Markouts: Measuring Adverse Selection | Udit Samani
How to measure whether your market making fills are profitable or toxic — deriving the markout profile, the breakeven spread, and the connection to Avellaneda-Stoikov.
In Avellaneda-Stoikov , the optimal spread decomposed into two terms: an inventory risk premium and an adverse selection component. The adverse selection term — 2 γ ln ( 1 + γ / κ ) \frac{2}{\gamma}\ln(1 + \gamma/\kappa) γ 2 ln ( 1 + γ / κ ) — is a constant that compensates the market maker for trading against informed counterparties. It appears in the formula almost by magic. But where does it come from, and how do you measure it in practice? The answer is the markout . A markout is the realized PnL on a trade measured at some horizon τ \tau τ after execution. It is the simplest diagnosti
Explore this link on the map →related reading
- From PIN to VPIN: An introduction to order flow toxicity | The Spanish Review of Financial Economicselsevier.es
- Trillions of Markets — Grant Stengergrantstenger.com
- Order Flow Toxicity of the Bitcoin April Crashjheusser.github.io
- The Market for Lemons - Wikipediaen.wikipedia.org
- Toward a Broader Conception of Adverse Selectionbayesshammai.substack.com
- Toward a Broader Conception of Adverse Selectionbayesshammai.substack.com
- Prediction Markets Explained — LessWronglesswrong.com
- Designing Compute Markets | Kavish Gargkavishgarg.com
- Adverse selection - Wikipediaen.wikipedia.org
- Opportunity Marketsparadigm.xyz
- Building a Futures Market Making Bot - Winning the UChicago Trading Competition, Part 1 - Tianyi (Austin) Liu's Blogtianyi.io
- [1904.05234] Flash Boys 2.0: Frontrunning, Transaction Reordering, and Consensus Instability in Decentralized Exchangesarxiv.org