Michael Reeves
There is much written by football statisticians these days but personally I remain a traditionalist that the most important stat is your league position and that is determined by the only other number to really count and that is your points on the board.
The Swans travel to West Brom today with 12 points from 10 games and 16th place in the league. That’s how the table shows it to be and no statistic will tell me any different that – after 10 games – it is a true reflection of the season to date.
Possession of the ball seems to be one that many fawn about these days but how many times in recent seasons have we seen a domination of the ball during games but at the end of the 90 minutes the scoreline shows a 2-0 defeat.
Xg is another one of those statistics that seems to get much positive press with even the TV highlight shows using it as a gauge of the game (or performance) but it’s never one that I pay much attention to.
That maybe ironic given I am now writing a rare piece on here about the subject but I think that is as much to prove the statistic is not important more than anything else.
xG and xGA are used to assign a value to goal scoring chances based on past performance data. Comparing these values to actual goals scored and conceded can indicate whether a team is above or below average at taking or preventing goal scoring opportunities; or experiencing particularly good or bad luck around the goal.
Bet Victor have recently done some comparison of the Championship teams so far this season with the key highlights they have observed being
- League leaders Sheffield United are overperforming in both attack and defence by 23.39% and 55.39% respectively
- Preston are the most underperforming team offensively (-112.54%), and most overperforming side defensively (90.64%)
- Blackburn are top of the league when it comes to goals scored above expected with 3.45
- Middlesbrough are the most underperforming team defensively having conceded 13 goals from an xGA of 7.67
- Bristol City have the highest xG of any team with 14.8, but are underperforming in defence having conceded 2.07 more goals than expected
I keep re-reading the statements they made above when they finished their research and drew the conclusion that if you score more goals than you concede, you win more games than you lose and therefore you have more points than many teams. And the more teams you have more points than the higher you are up the league table. Surely it is no more complicated than that one?
A glance at the league table will tell you right now what you need to know without the comments above or the tables below surely? From our perspective the tables produced in the research show that we have scored 2 less goals than Xg says we should and we have conceded 2 more than Xga suggest we should have done. That feels about right based on the 10 games we have played. Wait, am I telling myself that Xg and Xga are actually things I should be thinking about? I’m confused now.
“Sheffield United are overperforming in both attack and defence” the opening line above tells us. But it also tells us they are league leaders. Surely we don’t need a separate statistic to tell us they are good at attacking and defending. They have more points than the rest of us so surely they have to be? Don’t they?
It’s a statistic that will not go away in a hurry but I still cannot make my mind up if it is one that I am paying attention to or not. Points, wins, goals matter more than anything else surely? Or is this telling us that our play really should see us in a higher league position ? If we scored more and conceded less…
The tables are below, judge for yourself and why not comment below and tell us your own thoughts on the stats themselves. Are they as pointless as I think? Or do I really think that? I give up!
