Scalping is the fastest-horizon trading style: a trader takes many very short-duration positions — held for seconds to a few minutes — aiming to capture a small number of ticks or fractions of a percent per trade, then closing flat (typically within the session and almost always before the close). The defining bet is frequency over magnitude: dozens to hundreds of trades a day, each with a tiny target and a tightly capped loss, so survival depends on a high win rate and ruthless cost control rather than on any single big winner. That is also its core tension — because targets are so small, the bid-ask spread, commissions, and slippage are not friction at the margin, they are the dominant term in the P&L. The edge in scalping lives almost entirely in execution and cost minimization, not in the cleverness of the setup, which is what makes it structurally hard for an under-capitalized, latency-disadvantaged retail trader to win at.

What this section covers

This is a section-overview node. It frames scalping as a style and points to the sub-topics that carry the depth; it does not duplicate their mechanics. The three children:

• Level 2 & Tape Reading — the real-time order-flow toolset (the order book = intent; the Time & Sales tape = executed ground truth) scalpers use to read near-term supply/demand, absorption, and tempo. This is the highest-resolution signal available on the seconds-to-minute horizon, and the node honestly separates the documented microstructure effect (order-flow imbalance, Cont-Kukanov-Stoikov 2014) from folklore "manual tape reading as durable retail edge."
• Spread Capture — the market-making side of scalping: posting passive two-sided limit orders to earn the spread itself rather than betting on direction. The node centers on adverse selection — why naive two-sided quoting is negative-expectancy without genuine queue priority, speed, and exchange rebates.
• High-Probability Quick Trades — the directional working philosophy of scalping: the recurring setups (trend pullback to VWAP/EMA, range fade, tape-confirmed support/resistance, confirmed oscillator reversal) and the discipline (hard stop every trade, tiny per-trade risk, brutal cost accounting) that the style depends on.

Together these map the two opposite roles in the same order book — providing liquidity (spread capture) vs taking it on a directional tilt (quick trades) — both read through the same lens (Level 2 + tape).

Who and what it suits

Scalping suits a trader who can give the screen continuous, undivided attention (it cannot be done alongside a day job, unlike swing trading — Admiral Markets), tolerates rapid-fire decisions, has low per-trade transaction costs, and trades liquid, tight-spread, high-volume instruments — index futures (ES, NQ), major FX pairs, large-cap equities. Liquidity is the first filter, not a preference: the spread is paid on every round trip, so a wide or unstable spread is fatal to a one-to-three-tick target. The style favors flat, range-bound or cleanly trending intraday regimes where small directional tilts repeat; it does not need (and is largely indifferent to) multi-day fundamental or chart context.

When it works vs when it fails

It works for a skilled, well-capitalized, low-cost operator in deep, tight-spread markets during clear trending or cleanly ranging conditions — small repeatable directional tilts (quick trades) or steady two-sided fills (spread capture). Strengths: minimal per-trade exposure, fast feedback, no overnight gap risk, many opportunities.

It fails when spreads widen, volatility turns choppy/whipsaw (tight stops get picked off repeatedly), liquidity thins, or news injects a gap the tight stop cannot survive — and market makers widen spreads precisely into the volatility a scalper most wants, so the easy fills vanish exactly when needed. The single most common misuse is treating scalping as a low-risk income stream while ignoring cost drag: a 55-65% win rate paired with near-1:1 (or worse) reward-to-risk plus commissions and slippage is mathematically a slow bleed. Closely behind: abandoning the hard stop on a losing scalp, which destroys the entire asymmetry the style is built on.

Adoption, debate & evidence

Scalping is widely practiced and widely taught, and at the institutional level the underlying mechanics are uncontested — designated market makers and HFT firms earn the spread and exploit order-flow imbalance continuously and profitably, on the back of speed, queue priority, and rebates. The contested claim is retail viability, and the academic weight leans skeptical:

• Order-flow imbalance is a real but very short-horizon effect — Cont, Kukanov & Stoikov (J. Financial Econometrics, 2014) document predictive content on the order of seconds, decaying fast. Genuinely useful to scalpers and HFT; irrelevant to multi-day positions.
• Day-trading outcomes are poor in aggregate — Barber, Lee, Liu & Odean (Taiwan) find heavy day traders earn gross profits before costs but transaction costs erase them for nearly all; only ~1% reliably profit. A Brazilian (USP/FGV) index-futures study is widely cited for the figure that ~97% of those who traded >300 days lost money (the exact percentage varies by source/sample).
• The margin is razor-thin — backtested retail scalping profit factors are frequently reported in the ~1.07-1.24 range (community backtests cited by Quantified Strategies; these are illustrative, not a validated population statistic); even minor slippage eliminates an edge that thin. Vendor-cited win rates of 55-70% recur precisely because near-1:1 reward-to-risk demands roughly that just to break even after costs.

Folklore vs measured: the folklore is "small risk, high win rate, easy income"; the measured reality is that the setup carries little durable edge, whatever edge exists comes from execution speed and cost minimization, and population data show only a tiny minority capture it. (Specific thresholds repeated by educators — e.g. 0.1% stop, 0.2-0.3% target — are conventions, not validated parameters.)

Sources

• Investopedia — Scalping (definition, seconds-to-minutes hold, many-small-wins principle); cross-referenced via Pepperstone and naga.com summaries.
• Admiral Markets — Scalping vs Day Trading vs Swing Trading (time horizons, trade frequency, time-commitment contrast).
• Cont, Kukanov & Stoikov — The Price Impact of Order Book Events, J. Financial Econometrics 12(1), 2014 (order-flow-imbalance effect and its short horizon).
• Barber, Lee, Liu & Odean — The Cross-Section of Speculator Skill (gross-vs-net day-trading profits, ~1% persistent winners); Barber & Odean — Do Individual Day Traders Make Money? (Taiwan).
• Quantified Strategies — scalping critique (structural disadvantage, costs/slippage erasing thin edges); profit-factor figures ~1.07-1.24 originate from Forex Factory community backtests cited therein (illustrative, not a population statistic).
• Chague, De-Losso & Giovannetti — Day Trading for a Living? (SSRN 3423101) — Brazilian mini-Ibovespa futures; 97% of those who day-traded >300 days lost money; no evidence of learning.
• Child nodes (this section): Level 2 & Tape Reading, Spread Capture, High-Probability Quick Trades — full mechanics, defaults, and per-claim sourcing.

Dispute flag: the institutional mechanics of scalping (earning the spread, exploiting OFI) are uncontested; the retail viability of scalping is genuinely contested, with academic evidence leaning negative after costs. The "97% lose" figure is widely repeated but varies by source/sample.