Diagnosing Adverse Selection in Market Making with Markouts
Summary
The document explains adverse selection as the risk that a market maker’s limit orders are more likely to execute just before prices move against the maker. One practical diagnostic is to place or observe limit orders near the best bid and offer, record fills, and measure the subsequent midprice movement. A negative average markout indicates that the average fill was followed by an unfavorable price move and can serve as a measure of adverse selection.
The discussion connects this effect to information differences among participants, including access to order flow or research, as well as examples such as insider trading and market manipulation. It also describes a response in which a market maker uses directional signals to adjust or cancel quotes when the expected price move is unfavorable. Markouts quantify average post-fill price behavior, but the document does not specify a measurement horizon, control for market conditions, or explain how to separate informed trading from other causes of price movement.
Key ideas
- Adverse selection occurs when fills tend to precede unfavorable price moves for a market maker.
- Post-fill midprice markouts provide a practical way to measure this effect.
- Information asymmetry can give some participants an advantage over resting quotes.
- Market makers may adjust or cancel quotes when signals predict an adverse move.
- Markout results depend on the measurement horizon and can have causes beyond informed trading.
Tags
Full text
# Adverse selection and market makers # Adverse selection and market makers What are practical examples of adverse selection market makers have to deal with?? I’ve read books but couldn’t really understand the concepts… ## Answer by databento (score 1) https://quant.stackexchange.com/a/80186 One way to intuit this is in terms of future (markout) PnL: - Place a bunch of limit orders randomly at BBO. - Get filled on them. - Measure their average markout PnL in terms of future midprice move. - Observe that the PnL is negative, i.e. price moves against you after the average fill. This move against you is one way you can quantify the adverse selection on a market maker (limit orders). ## Answer by HTF (score 0) https://quant.stackexchange.com/a/80177 adverse selection is something very easy to understand. A market maker should quote both on buy and sell side. When a bid order is executed, market maker wants the price to go up which makes the ask order executed as well, then he can make the money of spread. But once the price goes down, he's undertaking the float loss. To deal with it, market makers always build a model to predict the price movement, if the signal shows the price will go down, the bid order will be canceled to avoid loss. ## Answer by Evan Semet (score 0) https://quant.stackexchange.com/a/80185 Adverse selection is a result of information asymmetry in the market. Not every market participant has equal access to information/research capabilities. In other words, some firms/individuals know more than others. Your question more so pertained to actual examples of this, so I'll just list off a few: - Insider trading - Market manipulation tactics (banging the close, etc.) - Firms that have more direct access to flow information
Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.