I love the idea. Now, the problem of statistics is that they depend on population and samples. Even if expected goal is based historical data, I wonder if you have a league with better Goalkeepers, just to mention one variable, it wouldn't change the outcome then if that's true, I wonder if shoots on target wouldn't make more relevant metric to quantify. Although shoot on target doesn't give you the chances to score, it is relevant to the position you are shooting from.
I'm curious about an analysis of expected goals from Arjen Robben actually. I was pretty young when I watched him, but it always felt like he pulled in from the side and scored many goals outside the box on the 7% line that was mentioned
While the probability of scoring a goal from a shot taken on the edge of the 6-yard box might be lower than we think, many times, if the initial shot is blocked by a defender or saved by the goalkeeper, the rebound is tapped in. Is there a metric that would take into account the value of a shot from that position, even if the initial shot itself doesn't go in?
good point..... however the higher probability shots also have the added benefits of resulting in corners, rebounds etc and probably at a higher rate than the lower probability shots do
Could it be that Ronaldo gets into the box because he is perceived as an outside threat so defenders play him differently (press outside the box) which allows him (via dribbling skills) to then get into the box? So in a way - all of those outside of the box shots would be part of the longer strategy of pulling defenders out to allow him better opportunities in of the box? In other words, could there are game theory dynamics at play that build context beyond the statistical analysis of the singular events?
9:51 When you say "here he gets the ball and at this point he has a less than 1% chance of scoring", this refers to "if he shot from where he got the ball", not "for this attack he has a 1% chance of scoring", right?
I'm pretty sure it is a edit mistake, because at 5:06 we can see what he is showing. Besides, It doesn't infringe any copyright or something like that.
Expected goals is more of a fallacy. It is one of those things that sound so nice and revolutionary but is not realistic. So who calculates the expected goals? Is it the fans? Every game should have an officially released expected goal; we should start taking expecting goals seriously only when that happens. The expected goal will be next to the ball possession, cards, corners and offsides stats in every game
It's a statistical model, so there isn't such a thing as "official" expected goals becuase the models can be designed differently or use different datasets. you've shown an absolute lack of understanding of the thing you're complaining about.
Thabks for these videos. Find them really helpful.
I love the idea. Now, the problem of statistics is that they depend on population and samples. Even if expected goal is based historical data, I wonder if you have a league with better Goalkeepers, just to mention one variable, it wouldn't change the outcome then if that's true, I wonder if shoots on target wouldn't make more relevant metric to quantify. Although shoot on target doesn't give you the chances to score, it is relevant to the position you are shooting from.
Should this also be broken down by the stronger foot of the player getting into the position?
I'm curious about an analysis of expected goals from Arjen Robben actually. I was pretty young when I watched him, but it always felt like he pulled in from the side and scored many goals outside the box on the 7% line that was mentioned
While the probability of scoring a goal from a shot taken on the edge of the 6-yard box might be lower than we think, many times, if the initial shot is blocked by a defender or saved by the goalkeeper, the rebound is tapped in. Is there a metric that would take into account the value of a shot from that position, even if the initial shot itself doesn't go in?
good point..... however the higher probability shots also have the added benefits of resulting in corners, rebounds etc and probably at a higher rate than the lower probability shots do
A very thought-through discussion under a TH-cam Video. Great job!
I'd expect that a shot from a rebound is analyzed as it's own individual shot rather than as part of a set of actions and reactions.
Thanks for the great video! I'd like to add subtitles in Japanese if possible
Could it be that Ronaldo gets into the box because he is perceived as an outside threat so defenders play him differently (press outside the box) which allows him (via dribbling skills) to then get into the box? So in a way - all of those outside of the box shots would be part of the longer strategy of pulling defenders out to allow him better opportunities in of the box? In other words, could there are game theory dynamics at play that build context beyond the statistical analysis of the singular events?
It's called counterplay
9:51 When you say "here he gets the ball and at this point he has a less than 1% chance of scoring", this refers to "if he shot from where he got the ball", not "for this attack he has a 1% chance of scoring", right?
Sir which tools did you use to analyse
4:44 -- 5:05 is blacked out - on purpose?
I'm pretty sure it is a edit mistake, because at 5:06 we can see what he is showing. Besides, It doesn't infringe any copyright or something like that.
No. Thats annoying. Thanks for pointing it out. Luckily the picture is the thumbnail for this video! It's of Barcelona shots by MSN.
Expected goals is more of a fallacy. It is one of those things that sound so nice and revolutionary but is not realistic. So who calculates the expected goals? Is it the fans? Every game should have an officially released expected goal; we should start taking expecting goals seriously only when that happens. The expected goal will be next to the ball possession, cards, corners and offsides stats in every game
It's based on historical shot data.
It's a statistical model, so there isn't such a thing as "official" expected goals becuase the models can be designed differently or use different datasets. you've shown an absolute lack of understanding of the thing you're complaining about.