[Football] At the moment, Graham Potter is too good for Brighton. He needs better players.

[Swansman]

Is that why Richarlison and Werner have scored 10 goals each ?

First: no thats not "why" since thats not how xG works, second: none of them have scored 10 goals in the PL this season.

 

[Robinjakarta]

Exactly.

People saying 'sod xg stats, I want more goals and points' don't seem to realise that the xg stats just back up, at it's simplist level, pub chat: fans discussing the ins and outs of a game after it's finished - where it went right/wrong, why the season's going right or wrong overall, etc.

Surely all football fans analyse a game once it's over to cartain extent?

I don't need backing up of what I've seen for myself. So yeah, sod xg stats.

 

[keaton]

Is that why Richarlison and Werner have scored 10 goals each ?

I'm not sure of your point? In the premier League Werner's scored 5 goals and Richardson 4.Maupay has scored 7. So either by xg or goals scored he is outperforming two £50m strikers

 

[Swansman]

I don't need backing up of what I've seen for myself. So yeah, sod xg stats.

Understandable, information can sometimes be a threat to perception and prejudice.

 

[zefarelly]

Just going off BBC website stats, goals and games.

 

[warmleyseagull]

A more forensic analysis of what everyone knows: Brighton are too good to go down but they might because of poor defending early season and continued poor finishing that was evident last season and has become even more acute this.

The comparison with Newcastle, Burnley and Palace is clear: they are shit and they know they are. So they play percentage football; keep it tight and go for something on the break. Tuesday should be kept in the library of all coaches to show how to absorb pressure and win a game against all the stats.

Potterball is all about non percentage football: play expansively, keep possession, dominate. It works when you have the players; GP doesn't. It is unfair, as the article sets out, to blame Maupay solely; Lallana hit row Z a couple of times from great positions on Tuesday; Trossard and Connolly are equally wasteful. It is GP's misfortune to have serial non finishers in his squad.

And it is a call for us as fans too: consensus on NSC is that GP's style is good watching and that it will in the end lead us to promised land. But it is high stakes: relegation is a possibility when we should be challenging for top 8; a reversion to CH style might be safer. In fact, given the paucity of goals in last few games it seems to be working out slightly like that.

 

[Bry Nylon]

I'm ok with the concept of xG but of course it is still the output of a subjective opinion and should be treated a such.

I think part of the problem I have with it is the name 'Expected Goals' Are they really expected goals? I see our xG for the Palace game was off the scale but their keeper barely made a save all night, in spite of the promising positions we got ourselves into. So on Monday night, my personal 'expected goals' for us ended up being somewhere between zero and the one we scored.

I think a far better name for this methodology would be 'Potential Goals'. I definitely feel we created the potential to score 2/3 goals but on what I saw I didn't really expect us to convert that many opportunities.

But I guess xG sounds sexier, easier to sell and more scientific than pG?

 

[Uh_huh_him]

What I think this shows (and some of us have warned about for a while, is you simply cannot do the Premier League "on the cheap". We've tried this so far, especially up front, and rather than spending money on one proven striker we've tinkered with cheaper options (Locadia, Andone, Maupay, Ali J, Connolly), and while this has just about kept our heads above water to this point there's no guarantee it will continue to do so, indeed it might not even do so come this May.

The point that a lot of posters miss is that Tony is not doing anything "on the cheap".
What they mean is pay top dollar for an established premier player at his peak/on the decline.

Tony is unlikely to sanction these type of deals, for many reasons, but primarily because he's going to lose money on every one (see Parish, Steve).

The only way we are going to be able to compete financially in this division, is by shrewd purchases and generating income from player sales.
We aren't there yet, but we are getting closer.

 

[warmleyseagull]

In the pub with mates and a few pints, leave it there and move on.

not pouring over tedious stats on spread sheets all week.

old fashioned, yes. but we ( well most of us) still live in a variable natural world, sanitisation of sport is killing it.

Better get used to it, because it isn't going away. The problem with pub discussion is wrong analysis and wrong solutions. Fine for fans, not for professionally run PL clubs. As it turns out, this is an example where both accord: the stats back up what we all know, but it isnt always like this.

 

[A1X]

The point that a lot of posters miss is that Tony is not doing anything "on the cheap".
What they mean is pay top dollar for an established premier player at his peak/on the decline.

Tony is unlikely to sanction these type of deals, for many reasons, but primarily because he's going to lose money on every one (see Parish, Steve).

The only way we are going to be able to compete financially in this division, is by shrewd purchases and generating income from player sales.
We aren't there yet, but we are getting closer.

I don't deny there's a reason for it, and the reasoning makes sense, but up front is not a position where you can afford to skimp and save. Besides, if you added together the money we spent on punts on Jahanbakhsh, Locadia and Andone you could afford a decent forward.

 

[Uh_huh_him]

I assume xG ignores the quality of the striker/defence?

So a decent chance falling to Dan Burn is the same as the exact same chance falling to Mo Salah?

 

[Swansman]

I'm ok with the concept of xG but of course it is still the output of a subjective opinion and should be treated a such.

I think part of the problem I have with it is the name 'Expected Goals' Are they really expected goals? I see our xG for the Palace game was off the scale but their keeper barely made a save all night, in spite of the promising positions we got ourselves into. So on Monday night, my personal 'expected goals' for us ended up being somewhere between zero and the one we scored.

I think a far better name for this methodology would be 'Potential Goals'. I definitely feel we created the potential to score 2/3 goals but on what I saw I didn't really expect us to convert any.

But I guess xG sounds sexier, easier to sell and more scientific than pG?

xG is pretty much... xG. If situation x happens, how often does it turn into a goal? Pretty scientific. Potential goals? Every attack could potentially result in a goal, but every attack wont statistically be expected to result in a goal. Not saying that one of those stats are more valuable than the other but the numbers would be very different so it would be more than a name change. xG would remain somewhere between (usually) 0-3,5 and pg would be somewhere between 200 and 300.

About their keeper barely making saves.. its true. But if you shoot the ball over the bar from a great position its still a good chance even if the keeper does not have to save it. If you blow a penalty over the bar it doesnt mean its less of a good chance than if their keeper saves it.

 

[Uh_huh_him]

I don't deny there's a reason for it, and the reasoning makes sense, but up front is not a position where you can afford to skimp and save. Besides, if you added together the money we spent on punts on Jahanbakhsh, Locadia and Andone you could afford a decent forward.

There's an element of truth to that, but there are also some issues with that approach.
1) wages - paying top striker wages opens up negotiations to all other "high performing" players
2) We can only afford one marquee forward - injuries, suspension, loss of form leaves you high and dry ( see Wilf Zaha)
3) No guarantees the top striker will perform at a new club

 

[Swansman]

I don't deny there's a reason for it, and the reasoning makes sense, but up front is not a position where you can afford to skimp and save. Besides, if you added together the money we spent on punts on Jahanbakhsh, Locadia and Andone you could afford a decent forward.

1. Manchester City. Phil Foden: £0m
2. Manchester United. Marcus Rashford: £0m
3. Leicester City. Jamie Vardy: £1m
4. West Ham. Michail Antonio: £7m

 

[CheeseRolls]

The stat obsession imported from gridiron, is both fascinating and the dullest thing imaginable. It keeps dragging me in and as I move in to take a closer look, some dullard whacks me round the head with it to prove their point.

Take outs for me from the Monday night meltdown.

1 Our main attacking threat came from two centre backs deployed as wing backs. Superb effort going forward by both, but if we are one super striker away from world domination, is there maybe scope for more natural cover in these positions?

2 Against a team parking the bus our midfield movement becomes very laboured. Watching Alzate play his strength for me is close control, short passing and constant movement. What I don't see enough of is him getting his head up and playing the long pass.

3 That Lalanna chance from near the penalty spot was the worst miss of the season bar none. Absolutely unforgivable miss that was nowhere near the target.

4 Maupay didn't really have many opportunities to score. What was more telling, was that in the circumstances he didn't do enough for me to bring others into the game, either laying the ball off or making runs to create space for others. This is how good forwards work for your team, when they are tightly man marked and XG stats tell you nothing about this.

5 Getting good forwards is not all about big fees and big wages. We missed out on Darwin Nunez in the summer, because he saw Benfica as a better prospect than little old Brighton. This season has raised our profile in a number of ways, which make me more optimistic we can attract better players.

6 Who knows maybe there is an Emile Heskey type out there, who can chage our scoring fortunes, without scoring many himself.

 

[Bry Nylon]

xG is pretty much... xG. If situation x happens, how often does it turn into a goal? Pretty scientific. Potential goals? Every attack could potentially result in a goal, but every attack wont statistically be expected to result in a goal. Not saying that one of those stats are more valuable than the other but the numbers would be very different so it would be more than a name change. xG would remain somewhere between (usually) 0-3,5 and pg would be somewhere between 200 and 300.

About their keeper barely making saves.. its true. But if you shoot the ball over the bar from a great position its still a good chance even if the keeper does not have to save it. If you blow a penalty over the bar it doesnt mean its less of a good chance than if their keeper saves it.

I can 100% see how xG applies to penalties as it is a repeated format in a controlled environment so it is very easy to produce a conversion ratio of all penalties taken and then say what the 'expected' chance of a goal will be, either for teams in general or a player in particular.

But when the ball is in open play, with multiple external factors influencing any one particular outcome, then saying something is 'expected' becomes way more subjective. Like I say, I see it's value and I find it interesting, but I don't regard it as especially scientific.

 

[Audax]

1. So is our xG fail bad luck, bad skill, inflexible coaching or a mix of the last two (if you accept that "unlucky" is really an outlier where a really stupid deflection happens or the ref gets in the way)?


2. But Bielsa is also a coach who has got the most out of an average striker. Bamford had a career as a journeyman Championship striker, mostly on loan. Now he's already beaten Maupay's best tally with 13 games to play. So you can coach a striker - or, if you can't, then you can give him better quality chances. That may be the actual difference.

1. I think there's another possibility here that you've overlooked: Do the strikers available to us have further untapped potential or have they reached a peak from which no further coaching can improve them? And another possibility: do we simply not have the right mix of strikers - if they're all "too similar", then it makes it easier for opposition teams to train to counter those strikers. If we have significantly different options, then they can't focus their efforts as much (or if they do, they leave a door open).

2. As mentioned in 1 above: is Bielsa simply working with a player who always had that unlocked potential? Whereas maybe Potter's working with a player (players) who are already at or near maximum potential?


I have a feeling my second possibility in 1 above is part of what is costing us. All our players who can shoot are pretty similar. Apart from at set pieces where the tall defenders like Dunk can get involved, we're overly reliant on smaller, agile players, who we hope to get into a space where they can shoot. Opposition managers for me seem to have cottoned on that these players generally aren't quick enough or good enough to get a scoring shot away when the box is crowded, and we don't have any alternative. So they can focus training ahead of a Brighton game on ensuring they transition rapidly to defense and get players into the box to lock down the likes of Maupay, Connolly, Trossard before they can get a clear shot away. That leaves us either taking shots from further out, through a crowded box, or trying to one-touch-pass our way through. The latter, when it works, has produced a couple of brilliant goals with the final shot being a simple tap-in - but has otherwise generally seen the attack break down. The former seems to result in whoever takes the shot snatching at the chance, trying to hit it too hard to get it through before the defense closes the gap, and that's where the accuracy falls apart.

 

[A1X]

1. Manchester City. Phil Foden: £0m
2. Manchester United. Marcus Rashford: £0m
3. Leicester City. Jamie Vardy: £1m
4. West Ham. Michail Antonio: £7m

Foden and Rashford are academy products much in the same way Connolly is for us, I am not including them in the same bracket as the punts we've taken on signing cheaper options and hoping they come good. Vardy and Antonio are both punts which came off, much as Maupay sort of has, but all of those have benefitted from having already played in English football, a pool we seem incredibly reluctant to fish in.

 

[andy1980]

I assume xG ignores the quality of the striker/defence?

So a decent chance falling to Dan Burn is the same as the exact same chance falling to Mo Salah?

Yes .

 

[Swansman]

I can 100% see how xG applies to penalties as it is a repeated format in a controlled environment so it is very easier to produce a conversion ratio of all penalties taken and then say what the 'expected' chance of a goal will be, either for teams in general in a player in particular.

But when the ball is in open play, with multiple external factors influencing any one particular outcome, then saying something is 'expected' becomes way more subjective. Like I say, I see it's value and I find it interesting, but I don't regard it as especially scientific.

Its not perfect, but useful. I think the Liverpool data analyst explains it well here:

https://streamable.com/ay4qq

It provides managers and players with a lot of information on the most efficient areas of scoring and the generally best choices for a lot of situations. But when it comes to individual situations I dont think its saying a whole lot because of the multitude of factors playing into every single situation.