Using Trend Following to Approximate Market-Making PnL

Summary

The document raises a question about a reported historical modeling result: a trailing trend-following strategy was said to reproduce the quarterly profit and loss of gilt market makers to within roughly half a percent. It asks how this relationship could arise, given the intuition that trend following may benefit from price movement while market makers face inventory and adverse-selection risks. The author also questions whether a simple negative correlation could explain the approximation.

The text offers no derivation, data, or details about the simulation, so it does not establish that trend following generally models market-making returns. The reported result concerns a particular historical setting and market structure, when market makers had lost protected access. It motivates investigating the mechanics linking inventory management, quote changes, and directional exposure, but the quoted account alone cannot show what model was used or whether the fit generalizes across instruments, periods, or market-making regimes.

Key ideas

• A historical account describes modeling gilt market-maker returns with a trailing trend-following simulation.
• The reported fit is specific to a market and historical context and is not independently substantiated here.
• The document asks whether directional exposure or some other mechanism links the two strategies’ returns.
• A correlation-based explanation alone would not identify the process producing market-making profit and loss.

Tags

# Approximating Market Making PnL with a Trend Following Strategy


# Approximating Market Making PnL with a Trend Following Strategy












In an interview about the setting up AHL Michael Adams made the following quote (the quote relates to their pre AHL days when they acted as consultants):

> I think because we we re doing work for Gilt market makers who were really struggling with the fact that they'd lost their franchises and no longer had a protected access to that market. We did some work for them and discovered that we could model their P&L to within +/ 1/2% p er quarter by simulating it using a trend following trailing strategy. I always saw that as a huge insight because it told us that trend following is really a systematic way of thinking about market making.

I would have thought that market making and trend following would be negatively correlated, trend following benefits from price volatility while market making does not. Is this correct?

However using a negative correlation like this (if it exists) does not appear a very natural way to approximate the PnL generated from a market making strategy.

So my questions are is my interpretation of how the PnL from the market making strategy is inferred from the tend following strategy correct?

If not how how did they appreciate the market making PnL from the trend following PnL?

Thanks

Baz

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.