Why “Getting Picked Off” Has No Single Quantitative Measure
Summary
The document explains “getting picked off” as an informal market microstructure term rather than a condition with a universal empirical definition. In high frequency trading and market making, it often refers to a faster, better informed participant trading against a stale quote. The example describes a retail buy order met by a trader who sees a cheaper offer on another venue, captures the difference, and sells to the retail buyer at the displayed price.
This illustrates how fragmented markets and differences in data access or latency can disadvantage a slower participant. However, the example is only one possible mechanism, and the text gives no formula, dataset, or test for identifying when a particular trader has been picked off. It cautions that the term covers several market microstructure effects, so measuring it requires a more specific definition and context.
Key ideas
- “Getting picked off” has no universally accepted quantitative definition.
- In market making, the phrase often describes trading against a quote that is stale relative to broader market information.
- A faster participant may use prices on another venue to trade profitably against a slower participant’s displayed quote.
- The example illustrates one mechanism but does not establish a general measurement method.
Tags
Full text
# Definition of „getting picked off“ # Definition of „getting picked off“ Is there a precise definition of what getting picked off means. In particular I want to check whether I have been picked off. How can I measure this quantitatively? ## Answer by Theodore (score 4) https://quant.stackexchange.com/a/49040 > Is there a precise definition of what getting picked off means no, there exists no uniform metric defining what it means to get "picked off". You will hear this term often in the context of high frequency trading and market making, both of which are heavily latency-constrained trading strategies. All that really means is that their success is conditional on receiving and parsing market data very quickly, and determining what they want to do very quickly as well. The firms and institutions that engage in this kind of trading typically can (and will!) arb across multiple exchanges. What does this mean for, e.g., a retail investor? - retail investor has brokerage account and wants to buy X. say whatever brokerage service you have is displaying "real-time" quotes, and you see X's market (as far as you're concerned) which is \$100.20 @ \$100.35. - retail investor sends order for 50 shares of X. - now comes a HFT, and there are way too many differences to enumerate so we'll focus on the most relevant one: the HFT has a much better idea of the broader market's "opinion" on X; they're aggregating quotes from every single venue where X exists, including ones you don't even know exist. - now the HFT also sees your order; BUY 50 shares of X @ $100.33. say they see a sell order for 50 shares on a different exchange, for \$100.29. The HFT will: buy 50 shares for \$100.29 $\rightarrow$ sell 50 shares of X to retail investor for \$100.33 $\rightarrow$ profit is the difference between how much was paid between the exchange hops; $2. But you never even saw that quote! Given that getting "picked off" is not defined in any kind of empirical way, there does not exist any kind of certain way to know whether you were "picked off". I have only explained one (fairly trivial) scenario, however there are a multitude of other microstructure phenomenon to learn about that also play into this.
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.