2022 World Cup Discussion

[BlueMoonAcrossThePond]

I can't watch a game if it isn't live. The games have to kick off at the same time and the tension is built by that, so they have to reflect the other game to. It's like City v Villa. We knew Liverpool were losing, then drawing but we all felt as though they would win and so we needed to. The drama was enough in our game alone, but we needed to be aware of what was happening there for it to build further. Even though the title was never Liverpool's on that final day it seemed as though they'd get it at certain points. I know in this case there are likely to be winners and losers in both games, but it needs to be discussed as they play out and not ignored.

Yup - good points.

RE Live versus not - so long as you don't know the outcome, I think watching a DVR replay is just the same as live. Even better since you can skip commercials.

RE tactics due to results in the other game - good point. Still... for those wishing to watch both games, knowing the state of the other match should be optional. For example, I'm pretty sure that in some previous WC broadcasts in the USA, the announcers would say "I'm about to announce the results of the other game - please mute sound for the next 10 seconds if you don't want to know the other score"... but Fox almost certainly won't do this.

Better still - I'd be in favor of a rule (if it doesn't already exist) to ban all forms of outside/electronic communication from both sides in the final qualifying game. Each team should play without knowledge of what's going on in the other game IMO.

 

[SebastianBlue]

 

[BlueMoonAcrossThePond]

Prima facie, this appears to be a sensational post. Is this outcome particularly unusual?

If expected goals are very close in many games, then we'd expect that the outcome of games based on XG to be close as well. Hence 15 out of 32 might well not be at all unusual.

In the same vein, if this is a huge statistical anomaly, then how far off expectation is this outcome? What is the sigma?

 

[supercity88]

Yup - good points.

RE Live versus not - so long as you don't know the outcome, I think watching a DVR replay is just the same as live. Even better since you can skip commercials.

RE tactics due to results in the other game - good point. Still... for those wishing to watch both games, knowing the state of the other match should be optional. For example, I'm pretty sure that in some previous WC broadcasts in the USA, the announcers would say "I'm about to announce the results of the other game - please mute sound for the next 10 seconds if you don't want to know the other score"... but Fox almost certainly won't do this.

Better still - I'd be in favor of a rule (if it doesn't already exist) to ban all forms of outside/electronic communication from both sides in the final qualifying game. Each team should play without knowledge of what's going on in the other game IMO.

I'm just one of those weird people where if I know the game has already been played, I can't watch it. I'd end up fast forwarding through it to get to the good bits.

The communication point is interesting. It obviously has a huge impact on things. If QPR didn't know they were safe would we have beaten them in 2012? Would a team go for the win when a draw would have been enough? Or hold onto a draw when they need to win? Only Niall Quinn will ever know.

 

[SebastianBlue]

Prima facie, this appears to be a sensational post. Is this outcome particularly unusual?

If expected goals are very close in many games, then we'd expect that the outcome of games based on XG to be close as well. Hence 15 out of 32 might well not be at all unusual.

In the same vein, if this is a huge statistical anomaly, then how far off expectation is this outcome? What is the sigma?

I am not sure how it is sensationalist? They are simply stating the fact that the xG is normal based on actual goals scored but that the distribution is abnormal based on individual scores.

As the data universe grows, xG distribution should normalise. If it does not, it can be a sign of aberration. Variance will provide some indications as to the likely cause if that were to happen.

 

[BlueMoonAcrossThePond]

I'm just one of those weird people where if I know the game has already been played, I can't watch it. I'd end up fast forwarding through it to get to the good bits.

The communication point is interesting. It obviously has a huge impact on things. If QPR didn't know they were safe would we have beaten them in 2012? Would a team go for the win when a draw would have been enough? Or hold onto a draw when they need to win? Only Niall Quinn will ever know.

RE communication and knowing results of the other game...

Obviously not knowing the final result of the other match in case it matters is the entire motivation for simultaneous matches in the first place. A complete (not sure how this would be achievable, probably it isn't) communication ban with regard to other team results is even better. At any rate, this seems impossible, as some random spectator in the crowd could be tasked with giving hand signals to communicate results in another game. Confiscating all cell phones to prevent this seems like massive overkill.

 

[BlueMoonAcrossThePond]

I am not sure how it is sensationalist? They are simply stating the fact that the xG is normal based on actual goals scored but that the distribution is abnormal based on individual scores.

It's possibly sensational in that, while true, the implication is that this outcome is abnormal - else why even mention it. The lack of follow on analysis to prove that this outcome is unusual might well be an artifact of Twitter - or more likely, simply the result of wanting to attract attention.

Regardless - how unusual is this outcome? - based on statistical analysis. Who knows based on the Twitter post?

 

[SebastianBlue]

It's possibly sensational in that, while true, the implication is that this outcome is abnormal - else why even mention it. The lack of follow on analysis to prove that this outcome is unusual might well be an artifact of Twitter - or more likely, simply the result of wanting to attract attention.

Well, the current state is abnormal. And I imagine they don’t go in to further analysis because anyone that follows their account and is remotely interested in xG based statistical analysis understands that current distribution is abnormal.

 

[BlueMoonAcrossThePond]

Well, the current state is abnormal. And I imagine they don’t go in to further analysis because anyone that follows their account and is remotely interested in xG based statistical analysis understands that current distribution is abnormal.

Huh?

Suppose that xG difference in these matches is very close - say 51% versus 49%. How unusual then, is the outcome cited? Not very unusual at all.

Based on the Twitter post, there's no indication about how unlikely these results are - perhaps extremely unlikely, perhaps within the realm of 50-50 expectation. Who knows? What's the sigma?

But for sure, it makes for a sensational post.

 

[SebastianBlue]

Huh?

Suppose that xG difference between all teams is 51% versus 49%. How unusual is the outcome cited? Not very unusual at all.

Based on the Twitter post, there's no indication about how unlikely these results are - perhaps extremely unlikely, perhaps within the realm of 50-50 expectation. Who knows?

…you do understand what abnormal means in statistical terminology, right?

It is used to describe something outside of a normal reference range.

I think you are trying to find some seedy motive where none exists. They were describing the current state, not making a determination as to why or what the following outcome would be.