A Peek Behind the Quant Curtain

This evening I had a flight from Sydney to Melbourne and, as usual, it is one that is full of business people who are either tired and just want to go home; or, have an early morning meeting and need to be in town the night before.

As soon as the seatbelt light turned off, the person next to me pulled down the tray table and plonked a large sheaf of loose papers on it.

Normally, I don’t pay attention to people doing their work on plane flights (it’s usually pretty boring stuff), however what caught my eye was that it was a a chart that I was very, very familiar with, namely the S&P 500 daily candle chart.

It’s one of those moments where casually observing the world around you but not paying attention in any great detail turns into something very distinctive catching your eye and your attention. It was unlabled, no X and Y numbers, however I could clearly see the run-up in 2019, the dramatic March 2020 crash, the 2021 bull run, the 2022-2023 dip, and the monster 2024 run-up.

As someone who spends a lot of time on the stock markets, I had to say hi and introduce myself.

Turns out the person sitting next to me was a quant for a large Australian superannuation firm. They were spending the time counting the number of days between S&P 500 local tops and local bottoms. They thought there may be some alpha in that space.

We started having a conversation about the tooling they use and the pace at which they make decisions.

What surprised me the most?

Majority of their tooling were Jupiter notebooks with pandas. Results from the pandas notebook were exported via screenshot into PowerPoint slides to present to the investment committee once a quarter. Decisions made were emailed to the trading desk for execution.

I asked him if they ever reviewed their workflows to see if there were better ways of working? The answer was no.

We talked for a while about how they went about finding alpha, how to manage risk, and how to take a large team of people and manage execution.

It was fascinating because this is a company managing billions of dollars of money and yet their tooling and processes were so ad-hoc.

Retail trader takeaways

As a retail trader, here are three things I took away from that conversation:

1. Keep it simple

It should be simple to explain why your trading strategy makes money.

At the end of the day, trading is simply the auction process with a price, a volume, and a date and time. A buyer and a seller must have agreed for a fair exchange of value for the trade to occur. There is an expectation that the price was fair for both parties.

I have evaluated hundreds of trading strategies that use various indicators and try to do backflips with complex maths. Almost all of them fail to be profitable, especially when accounting for slippage, fees, and commissions.

2. Whales are slow

Large institutions move slowly and they have to go through committees and processes. They react to significant events months after the moment passes.

As a retail trader, all I have to do is catch the wave early and let the whales carry through. That means timing is not sub-second critical. Leave that to the HFT firms.

For retail trading (that is not the person’s day job), my measure if a trading system is fit for purpose is whether it can work on the two-three week time frame.

Over the years, I have found that to be a sweet spot for busy people who want to optimise their trading returns. While I can certainly trade more often and with shorter time frames, by ensuring the system can work on the two-three week timeframe, it means I am forced to ask and answer difficult questions about risk, volatility, position sizing, and drawdowns.

3. Paper is king

Something I have done in all of my work notetaking is to put pen to paper.

There is something about putting pen to paper that cannot be beaten. I’ve tried everything digital: note taking apps and tools, paper style covers for iPads, and many more devices; however, nothing beats being able to work on paper.

Printing out charts, annotating all over them is a part of the cognitive process that helps in ways that I cannot describe. The quant I met on the plane also agrees.

I wonder if it has something to do with the “serious play”¹ approach. I still don’t have a good reason for why that is, but it works and I’ll keep doing it as long as it keeps working.

Short version: Keep it simple, be nimble, and write things down.