EPTFIFA World Cup 2022 - Knockout Stage - Page 79
| Forum Index > FIFA World Cup 2022 |
|
{CC}StealthBlue
United States41117 Posts
| ||
|
Stratos_speAr
United States6959 Posts
On December 15 2022 23:11 sharkie wrote: If France beats Argentina like 4-1 or something like that the loss is on Messi and him not tracking back or pressing. Just like if Argentina wins 2-1 it will probably be all thanks to Messi. I respect Messi a lot and he deserves most of the praise in the game but Griezmann is not just doing offensive and magical work but also a lot of tracking back, covering and hes everywhere at once. And while doing that he produced more chances than Messi has so for me Griezmann has been just the better player this tournament. I am so surprised to see so many of you disregard the most vital part of football. Not conceding goals/chances But France haven't looked amazing defensively. They've surrendered control of the game for two rounds in a row and gave up plenty of half-chances to Morocco and a whole lot of legitimate chances to England. Also pretty much every advanced stat that I can find says that Argentina's defensive performance has actually been slightly better than France's, whereas France has had a slightly better offensive performance than Argentina. | ||
|
nojok
France15845 Posts
On December 16 2022 02:27 Stratos_speAr wrote: But France haven't looked amazing defensively. They've surrendered control of the game for two rounds in a row and gave up plenty of half-chances to Morocco and a whole lot of legitimate chances to England. Also pretty much every advanced stat that I can find says that Argentina's defensive performance has actually been slightly better than France's, whereas France has had a slightly better offensive performance than Argentina. Football is not an American sports, stats are way less meaningful. Simple exemple : a player tries his chance, a 0.2 xG more for his team, a player misses his pass that would have been a guaranteed goal (Mbappé against Morocco when he had to pass) 0 expected goal. And I could carry on. Argentina lost to Saudi Arabia with their A team, they got a panic attack against Australia which was a walk for France and almost threw away their lead. A desperate kick&rush from Netherlands lead them to concede two goals very quickly. Argentina had all the penalties going their way, including light ones, France conceded penalties in situations which absolutely did not require that level of engagement. In fact, outside of penalties, France conceded no goal in the knock out stage and all of our goals come from open play, which is incredibly rare at this level (and a flaw tbh). And yeah France is happy to give up on the ball. And that's exactly what cost Croatia so much, they barely tried to attack against Brazil while they tried to take the control of the game against Argentina, don't expect us to do that. On the mentality side, I feel we're favored as well, I could see us recover from a 1 0 lead, not so sure Argentina can cope with the opponent having the lead for more than 15 minutes. But yeah hmm, "advanced stats". | ||
|
Stratos_speAr
United States6959 Posts
On December 16 2022 02:55 nojok wrote: Football is not an American sports, stats are way less meaningful. Simple exemple : a player tries his chance, a 0.2 xG more for his team, a player misses his pass that would have been a guaranteed goal (Mbappé against Morocco when he had to pass) 0 expected goal. And I could carry on. Argentina lost to Saudi Arabia with their A team, they got a panic attack against Australia which was a walk for France and almost threw away their lead. A desperate kick&rush from Netherlands lead them to concede two goals very quickly. Argentina had all the penalties going their way, including light ones, France conceded penalties in situations which absolutely did not require that level of engagement. In fact, outside of penalties, France conceded no goal in the knock out stage and all of our goals come from open play, which is incredibly rare at this level (and a flaw tbh). And yeah France is happy to give up on the ball. And that's exactly what cost Croatia so much, they barely tried to attack against Brazil while they tried to take the control of the game against Argentina, don't expect us to do that. On the mentality side, I feel we're favored as well, I could see us recover from a 1 0 lead, not so sure Argentina can cope with the opponent having the lead for more than 15 minutes. But yeah hmm, "advanced stats". Sorry to break it to you, but football is not special. This is the same nonsense Boomer mentality that you see in every sport; they did it in baseball before the advent of advanced anaylitics, they did it in American football before the current era, they did it in basketball before maybe ~5-8 years ago, and they still do it in hockey. It's little more than old school fans/pundits/analysts that don't want to let go of their vibes and feelings and instincts and "eye tests" about player and team quality because they're scared of/unfamiliar with statistics. The reality is that statistics, when used correctly, can be meaningful for any sport. Advanced footballing stats are in their infancy but have already proven to be useful and predictive in many ways. Things like the WC are vulnerable to small sample variability but it doesn't mean that the statistics are useless. The rest of your post is just homer rationalization. Pretty much everything shows us that France has an incredibly deadly attack but midfield and defensive vulnerabilities. They've got a great chance to win but there's also a good reason they aren't heavily favored (or favored at all) by betting markets or statistical models. I'm excited for the final and I think it'll be a good one. I'm just curious to see how people like you rationalize it if France ends up losing. | ||
|
Jockmcplop
United Kingdom9765 Posts
On December 16 2022 02:55 nojok wrote: Football is not an American sports, stats are way less meaningful. Simple exemple : a player tries his chance, a 0.2 xG more for his team, a player misses his pass that would have been a guaranteed goal (Mbappé against Morocco when he had to pass) 0 expected goal. And I could carry on. Argentina lost to Saudi Arabia with their A team, they got a panic attack against Australia which was a walk for France and almost threw away their lead. A desperate kick&rush from Netherlands lead them to concede two goals very quickly. Argentina had all the penalties going their way, including light ones, France conceded penalties in situations which absolutely did not require that level of engagement. In fact, outside of penalties, France conceded no goal in the knock out stage and all of our goals come from open play, which is incredibly rare at this level (and a flaw tbh). And yeah France is happy to give up on the ball. And that's exactly what cost Croatia so much, they barely tried to attack against Brazil while they tried to take the control of the game against Argentina, don't expect us to do that. On the mentality side, I feel we're favored as well, I could see us recover from a 1 0 lead, not so sure Argentina can cope with the opponent having the lead for more than 15 minutes. But yeah hmm, "advanced stats". There's like 4 teams who are in the Premier League right now because they switched to a stat based 'moneyball' style recruitment and coaching model and immediately got promoted. | ||
|
andrewlt
United States7702 Posts
| ||
|
Stratos_speAr
United States6959 Posts
On December 16 2022 05:41 andrewlt wrote: I'm curious, how advanced are advanced analytics in football. Do they correct time of possession and such based on the tendency of teams to sit with the lead (something I think is frowned on by advanced analytics)? In other sports, possession stats aren't "corrected"; instead there is a known effect for having a lead and therefore having less of the ball/fewer offensive chances than you would normally have. I think hockey is one of (if not the most) affected sports. Statistical analysis and models take this into account when determining things like win probabilities, ELO, etc. I'm 98% sure that this is the same in football. I think that the only reason this would be "frowned upon" (though I don't think this is really true) is because, statistically speaking, giving up possession when you have a lead still reduces your probability of winning when compared to maintaining possession. That's not to say that it can't be done, but in sports like football and hockey, the #1 way to make sure that your opponent doesn't have chances to score is to make sure they don't have the ball (or puck), meaning they physically can't take shots and therefore have less opportunities to score. | ||
|
andrewlt
United States7702 Posts
I'm fairly certain that analytics would recommend that a team with a 1-score lead would be better off playing normally and trying to get a 2-score lead over parking the bus. I've read something to that effect a while ago but I'm not sure if that is still the consensus among data minded analysts. However, it doesn't look like teams practice it. Just this knockout round, I've seen teams cede control of the flow of the game with a lead. France vs England, France vs Morocco, Argentina vs Netherlands, Brazil vs Croatia, Argentina vs Croatia looked like the team with the lead were on the back foot when they had the lead. The reason I bring it up is because France had 39% time of possession against Morocco and 43% against England. France scored within 5 minutes of the Morocco game and had a lead for most of the game against England. Without accounting for France having a lead, you could make the argument that they got outplayed both games. | ||
|
Stratos_speAr
United States6959 Posts
On December 16 2022 07:53 andrewlt wrote: I follow mostly basketball and American football. The common thread I've seen is that traditional game plans err on the side of being too conservative on offense. Analytics almost always recommends something more aggressive than what tradition dictates. I'm fairly certain that analytics would recommend that a team with a 1-score lead would be better off playing normally and trying to get a 2-score lead over parking the bus. I've read something to that effect a while ago but I'm not sure if that is still the consensus among data minded analysts. However, it doesn't look like teams practice it. Just this knockout round, I've seen teams cede control of the flow of the game with a lead. France vs England, France vs Morocco, Argentina vs Netherlands, Brazil vs Croatia, Argentina vs Croatia looked like the team with the lead were on the back foot when they had the lead. The reason I bring it up is because France had 39% time of possession against Morocco and 43% against England. France scored within 5 minutes of the Morocco game and had a lead for most of the game against England. Without accounting for France having a lead, you could make the argument that they got outplayed both games. Plenty of people are making this argument even when you consider that France had the lead. Playing better is much more nebulous in football than a lot of other sports so these debates never really get anywhere, but there is clearly a feasible argument for saying that France has been outplayed at various points. Between the examples you gave and the fact that the stats reliably show that more possession = more consistent chances of winning, I think it's pretty clear that "giving up possession to defend" is an old-school tactic that doesn't have a lot of empirical basis. EDIT: I actually finally found a singular collection of stats: ARGENTINA: Goals = 12, xG = 13.21, Goals Against = 5, xGA = 2.43 FRANCE: Goals = 13, xG = 12.05, Goals Against = 5, xGA = 7.51 France's defense has been significantly worse. This really could've bitten them in the ass if Kane didn't miss a PK. | ||
|
andrewlt
United States7702 Posts
Morocco played 5 European teams this World Cup and the France game was the only one where they had time of possession above 35%. It is obviously a clear outlier. https://www.espn.com/soccer/fifa-world-cup/story/4832549/2022-world-cup-tactical-trends-do-you-even-need-possession With that said, according to the article above, Argentina and France rank 3rd and 4th respectively in xG differential. France might not look invincible but they are clearly one of the top teams in the tournament. I think their stats are skewed downward from employing an outdated tactic while leading. In a tied game or a game where they are down, I expect them to perform better. | ||
|
RKC
2848 Posts
But WC is quite different from club football especially based on league format. Less prep time, packed fixtures, tournament format and rules, climate, etc. So there are some additional intangibles (or measurable data that haven't been properly tracked) to consider: player fatigue, risk of injuries, risk of suspension. One reason why even top teams don't dominate and pressure as much after taking the lead is due to long-term strategic considerations - conservation of resources for future matches. A lot of last minute goals and drama could be attributed to teams thinking ahead and losing focus. Maybe analytics from knockout round style tournaments are better comparables. Such as Champions League. A recent example is RM winning last season despite not dominating matches and conceding a lot of goals. | ||
Liquid`Drone
Norway28735 Posts
And indeed, these statistics are much more relevant for league play than they are for championships, primarily for one reason: sample size. If you play between 3 and 7 matches, football is such a low scoring game that results will frequently differ significantly from performance. During a 38 game or whatever season, stuff tends to even out fairly well (not perfectly though), and you get results that much more closely match overall performance. Championships are much more likely to be decided by individual moments of brilliance (or even luck) - that's part of their magic, but also why the winning team isn't necessarily the best one. | ||
Poopi
France12907 Posts
On December 16 2022 03:39 Stratos_speAr wrote: Sorry to break it to you, but football is not special. This is the same nonsense Boomer mentality that you see in every sport; they did it in baseball before the advent of advanced anaylitics, they did it in American football before the current era, they did it in basketball before maybe ~5-8 years ago, and they still do it in hockey. It's little more than old school fans/pundits/analysts that don't want to let go of their vibes and feelings and instincts and "eye tests" about player and team quality because they're scared of/unfamiliar with statistics. The reality is that statistics, when used correctly, can be meaningful for any sport. Advanced footballing stats are in their infancy but have already proven to be useful and predictive in many ways. Things like the WC are vulnerable to small sample variability but it doesn't mean that the statistics are useless. The rest of your post is just homer rationalization. Pretty much everything shows us that France has an incredibly deadly attack but midfield and defensive vulnerabilities. They've got a great chance to win but there's also a good reason they aren't heavily favored (or favored at all) by betting markets or statistical models. I'm excited for the final and I think it'll be a good one. I'm just curious to see how people like you rationalize it if France ends up losing. If we are honest with ourselves though, France is largely favored if Argentina doesn’t get free penalties (and we don’t give up dumb ones like Hernandez did vs England). | ||
Twisted
Netherlands13554 Posts
 , nice video (for Messi fanboys like me):Had a little smirk when I read a discussion about who's the better player between Griezman and Messi haha. That was entertaining. Problem with Messi is that he always gets compared to.. Messi. He was that much beyond anyone during his peak years, but he's still the best player bar none at 35. The gap is just smaller and the standards people hold him to are.. his own standards from 5-10 years ago. We're witnessing the last hurrah of the greatest player in history. Enjoy it as much as you can. | ||
|
Stratos_speAr
United States6959 Posts
On December 16 2022 12:07 RKC wrote: Generally, I agree that data and stats are useful tools to measure player and team performance. But WC is quite different from club football especially based on league format. Less prep time, packed fixtures, tournament format and rules, climate, etc. So there are some additional intangibles (or measurable data that haven't been properly tracked) to consider: player fatigue, risk of injuries, risk of suspension. One reason why even top teams don't dominate and pressure as much after taking the lead is due to long-term strategic considerations - conservation of resources for future matches. A lot of last minute goals and drama could be attributed to teams thinking ahead and losing focus. Maybe analytics from knockout round style tournaments are better comparables. Such as Champions League. A recent example is RM winning last season despite not dominating matches and conceding a lot of goals. None of those factors are unique to the World Cup. You see them in domestic play as well, particularly the Champion's League. We already discussed the line after that and how this is an outdated "vibesy" strategy that doesn't have much (if any) empirical support; giving up possession still reduces your chance of winning because you're giving your opponent more chances to score. Lots of last minute drama is due to the drastically increased added time in this tournament. There's nearly 11 minutes of total added time per game in this tournament compared to others, and both the amount of goals scored in that time and the xG in that time has increased in a likewise manner. World Cup play is unique in two ways: 1) teams aren't together much so they play simpler, less tactically complex (i.e. "worse") football, and 2) there are very few games so they are subject to extreme variability in performance/outcome. The latter is just a basic statistical concept that can be applied to literally anything. While I have mentioned possession myself, I must state that I think on its own, it's a fairly meaningless statistic, one that can boil down to style. xG is highly relevant, though - but if you base yourself on xG to determine who is the better team (which again, I have to some degree done - I just want to amend it a little), you're partially eliminating goalies and strikers from the equation. For me, it's mostly about the repeatability of a game - you can get a 1-0 victory from a game where you have 0.5 vs 3 xG, but it'll very rarely happen twice in a row, whereas if you get a 2-0 victory from a game where you have 3 vs 0.5 xG, a similar result is fairly likely to happen if you play a rematch. And indeed, these statistics are much more relevant for league play than they are for championships, primarily for one reason: sample size. If you play between 3 and 7 matches, football is such a low scoring game that results will frequently differ significantly from performance. During a 38 game or whatever season, stuff tends to even out fairly well (not perfectly though), and you get results that much more closely match overall performance. Championships are much more likely to be decided by individual moments of brilliance (or even luck) - that's part of their magic, but also why the winning team isn't necessarily the best one. Possession still has a pretty good correlation with odds of winning. It's not perfect (obviously nothing is), but it's definitely not useless. Also xG takes the odds of a shot being saved into account, but yes it doesn't take individual player skill into account; it take's the population's skill into account. This is part of the fun of actually having the game played out. | ||
|
MyLovelyLurker
France756 Posts
On December 16 2022 16:51 Poopi wrote: If we are honest with ourselves though, France is largely favored if Argentina doesn’t get free penalties (and we don’t give up dumb ones like Hernandez did vs England). Really ? I'm French and old enough to remember when we won 4-3 only thanks to Pavard seeing God that day. Since then France is shakier defensively (not surprising given all injuries) and Argentina plays a bit more compactly, so I'm really not comfortable calling it better than a coin flip tbh. I think Argentina is roughly on par with England, and we know how narrowly we won that one. | ||
|
MyLovelyLurker
France756 Posts
On December 16 2022 14:27 Liquid`Drone wrote: While I have mentioned possession myself, I must state that I think on its own, it's a fairly meaningless statistic, one that can boil down to style. xG is highly relevant, though - but if you base yourself on xG to determine who is the better team (which again, I have to some degree done - I just want to amend it a little), you're partially eliminating goalies and strikers from the equation. For me, it's mostly about the repeatability of a game - you can get a 1-0 victory from a game where you have 0.5 vs 3 xG, but it'll very rarely happen twice in a row, whereas if you get a 2-0 victory from a game where you have 3 vs 0.5 xG, a similar result is fairly likely to happen if you play a rematch. And indeed, these statistics are much more relevant for league play than they are for championships, primarily for one reason: sample size. If you play between 3 and 7 matches, football is such a low scoring game that results will frequently differ significantly from performance. During a 38 game or whatever season, stuff tends to even out fairly well (not perfectly though), and you get results that much more closely match overall performance. Championships are much more likely to be decided by individual moments of brilliance (or even luck) - that's part of their magic, but also why the winning team isn't necessarily the best one. I'm a math PhD and I approve of that post. For the technically inclined: 1. xG vs victory correl is determined by shooting on goal opps, over a single match you could have n=2 or n=3 with shoots on post to boot... doesn't get more random than that 2. *non-stationarity* due to deliberately shifting tactics intramatch (Italian catenaccio springs to mind : score the first goal, then defend with 11 players -> huge xG/outcome correlation drop right there) 3. variance of the xG/victory correlation over the last few world cups, that quantity would be 45% bid @ 85% offered, historically... meaning entire world cups of being generally non-predictive... This is not to diss analytics in general, totally valid to count % passes in a season to gauge a player's form and value, because then n=100 to n=1000. But we all know intuitively how rare events such as the ball hitting the post one way rather than another - which will never be predictable statistically - completely change how a single match goes. And for a top 8 team, a world cup really only is about 3 or 4 serious matches. A dozen key shots post group phase is realistically not enough for the law of large numbers to apply. | ||
|
Poopi
France12907 Posts
On December 16 2022 22:48 MyLovelyLurker wrote: Really ? I'm French and old enough to remember when we won 4-3 only thanks to Pavard seeing God that day. Since then France is shakier defensively (not surprising given all injuries) and Argentina plays a bit more compactly, so I'm really not comfortable calling it better than a coin flip tbh. I think Argentina is roughly on par with England, and we know how narrowly we won that one. It’s semi trolling and semi serious. How 2018 went is largely irrelevant given how different the squads / momentum are. The England game was only close because of the penalties though? France seemed able to score during the live game while England only scored during penalty / missed one. They looked impressive for 10-15mn after we scored, but as far as I know, it’s pretty normal in football to try and attack desperately after conceding a goal? France was able to navigate through the storm pretty well and then it calmed down. Argentina should have been eliminated vs Netherlands iirc, but the referees were bad throughout the game and the narrative is that Argentina is favored this WC to « give » Messi a World Cup, which is why I was poking fun at this. In all seriousness though, our squad looked a bit better overall so I stand with my call: if we can avoid penalties, we should win relatively comfortably. | ||
|
RKC
2848 Posts
I do still believe that for an evenly match game (even in terms of both teams having enough time to train and prep in advance like club football, not necessarily even in terms of quality), higher possession will more strongly correlate with higher chances, goals, and wins. But when the game is between two top teams closely even in quality, possession becomes less of a factor as some top teams deliberately cede possession and hit on the break and still become very successful with such tactic (eg Mourinho's Inter, RM in CL last season, etc). Football at the highest level involves a lot of intangibles (or data that we have yet to properly track). | ||
|
MyLovelyLurker
France756 Posts
On December 16 2022 23:13 Poopi wrote: It’s semi trolling and semi serious. How 2018 went is largely irrelevant given how different the squads / momentum are. The England game was only close because of the penalties though? France seemed able to score during the live game while England only scored during penalty / missed one. They looked impressive for 10-15mn after we scored, but as far as I know, it’s pretty normal in football to try and attack desperately after conceding a goal? France was able to navigate through the storm pretty well and then it calmed down. Argentina should have been eliminated vs Netherlands iirc, but the referees were bad throughout the game and the narrative is that Argentina is favored this WC to « give » Messi a World Cup, which is why I was poking fun at this. In all seriousness though, our squad looked a bit better overall so I stand with my call: if we can avoid penalties, we should win relatively comfortably. I hear what you're saying, I don't disagree about Argentina squandering their lead in poor fashion to the Netherlands. Not sure I agree with the 'it's only penalties' point - the second pen France conceded to England was clearly do-or-die, and reflective of the ton of problems Saka and others were giving our fullbacks. I felt like we were clearly dominated in second half, it wasn't a deliberate strategy to sit back and give possession like against Morocco. Also fundamentally I hope you're right  | ||
| ||
