- Thread starter
- #41
Country: Iceland
@Macho and you feel we barely got our attack out of first gear yet this season.
-
! ! ! IMPORTANT MESSAGE ! ! !
Discussions about police investigations
In light of recent developments about a player from Premier League being arrested and until there is an official announcement, ALL users should refrain from discussing or speculating about situations around personal off-pitch matters related to any Arsenal player. This is to protect you and the forum.
Users who disregard this reminder will be issued warnings and their posts will get deleted from public.
Squares, circles and triangles : The 2023/24 statistic thread
- Thread starter hydrofluoric acid
- Start date
- Replies 403
- Views 21,336
Country: Iceland
@Macho and you feel we barely got our attack out of first gear yet this season.
Country: England
Lol I remember when the data used to be a source of concern. Nowadays they just constantly look elite.
Room to grow like you say as well.
Dokaka
AM's resident Hammer
Actually quite interesting to me that there's a correlation between being towards the top right of that graph and being considered good and/or exciting teams.
Many people just don't respect defensive solidity, which is extremely evident on here as well. Not difficult to understand why as it's impossible to both play risky expressive attacking football and maintain that solidity at the other end, and people generally want to be entertained.
You cannot have both. To be free flowing and expressive, you vacate important space that can be exploited if you turn the ball over, which makes you "naive" etc. If you keep your shape, you create less and become much more methodical, which makes you "boring". This only applies to top teams, just to be clear, as they constantly have to unlock a deep block. It's easier for teams like us to create chances without leaving space as we can operate on the direct counter because teams have to push up to score against our low block.
Problem is some people will never be happy. Wenger was chastised over and over again for being naive defensively, but the truth is he was just giving the fans exactly what they're screaming for right now - risky, free flowing attacking football.
Country: England
Well I’m well aware of what I like and what actually works in top level football may not be the same hence why I like data to supplement the viewing. To try and combat my biases.
I’m really glad @hydrofluoric acid made this thread so we can constantly monitor the patterns in the league. Arsenal look really healthy when looking from certain perspectives and away from the emotions of matchday.
I think the goal to eventually become good enough that this becomes an art, rather than just business should always be the goal though. There have been plenty of flamboyant and successful teams I don’t see why that should end or be scolded.
Country: England
If it really just became about off the ball runs and defensive actions I’d honestly stop watching and become a hooligan fulltime.
Dokaka
AM's resident Hammer
Yeah I've stated a few times that I'm not necessarily a fan of what a statistical driven approach has done to the game in terms of the viewing experience, but at the same time I'm not going to join the choir of troglodytes who disregard all stats as "nerd ****" etc. Clubs wouldn't use endless amounts of money and build entire strategies around the results of those statistical results if it wasn't incredibly valuable and important.
I think entertaining football with the Pep/Arteta approach only really kicks in when you either face poor teams that simply get blown away, or you amass a squad of such immense talent that moments of individual brilliance makes you stand up and applaud. We've seen the latter a few times at Arsenal, but we see it somewhat often at City because.. well, they're a decade+ ahead in spending and have been relentless in their pursuit of the best players.
Country: Canada
Can someone explain to me how xG is actually calculated? Is it only based on shots on target? E.g., if Martinelli curves one just beyond the bar (shot off target), is that disregarded in xG calcs?
Country: England
It's calculated how you'd think, distance, angle, obstacles. The likelihood of the ball going into the net being 1 and a header from distance and obscure angle being 0.12 or something (random example not to be quoated).
A way I got familiar with it is I looked at iconic or goals I remembered and saw what the xg was.
I know there is a huge difference between header and kicking as well. How the data is actually captured though I actually have no clue lol.
Country: Canada
Right but does a shot off target get disregarded? It has to doesn't it? A shot off target has no chance of being a goal so how can it be assigned an "xG"?
- Thread starter
- #50
Country: Iceland
I think there are some different formulas and databases that different calculators go by. I have seen some different xG from another source than xGphilosopy.
One of the interesting things with the strongest chess computers(programs) is that even if chess could be explained as a game of math, they evaluate positions differently. Even positions 5 games into the opening. The reason Im saying this is that even if there are differences in formulas and maybe some might be wrong the data they are going by has become actually very good and it's a metric that models results and performance quite decently IMO.
Dokaka
AM's resident Hammer
The xG is mainly calculated based off where/when/how the shot was taken, not where it went. That's why a penalty kick has a fixed xG of 0.76, because the average percentage of PK conversions is 76%. In simple terms you just look at a chance and assign a number on how likely it was to become a goal based off the positioning of the player, opposition, teammates, goalkeeper, type of shot etc.
It's an attempt to remove the inherent noise in football, like teams being high up the table based off 3-4 incredibly lucky results or the direct opposite, down the table due to being on the receiving end of it.
It's pretty reliable, but somewhat useless on a single game basis. Gives a much better idea of how well a team has played across a season than just looking at the league table.
Country: South Africa
Player:Nketiah
For context and clarity (for me at least). It shows how we aren’t as vertical anymore, looking more at patient progression up the field.
It will be interesting to see how we adapt with a vertical, space creating player like TP5 returning, giving us our 22/23 setup again.
He allows us more central access, and the question becomes do we use that and open ourselves up more to transition, or do we continue in the same vein so far this season.
Country: Canada
Perfect, thank you! I understand the concept of xG I was just unclear on what parameters are used to calculate it specifically with regards to off target shots.
Synical
Miss Superstitious
Country: England
I'm surprised at Chelsea's placement
Synical
Miss Superstitious
Country: England
LOL at us literally on City's back
- Thread starter
- #56
Country: Iceland
Synical
Miss Superstitious
Country: England
Strange, because both we and City have 11 goals from the last five matches. Or am I missing something?
- Thread starter
- #58
Country: Iceland
I have made xG evolution graph for the first 8 games in PL this season.
MartiSaka
Join my "Occupy A-M" movement here 🗳
Im most suprised at Everton, sitting next to league leaders Sp**s. Excellent in attack supposedly, Everton for top 4.
- Thread starter
- #60
Country: Iceland
