Tape reading: speed, size and aggressor side without mind-reading

Start in plain English

Tape reading is the skill of drawing evidence from the stream of completed trades, known as prints, that a Time & Sales window shows. For a small-cap scalper the tape is often the fastest information on screen: it updates before a 1-minute bar closes. The window itself, its columns and its data are covered in US Day Trading Fundamentals: Time & Sales. This lesson is about reading it.

The core idea: the tape shows what traded, not who traded or why. Three things on it are observations:

• Location: whether each print happened at the bid, at the ask, or between them.
• Speed: how many prints arrive per second, and whether that rate is rising.
• Size: how many shares each print carried.

Everything else, "buyers are stepping in", "a big seller is done", "someone is loading up", is an inference. Inferences can be useful, but only as hypotheses you count, not as facts you act on.

Two paths from the same start

All prints, quotes, sizes and costs in this example are Illustrative. Merrow Quay Pharma is fictional.

The tape at 10:31:20 ET. Merrow Quay was quoted $1.86 × $1.87 at 10:31:00. In the next 20 seconds the ask lifted to $1.88 and then $1.89. Here is every print in that window.

What a reader could count at 10:31:20: 12 of 14 prints at the ask; the print rate rising from 1 to 5 per five seconds; the ask lifting twice. And one print that does not fit: 20,000 shares at $1.87, labelled at the bid, larger than all the ask-side prints combined (7,400 shares).

The plan, written before 10:31. If an ask-side burst lifts the offer (this lesson's definition: at least 10 prints in 15 seconds, at least 75% of them at the ask), buy 900 shares at the $1.89 ask. Stop $1.84. Target $1.99. Time stop 10:36. Planned risk $0.05 per share, $45 (1R). Commissions and fees $0.005 per share round trip.

The review note: "Burst as defined; plan followed; outcome A, B or C." The tape at 10:31:20 described what had traded. It did not say which story came next. The rest of this lesson covers how platforms label print location, what speed and size can and cannot tell you, how quote changes differ from prints, and how to count your own tape observations.

Print location: at the bid, at the ask, or between

Most platforms colour each print by comparing its price with the quote at the moment the print arrives:

• At or above the ask: usually labelled a buy, on the assumption that a buyer crossed the spread.
• At or below the bid: usually labelled a sell.
• Between the bid and ask: often uncoloured, or split by which side it is closer to.

That label is the platform's inference, not a field in the trade report. Trade reports carry price, size, time and condition codes; they do not say which side initiated the trade. Researchers infer the side with rules like the one Lee and Ready (1991) proposed, comparing the price with the quote and with the previous trade. The inference can be wrong when the quote is stale, when a print is reported late, or when the trade was negotiated off-exchange. Platforms also differ in which quote they use and how they colour between-quote prints, so two screens can colour the same print differently.

So "12 of 14 prints at the ask" is an observation about labels. "Buyers are aggressive" is one step further.

Speed and clustering

The print rate is easy to observe and hard to interpret. In the Merrow Quay window it rose from 1 to 5 prints per five seconds. A rising rate shows that more trades are happening. It fits several situations: buyers crossing the spread, a seller filling many small bids, algorithms splitting one order into pieces, or traders reacting to an alert. Many traders see the same acceleration at the same moment, which may be one reason bursts can reverse quickly.

Clustering, meaning several prints at one price within a second, often comes from one large order filling against several resting orders, or from one resting order absorbing many small ones. Both look identical on the tape.

Large prints and why size alone is ambiguous

The 20,000-share print at $1.87 is the most persuasive line on the tape, and the least clear. It can be:

• a seller hitting a large resting bid;
• a buyer's large resting order being filled by many sellers at once;
• a negotiated or off-exchange trade reported at a price inside the recent range;
• a late report of a trade that happened seconds earlier, at a different quote.

Off-exchange trades are reported to a FINRA trade reporting facility. FINRA's reporting rules require most of them to be reported quickly, generally within seconds, and late or special trades carry condition codes. A platform may show these prints in the same list as exchange prints. Off-exchange prints on a small-cap tape are covered in the Advanced course, Dark pool prints.

The safe reading of a large print is the observed one: "20,000 shares traded at $1.87 at 10:31:15." The rest is a hypothesis to log.

Quote changes vs. prints

The tape shows trades. The quote shows the best displayed orders, and Level 2 shows more of them. They change for different reasons:

• The ask can lift because buyers took all the shares offered at $1.88 (a print), or because sellers cancelled their offers (no print).
• A bid can vanish because it traded (a print), or because it was cancelled (no print).

In Path B the bids at $1.88 and $1.87 disappeared before most of the selling printed. A tape-only reader would have seen the flush later than a reader watching the quote. FennEdge's article Level 2 vs. Time & Sales covers the difference between displayed orders and completed trades in detail; US Day Trading Fundamentals covers how to read Level 2.

Observe vs. infer

At 10:31:20 ET in the Merrow Quay example (fictional; all figures Illustrative):

Odds box

The split was chosen to land near a small positive number, so it reads as neither a promise nor a warning. Moving about four to five percentage points from Path A to Path B flips its sign. As of 2026-10-05, FennEdge does not offer a tick-level tape archive, so the sample for this question is your own log.

What shifts the odds

These are hypotheses to record, not findings:

• Share of prints at the ask, by count and by volume. In the example they point in opposite directions (12 of 14 by count; 7,400 vs. 20,300 shares by volume).
• Print rate in the 15 seconds before the decision, compared with the previous minute.
• Large prints in the window, their location and their condition codes.
• Quote behaviour: did the ask lift through trades, or did offers disappear without prints?
• Spread before and during the burst.
• Time of day and session. Premarket tapes are thinner.
• Distance to the high of day and to round numbers.

Cost of being wrong

Merrow Quay plan, 900 shares, 1R = $45 (Illustrative):

Tape-based entries happen in fast markets, and exits in fast markets can slip: in Path B the stop slipped two cents because the bid fell through two levels at once. Slippage covers how to measure that.

How many trades before this means anything

Tape observations are numerous but not independent. Fifty bursts logged in one stock on one morning share one news event and one crowd. Suppose you log 30 bursts across many days and 12 follow Path A (40%): the 95% Wilson interval for the true Path A share runs from about 25% to 58% (computed on these hypothetical counts). That range includes the break-even point and a clear edge alike. How many trades before your numbers mean anything explains why.

What this lesson does not claim

• It does not claim that any tape pattern predicts price. Every split above is Illustrative.
• Print-side labels are a platform's inference from the quote, not reported facts.
• FennEdge shows no Level 2 depth and no live tape (as of 2026-10-05). Practising tape reading needs a platform or simulator that provides Time & Sales.
• Merrow Quay Pharma and its prints are fictional.

Go deeper

• FennEdge first.
  • Article: Level 2 vs. Time & Sales: displayed orders are not completed trades.
  • US Day Trading Fundamentals: How to read Level 2 and Time & Sales.
  • The Scanner (paid: Plus, Pro, Max; as of 2026-10-05) alerts on rapid rises and drops, regular-session highs and LULD halts of Nasdaq-listed stocks: moments when tapes accelerate. It shows alerts, not a tape, and has no Level 2 depth.
  • In-Play Days shows 1-minute charts of past active days. Without Max, including as a guest, you can open each ticker's most recent active day in full; earlier days and the two-year history need Max. It does not show individual prints, but it shows what the minutes around a burst did.
  • Gapper HOD Timeline for when gappers printed their highs (guests, Free and Plus see the latest completed session; earlier sessions need Pro or Max).
• Absorption. Prints trading into a price that does not move: covered in the Advanced course, Absorption.
• Trade condition codes. Odd-lot, extended-hours and out-of-sequence modifiers change which prints update the last price, high and low. US Day Trading Fundamentals covers them.
• Why platforms colour prints differently. Different quote sources, timing and rules for between-quote prints.
• Related lessons: Thinking in probabilities, Is it liquid enough to scalp?, Paper trading.

Exercises

Seven exercises follow this lesson: three concept checks, two replays of the Merrow Quay tape (separate observations from inferences; then select every outcome consistent with the tape at 10:31:20 before Path B is revealed), one scenario about a large print, and one practice task. Exercises, answers and progress are available after signing in.

Practice task: 20 tape observations, fact and inference. On a simulator or a platform's replay of a past session, log 20 tape observations at the moments you would consider a trade, whether or not you take it. For each, write what printed, in observable terms only (location, print rate, largest print, quote changes), in tape_obs; your inference separately in review_note; and what price did over the next two minutes. Use the FennEdge practice log: download the practice log (CSV), with a column guide, with setup_tag = tape-obs and mode = replay or skipped. Keep the log separate from the Trading Journal, which holds only real broker-synced trades.

Primary sources

• Lee, C. M. C. and Ready, M. J. (1991). Inferring trade direction from intraday data. The Journal of Finance 46(2).
• FINRA. Trade Reporting Frequently Asked Questions.
• CTA Plan. Consolidated Tape Association.
• UTP Plan. Nasdaq UTP Plan.

Sources checked 2026-10-05. Merrow Quay Pharma and all prints, quotes, outcome splits and costs in this lesson are fictional and Illustrative. This lesson is education, not a recommendation to buy or sell any security. Past patterns do not indicate future results.