Stat time

[RobTheRover]

I've never got the value of xG, Gaz.   Can you tell me how it's calculated please,  if you know?  Every match i check in Fotmob whilst it's in play rarely has an actual score that resembles the xG

[BillyStubbsTears]

In fairness, I suspect our xG yesterday was something like 0.5.

We created very few clear cut chances. Hurst's first goal was a bit of a freak, and his second was an excellent finish, rather than a stone cold unmissable chance.

That's fine as long as we keep taking chances. But I couldn't blame anyone who concludes that we've probably overperformed on results so far, compared to chances made and allowed.

But then we've had a bloody tough start to the season. Be interesting to see how the next two months pans out as the (on paper) difficulty of matches starts to level off.

[BillyStubbsTears]

Looks like I'm wrong again. Our xG yesterday was 1.4. Which is baffling compared to Sutton's xG of 0.52 against us. They created more clear cut goalscoring chances against us in the first 40 mins than we did all the match yesterday.

[IDM]

Sounds to me like that stat is one which is generated by the action in the game just happening/happened.  Therefore coaches could analyse why a team didn't score as many as they should have, poor finishing, better than expected goalkeeping?

However the circumstances and contexts of the following game are always different, which is why I argue that such stats are meaningless for predictions.

You don't need stats to tell you that if we're nicking results whilst playing badly, then its likely results will worsen if performances don't improve.  But the stats from those poorer games won't explain nor predict when  the performance will improve.