reasons for changes from the prior version:

1. more data
2. the 8th and 9th place teams in the West have changed

San Diego’s xGD trend against playoff opponents seem to be pretty close to their overall trend.

In contrast, Inter Miami’s stands out as one of the team’s whose numbers drop off the most against stronger opponents.

I have to do some estimations for extra time from match event logs, which leaves some leeway for error. This is why I favor the fbref's listed xGD/90 figure when showing the overall xGD per team.

Your math is probably not bad. It's supposed to be Leading xGD * Leading Duration % + Trailing xGD..., which does come out to -0.03. To "square" things with fbref, I end up using what they list as total xGD/90 on the right-hand side, which was +0.03 at time of scraping. (Looks like it's +0.01 now.)

lmk if this is along the lines of what you're looking for. happy to iterate

absolutely! i’ll post a follow up when i can

i believe the author’s name is Tyson Ni. (not sure if he is on bsky.)

i was made aware of these rankings from a @rwohan.bsky.social article

it was offseason for everyone, and now i'm rusty with my viz 😅

i feel like this points to a selection bias that pre-shot xG may not capture. players who are good with their weak foot are more likely to take more shots with it and balance out goals with misses, and those who are poor with their weak foot will only take shots with it in very favorable positions

🔗 tonyelhabr.rbind.io/posts/soccer...

xGD is live from Opta here (including the ongoing matches)