Initiative and Responsive Activity: Information Asymmetry at the Extremes

Abstract. Initiative (informed) traders initiate trades by placing orders; responsive (uninformed) traders accept those orders. This distinction shapes price dynamics and continuation patterns because the order flow itself reveals information. When aggressive trading concentrates at price extremes, highs, lows, or support/resistance levels, the source of that aggression (who is initiating) predicts whether the move will persist or reverse. This paper examines the classification of initiative versus responsive activity, the mechanics that govern who trades at extremes, and implications for assessing price continuation.

The Core Concept

Market microstructure research distinguishes initiative and responsive trading roles based on who submits an order first ^[1]. An initiative trader places a limit order and waits; a responsive trader places a market order and accepts that limit price immediately, removing liquidity from the book. This mechanical distinction carries an information implication: the trader willing to transact immediately (the responsive party) faces greater risk of adverse selection, so responsive traders are generally assumed to be less informed than the traders they trade against.

When prices reach extremes, local highs, lows, or technical support and resistance levels, order flow becomes asymmetric. If aggressive selling (responsive activity) concentrates at a local high, it may signal capitulation or distribution by informed sellers, suggesting the high is genuine. Conversely, if aggressive selling at a high comes from reactive liquidation (uninformed traders), the move may reverse as that forced selling subsides. This distinction between informed and uninformed aggression at extremes is relevant to continuation prediction, not because the pattern predicts reversals deterministically, but because the source of the aggression carries different information value.

How Initiative and Responsive Activity Are Distinguished

The primary method for classifying trade direction, and thus initiative, is the Lee-Ready algorithm and its variants ^[1]. When a trade occurs, the algorithm compares the trade price to the prevailing bid-ask spread. If the trade price is closer to the ask, the trade is classified as buyer-initiated; if closer to the bid, seller-initiated. When a trade is exactly at the midpoint, the algorithm uses tick rules: if the price is above the previous price, it is marked as a buy; if below, a sell. The trader who initiated the trade (the one who "hit" the bid or "lifted" the ask) is the aggressive side; the other is responsive.

Informally, traders who participate in price extremes fall into two broad categories. Informed initiative traders anticipate further movement in the direction they trade and aggressively buy or sell into resistance or support, signaling conviction. Reactive responsive traders are drawn into the extreme because stops are triggered, margin calls force liquidation, or algorithms react to price levels mechanically. These two sources of aggression produce different continuation profiles.

When aggregate order flow at extremes is heavily initiated (aggressive), it tends to predict continuation in the initiated direction, because informed traders are willing to pay or accept less favorable prices to execute. When order flow at extremes is heavily responsive (reactive), continuation is less reliable, because the reactive flow may be exhaustible, once stops clear or forced sellers exit, the move may stall. This pattern is conditional on information asymmetry and market state; the effect is not deterministic.

Worked Example: Initiative Demand at Support

Consider a stock that has declined steadily and is testing a major support level for the second time within weeks. At the support level, large passive buy orders accumulate on the bid. The stock price drops to within 2-3 cents of that level, and several outcomes are possible.

Scenario A: Initiative buying dominates. Large market buy orders execute against the support-level limit sells, pulling price upward. In high-frequency data (tick-by-tick or order-level), these buyer-initiated trades would be classified as aggressive buys. The traders who placed these market orders (the initiative buyers) are willing to pay near the worst levels at support, signaling conviction that support will hold and prices will recover. This is consistent with informed trading: the buyers know more than the passive limit sellers and are willing to pay for speed. In this scenario, the stock often reverses and rallies, because the initiative buying reflects genuine repricing demand.

Scenario B: Responsive selling dominates. Instead, the price tests support and trigger cascading selling from algorithmic stops and liquidation orders. These are responsive sells: they are reacting to the price level, not anticipating a further decline. The sell orders themselves do not originate from informed traders; they originate from mechanical processes and forced exits. If the passive buyers at support absorb this reactive selling quietly, and the reactive sellers complete their liquidation, then buying pressure reasserts and the stock rallies. If instead reactive selling accelerates (panic or algorithmic feedback loops), the support level breaks. In either case, the critical information comes from the source of the aggression: reactive sellers are less informative than deliberate initiative sellers.

This distinction is subtle but consequential. A trader or algorithm observing that the order flow is heavily responsive at support is in a weaker position to predict continuation than one observing that the flow is heavily initiative. Practitioners sometimes assess this by noting the speed, size, and persistence of aggressive orders, and by asking whether the aggression appears forced (responsive) or deliberate (initiative).

Why It Matters for Continuation

Initiative order flow is positively associated with price continuation in the initiated direction ^[1][2]. When aggressive traders buy into resistance or sell into support, they are taking positions at unfavorable prices, which implies confidence, or at least information, that the move will continue. Conversely, when order flow at extremes is reactive and exhaustible, the continuation of the move becomes dependent on whether the reactive flow clears and whether underlying demand or supply reasserts.

From a market efficiency perspective, initiative trading at extremes is more likely to move prices permanently, while responsive trading may move prices temporarily. This distinction is relevant to both short-term tactical decisions (whether to fade or follow the move) and to longer-horizon price discovery: if informed traders are aggressively trading at extremes, the extreme may represent a genuine repricing rather than an overshoot.

However, this relationship is probabilistic and conditional. It does not hold uniformly across market states, asset classes, or time horizons. During stress events or liquidity crises, the classification itself becomes noise; the forced seller is indistinguishable from the informed seller by price alone. The Lee-Ready algorithm makes assumptions about bid-ask spread stability and tick directionality that can break down in fast markets.

Limitations

The distinction between initiative and responsive activity, while conceptually sound, has several material constraints in practice.

First, the Lee-Ready algorithm (or any simple price-based classifier) is indirect. It infers intent from price and assumes that aggressive orders represent informed trading, but this is not always true. A large motivated buyer may hit the ask at resistance not because they are informed that the move will continue, but because they have a fixed demand (a hedge, a rebalancing requirement) and the price level is incidental. Conversely, a reactive algorithmic sell at support may contain genuine information about liquidity supply that the passive limit sellers lacked. Classification by price is faster than classification by intent, but it is also noisier.

Second, the identification of "extremes" is itself subject to data-mining and look-ahead bias. A support or resistance level that was obvious in hindsight may not have been obvious in real time, and the casual application of this framework can lead to circular reasoning: "prices tested support, and since the order flow was initiative, I knew it would hold" versus "prices held support, so the order flow must have been initiative." Research on this topic often mitigates this by using ex-ante defined levels or long historical samples, but practitioners must be cautious.

Third, the framework assumes that informed trading concentrates at extremes. But information is heterogeneous and asymmetric: the informed buyer at resistance may be informed that demand will arrive, not that the local move will continue. The distinction between level-specific information and directional information is often glossed over.

Fourth, empirical research on continuation following initiative versus responsive activity is sparse and mixed in quality. While the intuition is sound, demonstrating a solid, actionable edge requires controlling for confounds (volatility regime, leverage, sector rotation) and accounting for transaction costs. Many papers on order flow and price discovery measure continuation over microseconds or seconds in liquid markets; the result may not transfer to less liquid assets or longer horizons.

Summary

Initiative and responsive activity reflect a distinction in market microstructure between traders who aggressively demand immediacy and those who supply it. At price extremes, the composition of order flow, whether the aggression is initiative (informed, deliberate) or responsive (reactive, forced), carries information about the durability of the move. Initiative aggression at support or resistance is more likely associated with genuine repricing and continuation; responsive aggression is more likely temporary and reversible. However, the classification is indirect and based on price alone, "extremes" are difficult to identify in real time, and the empirical evidence on continuation edge remains conditional and subject to model assumptions. The framework is useful for qualitative assessment of market microstructure but should not be mechanized without careful validation on the specific market and horizon in question.

Key Definitions

Initiative order: An order that demands immediacy and liquidity, typically a market order that accepts the quoted bid or ask price; the trader who submits an initiative order is exposing themselves to adverse selection and is assumed to be more informed.

Responsive order: An order that supplies liquidity by accepting or taking the opposite side of an initiative trade; typically a limit order already on the book or a market order reacting to a price level or news event.

Lee-Ready algorithm: A price-based method for classifying buy-initiated versus sell-initiated trades, developed by Lee and Ready (1991); compares the trade price to the bid-ask midpoint and uses tick direction when price is exactly at the spread midpoint.

Adverse selection: The risk faced by liquidity providers that the trader on the other side of their quote is better informed, likely to profit at their expense, and thus the quote is mispriced; a primary cost of market-making.

Order flow: The sequence of buy and sell orders executed in a market, distinguished by size, direction, and aggression; in microstructure research, order flow is examined as a predictor of price discovery and continuation.

Price extremes: Local highs, lows, or technical support and resistance levels at which traders or algorithms concentrate order flow; these regions are examined because they often precede reversals or continuations depending on the nature of the order flow.

References

• Lee, C. M. C., & Ready, M. J. "Inferring Trade Direction from Intraday Data." Journal of Finance, vol. 46, no. 2, 1991, pp. 733-746. Https://doi.org/10.1111/j.1540-6261.1991.tb02683.x
• Hasbrouck, J. "Empirical Market Microstructure." Oxford University Press, 2007., A full reference on order flow classification and price discovery; the Lee-Ready method and its extensions are covered extensively.
• Easley, D., & O'Hara, M. "Information and the Cost of Capital." Journal of Finance, vol. 59, no. 4, 2004, pp. 1553-1583. Https://doi.org/10.1111/j.1540-6261.2004.00672.x, Examines the link between order flow, information asymmetry, and asset pricing; foundational for understanding why initiative trading predicts continuation.
• Chordia, T., & Subrahmanyam, A. "Order Imbalance and Individual Stock Returns." Journal of Financial Economics, vol. 72, no. 3, 2004, pp. 485-518. Https://doi.org/10.1016/j.jfineco.2003.10.003, Analyzes the predictive power of order imbalance on price continuation across equities.
• U.S. Securities and Exchange Commission. "Market Structure Literature Review, Part I: Equity Markets." SEC Division of Economic and Risk Analysis, 2014., Provides regulatory perspective on market microstructure and order classification in U.S. Equities.

Educational research on historical data only. Not investment advice, not a signal, and never a performance promise. Past results do not predict future performance. Every reference is link-verified before publication and every paper is re-audited weekly against the library's editorial standard.

Last reviewed by the PropLedger research pipeline: 2026-08-30. Educational research on historical data, not financial advice.