I think XG gets a lot of stick on here because people try using it as a stick to beat the team with even though we're top of the league. personally I don't get why we need the XG stat but because of that I try not to comment on it and just leave it
Thoughts v Cardiff
- Thread starter kirsikka
- Start date
wingsnwheels
Star Player
Football is not the only sport to get stupid, unnecessary stats. Watching cricket the other day on Freesports they came up with shot efficiency (%) and watts created by the shot. Is the shot power fed back into the national grid by any chance? IMO stupid facts that mean nothing.
DJ
Moderator
Man of the match poll - https://bournemouth.vitalfootball.co.uk/vital-verdict-afcb-v-cardiff-city/
DJ
Moderator
Nothing is worse than the "expected result" stat, which the Cricket uses and Amazon Prime began to use for their Premier League coverage. Completely pointless.
Matt Stevenson
First Team
Hi all,
Here's the much-loved
xG timeline.
At 44 mins you could say the Cardiff game plan of sitting back and hoping for a smash and grab had been successful. They had created two big chances (both rated about 1 in 2) but nothing else, and we had created little in terms of clear chances. Their first was a header in the 6-yard box from Moore that Travers reacted to well, although Anthony did enough to stop Moore getting a firm header on it, which hasn't been commented on. The second was a slide by Moore to connect on the edge of the six-yard box and Travers kept it out.
In between Cook had a chance to shoot first time but opted to steady himself with what Kris Temple called a terrible first touch and the chance disappeared. And we also scored, although the chance rated one in 33 was covered by the keeper before the deflection. Let's hope this bit of luck sets Christie on his way.
The sending off in the 45th minute (which was a clear red, but I was still surprised that the ref didn't bottle it) changed everything, with Cardiff sticking to the game plan but unable to get near our goal. As they tired we dominated the last third of the game with five big (rated above 3 in 10) chances. In order.
A Lerma header where he rose well but couldn't keep it down, Solanke's goal after a great one-two involving a backheel from Billing, a Solanke chance from a Christie pullback although the ball was slightly behind him, the third goal which is a shame won't be credited to Jeff after a thunderous shot and a late chance for Christie when he couldn't sort out his feet in time to meet an excellent Marcondes cross.
Cardiff looked like they had given up at 2-0, but it was pleasing that we were still going after them rather than settling for that scoreline.
It looks like our confidence is back and have a nice set of fixtures in January.
Here's the much-loved
xG timeline.At 44 mins you could say the Cardiff game plan of sitting back and hoping for a smash and grab had been successful. They had created two big chances (both rated about 1 in 2) but nothing else, and we had created little in terms of clear chances. Their first was a header in the 6-yard box from Moore that Travers reacted to well, although Anthony did enough to stop Moore getting a firm header on it, which hasn't been commented on. The second was a slide by Moore to connect on the edge of the six-yard box and Travers kept it out.
In between Cook had a chance to shoot first time but opted to steady himself with what Kris Temple called a terrible first touch and the chance disappeared. And we also scored, although the chance rated one in 33 was covered by the keeper before the deflection. Let's hope this bit of luck sets Christie on his way.
The sending off in the 45th minute (which was a clear red, but I was still surprised that the ref didn't bottle it) changed everything, with Cardiff sticking to the game plan but unable to get near our goal. As they tired we dominated the last third of the game with five big (rated above 3 in 10) chances. In order.
A Lerma header where he rose well but couldn't keep it down, Solanke's goal after a great one-two involving a backheel from Billing, a Solanke chance from a Christie pullback although the ball was slightly behind him, the third goal which is a shame won't be credited to Jeff after a thunderous shot and a late chance for Christie when he couldn't sort out his feet in time to meet an excellent Marcondes cross.
Cardiff looked like they had given up at 2-0, but it was pleasing that we were still going after them rather than settling for that scoreline.
It looks like our confidence is back and have a nice set of fixtures in January.
New Forest Cherry
Star Player
do you know how the XG stats are compiled? does the algorithm include relative ability of the player (left foot, right foot, head), speed and bounce of the ball, position of the keeper, covering defenders etc?
when available, I'm looking forward to comparing Anthony's 54% chance at QPR with Moore's two chances last night.....
thanks
Neil Dawson
UTC Legend
I love stats and get bantered at work for forever pulling out spreadsheets. My lack of love for Xg isn’t that I don’t like stats it’s that it’s not statistically enough!
I don’t think you can have a goal prediction stat that doesn’t take into account the quality of the potential scorer or keeper. If you could factor that in I’d believe it. Harry Kane fifteen yards out against Darryl Flahaven would get the same Xg as Chukki Eribenne against David de Gea.
That will never wash for me….
RobTrent
UTC Legend
Confession time: Although watching when that challenge on Billing went in, I completely didn't see how bad it was. I was really surprised when a scrap ensued and even more when the ref brandished his red card. My eyesight is failing!
Glad to see waz is back with his trolling negativity.
Glad to see waz is back with his trolling negativity.
to be fair at the moment pretty much anyone fancied there chance against de gea so that should always be a higher xg
Matt Stevenson
First Team
Here is probably your best quick introduction - for most of your questions the answers it tries to, and some are probably too difficult to include (we don't want to make it even more complicated!)
What is Expected Goals (xG)? An Introduction | Education | Infogol
From putting together the timelines, my instinct is that the level of uncertainty is about +/- 5% and it gets worst with headers. So, take all of the stats with a pinch (or two) of salt.
To me, it's an interesting way of seeing how the game panned out - would I be surprised it a team with an xG 1 lower than another won the game, not at all. For me, it sense checks my natural AFCB bias.
Re Anthony vs Moore *2, I'm not sure which chance is easiest - Anthony's would have been if he had controlled it first, but he didn't. With the margin of error we are working too, all are about the same. Did you have a clear view on what was easiest?
Matt Stevenson
First Team
I understand this logic, but equally you should say the same thing about all the other stats, such as big chances and penalties, because of the different level of forward / keeper, shouldn't you? If you throw them all in the bin, then that is consistent. If you're happy to discuss the others, then xG is just another to add into the mix.
We all have our own internal xGs, based on an eye test. New Forest Cherry, has raised which chance of Anthony's vs QPR, and Moore's two last night is the easier. I think if we asked about we would get different answers. xG says Anthony would score about 1 goal more in 20 such chances, than Moore would have done, but this is well within the margin of error that xG has. I guess I find it strange that people are happy to have internal xG (with limitations) but not externally researched xGs (with limitations)
But I'm using xG purely as an added bit of context to the game, and for fun discussion, certainly not trying to make points one way or another. (or start arguments!)
I said we were poor in the first half but played some nice football in the second. I have a different opinion to you, get over it. Ironically you are the troll in this situation.
Neil Dawson
UTC Legend
Yeah, more than happy to have anything to have a bit of fun with, it’s the people that use it to make a point (when that point suits their narrative of course) that I take umbrage with.
For me to see it as anything but fun though it would need to take into consideration the talent of the players involved in finishing. When the pools panel adjudicate on say Liverpool v Norwich, or the bookies give odds they factor in the quality of the player. Similarly first goal scorer odds. In this day and age it wouldn’t be too hard to write an algorithm that factors in finishing ability so we could distinguish between a chance that fell to Stanislas and Lewis Cook. This would give it more credence .
Neil Dawson
UTC Legend
You were not alone Rob. I said ‘well if Lerma was sent off for that I’d have been fuming’ then watched it back on my phone at half time and it looked like a different tackle! Two reds!
Maybe we shouldn’t be so quick to judge referees.
Neil Dawson
UTC Legend
Sorry meant to reply to this post
RobTrent
UTC Legend
ok we'll all stop posting because we have different opinions!
as soon as the challenge went in I thought it had to be a red, but then the handbags after I thought might make the ref just dish out a load of yellows instead. After he put up the red red from where I was sat when he point the player off I thought he gave another red so I instantly looked to see where lerma was.
You must be intelligent enough to realise that you can disagree with an opinion without labelling somebody a troll?
