Introducing a new feature on TCS: data-driven match analysis threads with visualisations by @MNazarian10! https://abs.twimg.com/emoji/v2/... draggable="false" alt="📊" title="Balkendiagramm" aria-label="Emoji: Balkendiagramm">

Starting with yesterday’s 4-1 win over Crystal Palace... https://abs.twimg.com/emoji/v2/... draggable="false" alt="⬇️" title="Pfeil nach unten" aria-label="Emoji: Pfeil nach unten">

This viz uses a metric called Possession Value to illustrate which players contributed most effectively with the ball.

We can see that Mason Mount had the highest PV of any Chelsea player, with Callum Hudson-Odoi and Ben Chilwell close behind.

Here, the metric Expected Threat is used to determine and display which players were the most dangerous offensively.

The right-hand-side partnership of Mount and Hudson-Odoi was particularly effective: two of Chelsea’s three open-play goals came from that flank.

Hudson-Odoi’s best offensive contributions in this game came in the form of progressive passing, but those high-risk, high-reward balls didn’t come at a cost: he boasted an impressively high pass completion rate of 87.8% despite his team-high tally of 9 progressive passes.

Mount’s assist in this game came from a set piece, and a large part of his xT came from his other dead ball deliveries, creating multiple big chances from corners in particular. He also progressed the ball more than any of Chelsea’s other attacking players. Well deserved MOTM.

Finally, on to Kai Havertz.

There has been much discussion around whether he has been a false 9 or just a conventional one for Chelsea, but his heat map shows that in this game he had a very free role, often dropping into midfield to receive the ball in central areas.

He took more shots than any other Chelsea player with 5, but the one goal he did score actually had the lowest Expected Goals value of them all at around 0.07, while Guaita saved 3 opportunities of his that would be classed as ‘big chances’ at over 0.3xG. A promising display.