• Dear Guest, Please note that adult content is not permitted on this forum. We have had our Google ads disabled at times due to some posts that were found from some time ago. Please do not post adult content and if you see any already on the forum, please report the post so that we can deal with it. Adult content is allowed in the glory hole - you will have to request permission to access it. Thanks, scara

**** Tottenham Hotspur v SCBC ****

Personally, I think the whole xG debate is a sign of the times. People want 'science', 'facts' and 'objectivity' (or so those who decide which stats you get believe, anyway). If you enjoy that, more power to you. For me, football is all about perception and nothing beats watching the game, preferably in the stadium. History is full of examples of donkeys being made to look like supermen by astute managers because, no matter what game theory tells you, a player will never be the sum of his stats (just ask Lukaku).

As for the game itself, it was fairly enjoyable. We really looked like we know what we're doing. The players didn't lose their nerves when Soton scored on their first shot of the game and that's a good thing too. I also think we looked more fluid than last season, with different players able to make things happen in different areas. Judging by this game, it looks like we have different options to deal with a problem and that's good news. Now, it's just one game and it's just Southampton. Only when the schedule gets more busy will we know if we're really up for it but, at least, it looks like we will be challenging for the Top 4 again which, one year ago, seemed like a dream.

Personally I like a mix between stats and watching the game. But I agree that if things like xG are brought up as "facts" that somehow disprove what one sees that misses the point. I find it interesting when stats don't match my impressions, gets me thinking and talking about the game and I enjoy that.

Agree fully on the game. Impressive, but just one game and at this point of the season it's very difficult to evaluate the opposition. Southampton are really inconsistent even during the season so that doesn't make it any easier.

When the fixture list congests and we go up against teams that are better we'll get different and harder tests. Hopefully we'll still see that fluidity and multiple solutions.

Right now for any opposition we should be a difficult team to face. So much quality and range of different qualities too. Hopefully that will still be the impression as the season progresses.
 
I think xG is useful, but less useful when looking at a single game. Even for a single game it can give additional information compared to pure shots/shots on target stats, but still.

Not quite sure how those xG models are produced. But looking at the Sessegnon header... Most headers in that position will be significantly worse chances.

Perfect cross with a lot of power. Attacking player able to really attack the ball with the defender backtracking. Compared to most headers from that position from a corner for example it's a significantly better chance. Much worse for the defenders, much easier to attack and score. I don't think the xG models factor that in.
I don’t believe XG takes into account (because it’s kinda impossible) key variables in football
Off the top of my head….
The time of the game - huge impact on pressure of a shot which affects the shooter
Weather - again rain, wind or even a dry pitch as Klopp likes to complain about
The shooter - we all know a striker will score more chances than a defender
The opposition - standard of opposition is so varied and some are better than others at blocking or saving shots for example

But it goes use historical info to predict the future which is a key part of statistics
 
I don’t believe XG takes into account (because it’s kinda impossible) key variables in football
Off the top of my head….
The time of the game - huge impact on pressure of a shot which affects the shooter
Weather - again rain, wind or even a dry pitch as Klopp likes to complain about
The shooter - we all know a striker will score more chances than a defender
The opposition - standard of opposition is so varied and some are better than others at blocking or saving shots for example

But it goes use historical info to predict the future which is a key part of statistics

It would be interesting to see the career attempts to goals ratio of our squad, I expect Dier and Sanchez score quite highly
 
It would be interesting to see the career attempts to goals ratio of our squad, I expect Dier and Sanchez score quite highly
That should be a factor then shouldn’t it? The success rate and accuracy of the individual
That would make it more accurate but is also it impossible to do
 
Personally I like a mix between stats and watching the game. But I agree that if things like xG are brought up as "facts" that somehow disprove what one sees that misses the point. I find it interesting when stats don't match my impressions, gets me thinking and talking about the game and I enjoy that.

We're on the same page: stats are only interesting if they challenge your conclusions - not when they're used to dismiss different opinions. Some players, for instance, are deceptively lazy. When you look at the numbers, they tend to cover a lot of ground. On the flip side, stats tend to deal with higher volumes when football is (also) all about moments of brilliance that change the course of a game. Would Gascoigne be held in the same esteem if you had the same level of details on his game as we do now? I don't know but I sure wouldn't swap that free-kick against Arsenal in the FA Cup for all the xGs in the world!

From that point of view, football is different from American sports. All in all, I think there's been a shift in the last few years. Stats used to be an analysis tool; nowadays, they tend to be used as a predictive tool.

As for the same itself, at least, it makes me eager to see more, which is probably as much as you can demand from a season opener.
 
We're on the same page: stats are only interesting if they challenge your conclusions - not when they're used to dismiss different opinions. Some players, for instance, are deceptively lazy. When you look at the numbers, they tend to cover a lot of ground. On the flip side, stats tend to deal with higher volumes when football is (also) all about moments of brilliance that change the course of a game. Would Gascoigne be held in the same esteem if you had the same level of details on his game as we do now? I don't know but I sure wouldn't swap that free-kick against Arsenal in the FA Cup for all the xGs in the world!

From that point of view, football is different from American sports. All in all, I think there's been a shift in the last few years. Stats used to be an analysis tool; nowadays, they tend to be used as a predictive tool.

As for the same itself, at least, it makes me eager to see more, which is probably as much as you can demand from a season opener.

I agree with your point of American sports which are dominated by stats and analysis. All sport now is dismantled by experts and players and teams drilled in game plans, which to my mind takes something away from it, that element of surprise is what makes the competition. Listening to radio at weekend there seemed to be a slight disappointment that Fulham and Brighton spoilt the days for us.
 
That should be a factor then shouldn’t it? The success rate and accuracy of the individual
That would make it more accurate but is also it impossible to do

It becomes an output rather than an input. With a large enough sample size players overperforming compared to their xG says something about their finishing.

As an input it would be less useful. Need a big sample, takes years. Won't be possible for many players.

If we're overperforming our xG by a bit, that can be sustainable because that's just Kane and Son being Kane and Son.

One of the interesting results of xG to me is just how rare it is for players to significantly outperform their xG over time.
 
so we're media darlings after one nice win over what will be a bottom 10 team this year. There are good reasons for optimism, we went a goal down and effortlessly got back on top, how often have Tottenham teams of recent years done that. Good teams spread the wealth and it was great to see 3 different scorers and not our big 2 on the scoresheet. Our new signings didn't have to start, as long as we stay healthy, we have a decent squad and this new 5 sub rule should greatly benefit us.
Proper test next week and everyone knows the abysmal record we have at the small club in fulham.
 
I agree with your point of American sports which are dominated by stats and analysis. All sport now is dismantled by experts and players and teams drilled in game plans, which to my mind takes something away from it, that element of surprise is what makes the competition. Listening to radio at weekend there seemed to be a slight disappointment that Fulham and Brighton spoilt the days for us.
American sports are very repeatable… hence the set plays and why the Americans struggle to make films about football (read Ossies story about escape to victory for a great laugh on that)
 
Some observations from Saturday:

Kulusevski MOM hands down for me. Glad he capped his performance with a goal.

Thought we reacted really well to going behind and bossed the game after they scored.

Was great to hear Enter Sandman before the players came out of the tunnel. Been wanting the club to play a song like that for years. Hopefully it’s not a one off.
 
Personally, I think the whole xG debate is a sign of the times. People want 'science', 'facts' and 'objectivity' (or so those who decide which stats you get believe, anyway). If you enjoy that, more power to you. For me, football is all about perception and nothing beats watching the game, preferably in the stadium. History is full of examples of donkeys being made to look like supermen by astute managers because, no matter what game theory tells you, a player will never be the sum of his stats (just ask Lukaku).

Good post, as someone who worked in coaching for several years stats can be useful but only to a point. Far too many folks put too much faith in stats and they can be used wrongly to try and make a point they are trying to push.
 
Back