Chelsea vs Aston Villa Post-Match Thread

[Sexual Ealing]

Expected goals. It's a stat that the kids are into.

[Smithy]

Not being funny, and I hope not stupid, but what's Xg?

It stands for Expected Goals.  It's a stat used to track the "quality" of the chances you create in a game. "Shot on target" will obviously tell you something about how a team has played, but if they're all 30-yard speculative efforts it can be misleading about the quality of the performance.  Xg tracks where the shot was taken, whether it was with the foot or head, how the ball arrived etc and so on, and applies a "likelihood of a goal" to it (i.e. how often does a shot from this position actually result in a goal).  A chance with an Xg of 0.35 would be scored 35% of the time.  So a pull back to a striker on the six yard line with an open goal is probably something like 0.9 - whereas a header from the edge of the box from a corner clearance is probably 0.05.  Both would technically be a "shot on target".

The fact we both had 1.37 expected goals means Chelsea did really well with their chances, and we did not.  That doesn't tell us anything we didn't already know, certainly not if you saw the game, except to add a little weight to the argument that we actually played OK, barring a couple of defensive lapses and our finishing (which are both obviously very important!).

To put a bit of context on it, when we beat them 2-1 at the end of last season our expected goals was 0.98, and theirs was 3.43.   We just did great with our chances, and they did poorly with theirs.

I like the Xg stat purely because it's useful track how creative the team has been.

Our expected goals for last season was 56.72, we actually scored 55 - so it was pretty accurate.  The better your attacking players are at finishing, the more likely your actual goals will be higher than your expected goals.  The team with the biggest difference between actual goals and expected goals last season was Spurs, who had Kane up front.  Probably not a coincidence.

It's just another way of looking at a game after the fact, as sometimes it's easy to let emotions decide whether a performance was good or bad.

[Hookeysmith]

Thanks for the explanation but i do feel it is statistics for statistics sake. Anyone who sees a full game and has a basic understanding of it can see who is on top, who was lucky etc

I thought MOTD gave us very little credit

[cdbearsfan]

I don't think it is that useful for matches but imagine it could have a purpose for scouting players. If a player scores fifteen a season, but his xG is thirty, that means his teammates are creating loads of chances that he messes up. If he he scores fifteen a season but his xG is six or seven, that means he is capable of creating goals out of nothing and might be worth signing, I imagine.

[Smithy]

Thanks for the explanation but i do feel it is statistics for statistics sake. Anyone who sees a full game and has a basic understanding of it can see who is on top, who was lucky etc

I thought MOTD gave us very little credit

Like all statistics, the smaller the sample, the harder it is to draw any proper conclusions. That's why Xg is just an interesting aside on a game-by-game basis, nothing more.  Xg is often miles away from the actual result in an individual game, but over time, over a season, it becomes more accurate for a team's performance.  It also becomes useful to see how many "quality" chances you're creating over time, vs those you're converting.  As others have said, this is also useful for scouting players. 

As you rightly pointed out, anyone watching the Chelsea game will know it wasn't 3-0, we probably should have scored at least once, and they shouldn't have got 3.  But over the longer term, when we're 10 games in, it's harder to gauge how we've performed beyond 'gut feel' (which if this forum has taught us anything, it's that people rarely agree on stuff like that).

In a game that is increasingly influenced by statistical analysis, I think it's useful for us as fans to have a basic understanding of the metrics being used by the teams themselves.

[dcdavecollett]

If you watch the match at 88 mins, the bit where McGinn appears to be fouled but the ref (the egregious Atwell) awards Chelsea a goal-kick -one of the many decisions to go in the home side's favour that afternoon- the camera cuts away to the touchline where Deano is having a conversation with the fourth official.

Unless my GCSE(failed) lip-reading is wrong, Deano appears to be making a reference to propelling spheroids, a subject that we all know is close to his heart.

But he didn't get sent to the stand and he wasn't fined £25,000, either.

I presume from this that the FO was an adult.