Our 'form' is incredible

I understand that bookies do use it when pricing games up, among a load of other factors. It’s not particularly good at predicting individual games as, like you say, football is far too unpredictable and a game of very small margins, but it works well for long term trends across a full season.

 

On any given game you're right you can't beat the bookies simply using xG. However a long term approach, without trying to (a) claw back losses or (b) bump up wins, I reckon you'd end up with a modest seasonal profit using this stat. @Wellsie will do the analysis to prove me wrong by the end of next week

 

I don't have analysis on that, but I am a believer that you can beat the bookies long term if you're clever and selective enough (I'm neither). I did matched betting for a while and have had my money's worth out of the bookies, trading or professional betting is too risky for me.

 

I’ve always been of the opinion that ‘the bookies will always win, ergo gambling is a fools game.

But then I went to work amongst a load of analysts, really clever people - most of whom had a gambling habit.

Then it struck me that it’s much more complicated than I’d previously believed and that bookies aren’t working with probability alone, they have to temper that with betting trends, which means sometimes odds are illogical and the nerds can make a profit.

 

XG definitely has merit.

I started following it in the last league 1 promotion season. From about a third of the season in it correctly predicted that we would eventually rise to the top based on our XG for and against - and that Peterborough were in a false position and would fall away.

Also handy for Fantasy Prem striker choices

 

Remember ! "There are three kinds of lies: lies, damned lies, and statistics."; this was popularised by Mark Twain