Watford (Home) Saturday August 17th

[SpartyBlue]

On 18/08/2019 at 07:25, Bailey said:

If Richarlison should have scored 1 of his 2 then that is already 0.5 for each attempt taking us over 1 expected goal even forgetting about the other chances. 

I think people overestimate how likely a goal is from the positions Richarlison found himself in. Those were good chances but no way is he supposed to score 50% of the time from there. 

[Romey 1878]

2 hours ago, Cornish Steve said:

Thanks for the explanation. Given the infinite number of combinations of player positions, direction, and speed, the effect of wind and rain, the effects of tiredness and how long a player's been on the field, etc., this would seem to be a very subjective statistic. Pattern matching of that sort would need some hefty computing power, which makes me very skeptical. 

The sentence you’re looking for is “it’s a load of shite.”

[MikeO]

44 minutes ago, Romey 1878 said:

The sentence you’re looking for is “it’s a load of shite.”

Succinctly put as always. 

[markjazzbassist]

yeah americans love stats, like really LOVE stats.  most of our popular sports are super stats heavy (sabremetrics).  But i've always felt soccer is the one sport where they just don't work well.  i have thought about it and tried to figure out why, but the off the ball stuff isn't easily quantifiable and most of that is extremely important stuff to the game.  positioning, movement, dragging defenders, blocking passing lanes, etc.  that's why the stats fall short.

[Romey 1878]

2 hours ago, MikeO said:

Succinctly put as always. 

I try  

[Bailey]

On 19/08/2019 at 15:02, Cornish Steve said:

Thanks for the explanation. Given the infinite number of combinations of player positions, direction, and speed, the effect of wind and rain, the effects of tiredness and how long a player's been on the field, etc., this would seem to be a very subjective statistic. Pattern matching of that sort would need some hefty computing power, which makes me very skeptical. 

All stats are to be fair. Is it a completed pass if its that bad it goes to the wrong player? Or if it hits him at knee height when he cant possibly control it? Is it an interception if the ball goes to a teammate or just a tackle? What about if 2 players going to the tackle and the bump from player 1 allows player 2 to take the ball away? Even some possession stats differ as some only take into account the time the team is on the ball whilst it is in play whereas others include the time the ball is not in play but the clock is going (ie a goal kick or throw in). All stats can only show black and white.

As a different example, in boxing I remember Pauli Malinaggi slating punching stats such as whether a punch connects or doesn't. He said it all depends on who is pressing the button and whether they see it in the first place and then how they interpret it.

On 19/08/2019 at 15:13, SpartyBlue said:

I think people overestimate how likely a goal is from the positions Richarlison found himself in. Those were good chances but no way is he supposed to score 50% of the time from there. 

That is true in a sense. I think it would be hard to say that a professional footballer at that level shouldn't be hitting the target. Then you have to factor in the chance of the keeper saving the shot. One of the chance was plum though, a good header would have given the keeper very little chance to make the save unless it was right at him. The other one was a little further out and been more difficult but it would have been far higher than 0.05.