XTX Trading Strategy Explained: How the Quant Giant Makes Billions
Inside XTX Markets’ machine learning models, order book signals, and high-speed trading edge.
How would you like a £600m tax bill?
Well, that’s exactly what one 45-year-old trader had in 2023.
(Last year he managed to trim it to a mere £202m 
)
The guy lives in London, runs a trading firm called XTX, and is one of the wealthiest people in the UK.
So… what does he actually trade?
XTX Markets
XTX is a quantitative trading firm, like Citadel or Jane Street, making tiny slivers of profit across millions of trades.
On their site they put it like this:
“We are a leading algorithmic trading firm that uses state-of-the-art machine learning technology to produce price forecasts for over 50,000 financial instruments across equities, fixed income, currencies, commodities, and crypto.”
Source – XTXMarkets
And being the market nerd I am, I wanted to know:
What does an actual trade look like?
(Because, spoiler alert: it ain’t no bull flag on the 1-min Nasdaq 
)
Turns out their signals are probably a lot about the order book:
Imbalances — bids vs offers at each level
Quote dynamics — how often prices update
Liquidity shifts — size pulled just before a move
Sweep patterns — aggressive trades hitting multiple levels (aka momentum ignition I guess)
Then add tick-by-tick patterns like:
➡ Micro-momentum (eg. 5 buy sweeps in a row)
➡ Reversion signatures (“last sweep was overdone, odds of bounce = x%”)
➡ Intraday seasonality (NY open, close, liquidity pockets)
All of that goes into their 25,000 Nvidia AI chips, chewed up in a data centre, and spat out as trades… filled before you or I even open a ticket.
(Bet they made a killing over NVDA earnings yesterday!)
Anyway, it’s not how we trade but I love anything like this…
So, if you want to do some digging for yourself.
> The article that caught my eye was on the FT here.
> But that’s paywalled, so if you don’t get FT I found a similar article here.
> And of course XTX’s website here.