*******Spurs v PSG*******Official OMT*******

[K.D.D.D.D.Soc]

Possession says more about style of play and game flow than quality.

Shots, and somewhat better xG tells a bit. It won't tell if a team will win or lose a game. But give me an xG for of 0.1 or 2.0 in any given game and I'll take the latter. Give us an average xG for of 1.5 for the rest of the season and we're almost certainly going to get better results than if it's 1.0 (assuming no changes to xG against).

Results are what matters. But results are over time clearly linked to performances. xG gives an indication of performance while obviously not being some perfect factual answer. I'd rather see a loss with xG numbers like against PSG than a loss with xG numbers like against Arsenal. At least the PSG game gives us some realistic reasons to think we'll do well in the future.

The xG thing reminds me of the formulas used by all those blokes ripping up their betting slips in the bookies.

 

[Mikey10]

I have to admit, i wasn't impressed yesterday once it went to 3-2....

i felt the players were starting to tire and we needed fresh legs to try and wrestle possession back and go back on the attack to get an equaliser.

Instead he seemed to freeze and sit on his hands while things unfolded and his subs i think actually made us worse and largely didn't make sense to me.

I'll be much more upbeat if we start on the front foot vs Fulham and attack well and with some creativity. But i wasn't impressed with how Frank looked like a rabbit in the headlights yesterday once PSG started to turn the screw from 2-2

I was less impressed with the incredible levels of stupidity displayed by a couple of our players, to be honest.

 

[K.D.D.D.D.Soc]

Is it?

Cricket was the first stat heavy sport I believe.

I think Test "stats" give a reasonable judgement on players and performance but still don't reflect condition of pitch or weather. First class average are a useless tool to gauge the suitability of a player to play test cricket.

 

[K.D.D.D.D.Soc]

It's a somewhat glorified shot statistic.

But it also tells us more than a shot statistic.

A stat doesn't have to be 100% accurate to be useful.

Exactly as demonstrated by politicians, pundits and management consultants.

 

[Bedfordspurs]

I have to admit, i wasn't impressed yesterday once it went to 3-2....

i felt the players were starting to tire and we needed fresh legs to try and wrestle possession back and go back on the attack to get an equaliser.

Instead he seemed to freeze and sit on his hands while things unfolded and his subs i think actually made us worse and largely didn't make sense to me.

I'll be much more upbeat if we start on the front foot vs Fulham and attack well and with some creativity. But i wasn't impressed with how Frank looked like a rabbit in the headlights yesterday once PSG started to turn the screw from 2-2

Fulham is the test
PSG was a free hit and we made it a contest

 

[braineclipse]

Not to be pedantic but I’m gonna be pedantic.

If you have:
- 25 shots
- Total xG ≈ 2.0
- No individual chance over, say, 0.15 xG

Then what you really had was:
- Lots of low-probability efforts
- Probably many blocked shots, tight-angle attempts, or pot-shots from distance
- Pressure but not threat


So while the cumulative xG says “you created enough to score twice”, the experience of the match will tell us differently.
“We never really looked likely to score — we just took a lot of low-quality shots.” xG per shot, shot locations, big chances created, etc, is more informative not just the raw xG total.

What I find interesting is that in the period when we had Kane and Son I don't think we appreciated how far above the statistical norm their finishing was. Kane & Son were absolutely xG overperformers. xG models tell you what an average player would score. Kane and Son were not average players. We often had games where the xG said “1.1 expected goals,” but Kane would stick one in the top corner or Son would curl in something from 0.06 xG.

I think we're in agreement? Create enough of those low quality chances to get an xG of 2 and on average you'll score two goals in that game none the less. And I'd rather have that than an xG under 1 or close to 0. For the experience of the game, particularly if we don't end up scoring, it's perhaps not great. But not so bad for the games where we do score twice?

An xG of 0.15 means a 15% chance of scoring given average finishing, no? 15% isn't that bad really for a chance. But of course if it's somehow a cumulation of really really low probability efforts that's not useful. But that's more of an edge case. It's not like teams are trying to pad their xG numbers by taking low probability shots.

Yes xG is also a way of quantifying how good players are at finishing/shooting, given a large enough sample size. And if you have as a whole as a team either really good or poor finishers that will make xG stats a bit misleading.

 

[Bedfordspurs]

I think we're in agreement? Create enough of those low quality chances to get an xG of 2 and on average you'll score two goals in that game none the less. And I'd rather have that than an xG under 1 or close to 0. For the experience of the game, particularly if we don't end up scoring, it's perhaps not great. But not so bad for the games where we do score twice?

An xG of 0.15 means a 15% chance of scoring given average finishing, no? 15% isn't that bad really for a chance. But of course if it's somehow a cumulation of really really low probability efforts that's not useful. But that's more of an edge case. It's not like teams are trying to pad their xG numbers by taking low probability shots.

Yes xG is also a way of quantifying how good players are at finishing/shooting, given a large enough sample size. And if you have as a whole as a team either really good or poor finishers that will make xG stats a bit misleading.

Its measure is using all the times that has gone it from that place, from shots taken.
So it could have been a top drawer finisher who’s scores one and misses one in identical situations making it a 0.5 probability (if that makes sense)
The key that elite players score those chances more often and less elite ones don’t, create the average

 

[Bishop]

I think we're in agreement? Create enough of those low quality chances to get an xG of 2 and on average you'll score two goals in that game none the less. And I'd rather have that than an xG under 1 or close to 0. For the experience of the game, particularly if we don't end up scoring, it's perhaps not great. But not so bad for the games where we do score twice?

An xG of 0.15 means a 15% chance of scoring given average finishing, no? 15% isn't that bad really for a chance. But of course if it's somehow a cumulation of really really low probability efforts that's not useful. But that's more of an edge case. It's not like teams are trying to pad their xG numbers by taking low probability shots.

Yes xG is also a way of quantifying how good players are at finishing/shooting, given a large enough sample size. And if you have as a whole as a team either really good or poor finishers that will make xG stats a bit misleading.

You're mis understanding a pretty important distinction. Every shot is its own event with its own probability:

• A 0.15 xG chance means:
  

  “An average player would score this 15% of the time.”

• A 0.03 xG shot from 25 yards might look nice in the highlights, but:
  

  “You’re scoring that about 3% of the time.”

This is the problem with looking at xG as a cumulative stat.

But those probabilities aren’t additive events in the real, lived match. They don’t reinforce each other. Each shot is an independent low-probability moment.

If you take 20 shots at 0.10 xG each, you reach something like 2.0 xG.

But what does that really mean?

• Each of those shots is 90% likely not to be a goal.
• The match can easily end with 0 or 1 goals, and that’s not surprising.

xG isn’t telling you what should happen in the match — it’s describing the quality of each chance independently.

xG is a per-event probability; total xG is just the sum of many small independent events, not a prediction of the final score.

The way you're thinking of xG you would think a higher number would better when in reality
it just means more summed probability. A team with 1.0 xG from a few big chances can be far more dangerous than a team with 2.0 xG from 25 weak ones. Cumulative xG doesn't tell you anything about the quality of the individual chances.

 

[Mikey10]

You're mis understanding a pretty important distinction. Every shot is its own event with its own probability:

• A 0.15 xG chance means:
  
• A 0.03 xG shot from 25 yards might look nice in the highlights, but:
  

This is the problem with looking at xG as a cumulative stat.

But those probabilities aren’t additive events in the real, lived match. They don’t reinforce each other. Each shot is an independent low-probability moment.

If you take 20 shots at 0.10 xG each, you reach something like 2.0 xG.

But what does that really mean?

• Each of those shots is 90% likely not to be a goal.
• The match can easily end with 0 or 1 goals, and that’s not surprising.

xG is a per-event probability; total xG is just the sum of many small independent events, not a prediction of the final score.

The way you're thinking of xG you would think a higher number would better when in reality
it just means more summed probability. A team with 1.0 xG from a few big chances can be far more dangerous than a team with 2.0 xG from 25 weak ones. Cumulative xG doesn't tell you anything about the quality of the individual chances.

So, in short, it’s absolute bunkum.

As I suspected.

 

[Bishop]

So, in short, it’s absolute bunkum.

As I suspected.

As I keep saying if you look it as a way to evaluate the quality of your finishing it says a lot, but if you're looking at it as some sort determination of chance quality, especially if looking at the cumulative figures then you're kinda missing the point. So yeah maybe bunkum.

 

[Muttley]

Here's a question for all you xPerts.

When Spence got to the by-line last night, refused to use his left foot and then got his right foot cross blocked at source what does that do for xG. He could clearly slide it into the 18 yard box with his left if he wasn't being such a numpty. There should have been a chance created.

 

[Mikey10]

Here's a question for all you xPerts.

When Spence got to the by-line last night, refused to use his left foot and then got his right foot cross blocked at source what does that do for xG. He could clearly slide it into the 18 yard box with his left if he wasn't being such a numpty. There should have been a chance created.

If you adjust his intention to cross by 0.7 and multiply it by his inability to do so - which works out at 0.38 - you are left with a number which is twice the value of the hypotenuse of a triangle.

That’s what Frank and his coaching team should be thinking about before Saturday.

 

[Bishop]

Here's a question for all you xPerts.

When Spence got to the by-line last night, refused to use his left foot and then got his right foot cross blocked at source what does that do for xG. He could clearly slide it into the 18 yard box with his left if he wasn't being such a numpty. There should have been a chance created.

All the xG models I know of only note chances when a shot on goal was taken. This can be subjective but I would assume that the moment you're talking about wasn't included in any xG stats.

 

[Bedfordspurs]

Here's a question for all you xPerts.

When Spence got to the by-line last night, refused to use his left foot and then got his right foot cross blocked at source what does that do for xG. He could clearly slide it into the 18 yard box with his left if he wasn't being such a numpty. There should have been a chance created.

Nothing. It’s not a shot mate
There is data on passing and creation using XG chain
It’s not a chance unless it results in a shot

 

[K.D.D.D.D.Soc]

I'd have loved to have seen Terry Venables explaining this to Gazza so that he could be more effective in games.

 

[Bedfordspurs]

I'd have loved to have seen Terry Venables explaining this to Gazza so that he could be more effective in games.

TBF Terry enables couldn’t even ensure Gazza turned up to training without wild animals, a mobile home, or bricking in Erik Torsvedts gloves

 

[K.D.D.D.D.Soc]

So, in short, it’s absolute bunkum.

As I suspected.

No its showing us thicko's that you got a better chance of scoring from 1 yard than 20 yards.

 

[K.D.D.D.D.Soc]

TBF Terry enables couldn’t even ensure Gazza turned up to training without wild animals, a mobile home, or bricking in Erik Torsvedts gloves

But you could rely on him turning up for training and performing in games, Gazza in his prime would be lucky to get a game for Hartlepool now, no tactical nous and tried to go forward too much.

 

[Bedfordspurs]

But you could rely on him turning up for training and performing in games, Gazza in his prime would be lucky to get a game for Hartlepool now, no tactical nous and tried to go forward too much.

Hell of a player though

 

[billyiddo]

As I keep saying if you look it as a way to evaluate the quality of your finishing it says a lot, but if you're looking at it as some sort determination of chance quality, especially if looking at the cumulative figures then you're kinda missing the point. So yeah maybe bunkum.

If a shot taken has a high xg then surely by definition it is a good chance that has been created? Whether or not the person on the end of the shot has good finishing ability doesn't change that. Agree on the cumulative figures.