a summary report. They will not explain a single order, but they let you see how your broker's executions compare with the quote across many orders of similar size. Where to find these reports and what their columns mean is explained in hi2morrow's guide to how to read a Rule 605 execution quality report.

Alexander Styopin's professional view: in practice, most bad fills are sizing decisions made before the order was sent, not execution failures after it. Daily volume tells you which stocks deserve a look; it cannot approve a specific order. The liquidity that matters is what you can reach, on entry and on exit, inside the price range that keeps the trade worth taking. If the planned size pushes the expected average outside that range, the position is too large even when the idea is right, so cut the size or cap the price before the market decides for you.

Educational material only. Not investment or legal advice. Order routing, market-data coverage, extended-hours access, and execution practices vary by broker, venue, and account type.

Author: Alexander Styopin, hi2morrow analyst and economist with 25 years of experience in the US stock market