it’s a loss. It’s a fifa regulation since they introduced shootouts in 1978. ( the first shootout was in ‘82). https://backyardsidekick.com/can-soccer-games-end-in-a-tie-how-it-affects-scoring/ in most shootouts it’s a tie and you move on. There is no “moving on” in the final, only the winner being declared the winner.
You're probably right but the article doesn't specifically address the question of how FIFA treats the outcome for the purposes of the ratings. I mean, I'm assuming they're going to score it as a 2-1 win for Canada (and 1-2 loss for Sweden) as opposed to a 1-0 win (I might've said 3-2 earlier or maybe I was only talking about 2011. I'm getting confoozed!) but I don't know. I've never read it anywhere.
Does someone have access to the actual FIFA rule on PK-decided games and FIFA ratings? I know that FIFA says that if the World Cup championship game is decided by PKs, the prevailing team is the winner. But I cannot find anything that says for rating purposes that PK-decided games, including in the WC or any other championship, are treated as anything other than ties. I think the likelihood is that, for rating purposes, they are treated as ties. PS - This is how the NCAA does it for college soccer and their rating system. From a logic perspective, this would be the correct way to do it.
the men use the points system in their new Elo version of the ranking system. They did away with the silly federation weighting. You get 3 points for a win, 2 points for a shootout win, 1 point for a shootout loss, ( so basically a group stage tie) and a goosegg for an outright loss. I haven’t found how the women do theirs. I just assumed it was reflected in the actual vs predicted result component of the formula.
No no no no no The men moved away from the 3pt/1pt/0pt system for calculating rankings several years ago and to an Elo-based system like the women had since the beginning Also, as @kolabear mentioned in a previous post, the way the ratings handled the 2011 shootout final was giving both teams the draw Sweden and Canada should both be credited with draws for today's result when it comes to rating calculations Wikipedia is a thing, y'all, and the editors there do a great job of keeping high-profile things like the FIFA ranking system very up-to-date and thorough, these things aren't hard to check.
It’s not just like the women’s system and it’s not exactly Elo. here is directly from the ranking procedure : those would be a shootout after a home and home qualifier where each team won a game, for instance. the weighting of results is the same as the points system.
True, it's not "just" like the women, I was editing my post while you were replying. However, saying "it's not exactly Elo" is wrong because there is no single "correct" Elo claculation, it's an entire class of rating systems. That said, you pulled the actual result values, and that's a good place to go for the women to answer the question. Nowhere in the women's actual results lookup table does it specify whether a match's actual result value changes based on PSO or match status. (There is the "match importance factor M", but that's a different quantity from the actual result itself.)
...and this is very much not correct. If you look at how those "W" values work in the new men's calculation system, they aren't "weightings". It's entirely different than the 2006-2018 3/1/0 system. The old system just awarded that number of points based on the W/D/L results before applying various multipliers. It was impossible to lose points for any given match; you only lost points when matches "phased out" of consideration. (Even the system used before 2006 made it impossible to earn negative points from a match because otherwise-negative scores got rounded up.) In the new men's system, an expected result "W_e" based on the difference in teams strengths gets subtracted from the actual result "W" - which I guess you could call the "win fraction" if you wanted to, to explain why they use that letter for the variable - so that you can gain or lose points from a single match, and how much depends on the difference in team strengths. It's a massively different equation and philosophy.
This was a discussion of how credit is awarded in a outright win or shootout win as opposed to a shootout loss or outright loss. The other multipliers are irrelevant to shootouts, or at least not any more relevant than if a shootout doesn’t occur. In that respect, under both systems, you get full credit for an outright win, no credit for an outright loss, credit for a tie if you lose a shootout, and credit halfway between a win and a tie for a shootout win. This is irrespective of any weighting for level of match, confederation multiplier, or relative rankings of the teams involved. A discussion of relative ranking, confederation rankings, match importance, etc. is irrelevant to the discussion, except that shootouts only generally happen in more important matches. in that respect they are identical regardless how you wish to express the maths. Whether you award 3,2,1 or 0 points or credit 1.0, .75, .5, or 0 weighting, the relative impact on the ranking is exactly the same. You get credit for a win, half win, tie, or loss in both cases. Your bringing up things like weak teams gaining points for a loss is just a red herring, and cause by other provisions in the old and new ranking systems, not anything to do with awarding ranking points in a shootout.
There's a lot of complaints about the FIFA rankings with the release of the new men's rankings. Although we generally only concern ourselves with the women's rankings here, because the method for the men was supposedly made more similar to the one for the women, I was curious about the widespread derision towards the new release. Some of the main complaints I heard: USA made it to #10 USA still below Mexico even though US beat Mexico twice in recent months Germany has fallen to #16 Winners of recent confederation championships wound up below teams they beat in winning those championships; eg US ranked below Mexico; Argentina below Brazil despite defeating Brazil to win the Copa America; Italy below England and Belgium despite defeating them to win the Euros (in the case of England, winning on PKs) How much of the whining (and I'll use the American whining instead of the English whinging since England failed to win the Euros) is the usual failure of fans to understand how an Elo/chess rating system is supposed to work? Or how much is a problem of the ranking system itself that fans of the women's game should be concerned about? First, rest assured that a lot of the fans whining simply don't understand how the system is supposed to work. They are heavily influenced by a presumption that anytime one teams beats another, the ranking should immediately reflect it by making that team higher ranked. However, there are major differences still between the men's and women's system which I think tend to make the men's system more susceptible to anomalies and flawed ratings. One of the most puzzling differences is that, in the men's system, there's no allowance for homefield advantage. So the US (men) just won the CONCACAF Gold Cup played on American soil and it's treated the same as if the games were played away or on a neutral pitch. Well, that's a huge boost in the ratings for the US. With the US women, when they play at home, the rating of their opponent is effectively lowered 100 points for calculating their rating. When they played Mexico at home a month ago, Mexico's rating of 1678 was treated the same as playing an opponent rated 1578 on neutral soil. Some other differences, whose benefits are more debatable, include weighting these confederation championships more heavily (4 times the typical friendly as opposed to 3 times in the women's ranking) and scoring a victory in a PK shootout as a half-victory (.75 points) instead of as an official tie in the women's system (with the exception of the championship match in Women's World Cup or Olympics, as we've been talking about) I'll probably add some of my thoughts about those later on
One of things I do is evaluate different rating systems in relationship to each other. Or, to be more accurate, I have developed a program that evaluates different rating systems in relation to each other. I do this only for NCAA Division I women’s soccer and the rating systems that are out there for it. But, I will express some thoughts about how one should evaluate the FIFA rating system, whether the one for the women or the one for the men. These thoughts should apply just as well to FIFA ratings as they do to the ratings for the DI women. First: A good rating system does not tell you what the results will be or even predict the results. Rather, it will tell you the likelihood of a team winning or losing a game. Thus if a better rated team, after making rating adjustments for home field advantage, loses a game it does not mean the rating system is bad -- the system contemplates that the better rated team sometimes will lose. Further, if a better rated team loses a game, the new post-game ratings will not necessarily have the winning team surpass the losing team in the ratings. It might, but in addition the losing team still might have a better rating but just not by as much. I believe it is especially important to understand this in relation to soccer rating systems, as to me it seems likely that soccer has significantly more results that are contrary to ratings than other sports. (For soccer overall, without regard to the amount of the rating difference, for the very best of the rating systems the better rated team wins only about 73 percent of the time.) Second: It is possible to tell whether a tweak to a basic Elo system improves it or makes it worse. This is done by comparing the system performance without the tweak to the system performance with the tweak. Again, it takes a lot of game results to do it to a high degree of precision, but it can be one. In this regard, one can look first at how well the basic system performs in terms of game results matching the rating differences of opponents and then compare this to how the tweaked system performs in terms of game results matching the rating differences. Strictly from a rating system performance perspective, tweaks should be used only if it can be demonstrated that their ratings match game results better than the ratings for the un-tweaked system. Third: Elo-based systems do about as good a job of rating teams as do other sophisticated systems. The variability of performance of the better systems, including Elo-based ones, is very small. Thus one should not expect to be able to find a rating system that will do significantly better than an Elo-based one. It might do slightly better, but the difference will be marginal.
More odd things on the men's ranking method. There's this on the Wikipedia page: ??? What does this mean?!! You can't lose points from a loss in the knockout rounds? (I suppose I should look at FIFA's own documentation but usually Wikipedia does a good job of summarizing and simplifying them, even correcting some of the errors in them) *** Let me go back to a couple of the other deviations the men's system makes from the women's. By increasing the "Match Importance" multiplier for World Cup and confederation championships and increasing the value of winning a PK shootouts (to 3/4 of a win in regulation instead of treating it as an official tie), what FIFA seems to be doing is trying to "reward" the winners of the World Cup, the Euros, the Copa America, and other confederation championships with higher ratings and rankings. It's like they're scared of the unhappy fans of teams who won the World Cup and see their team isn't ranked #1. "We won the World Cup! How can we not be ranked #1?" "We won the Euros! How can we be ranked lower than other teams who didn't win the Euros?" I dunno... Are we going to have an uprising of Canadians next week when they find out Canada isn't ranked #1 after winning the Olympics?! Had we better keep a lookout for the torches and pitchforks?! By (apparently) not counting losses in knockout stages against teams, FIFA also seems to be "rewarding" the teams which make it to a knockout round. In doing so, they're losing the point of what the ratings system is (you'd think) supposed to do, which is offer an imperfect but reasonable projection on how teams will perform in the future. How does "rewarding" a team for winning a PK shootout (with 3/4 point instead of 1/2 point) hep the ratings project the likelihood of future results? By winning a PK, is Team A now 3/4 likely to beat Team B in a future match as opposed to having even odds? Why have an Elo-type rating at all if you're going to bastardize it like this? Just say the World Cup champion will be the team most likely to win the next World Cup because that's how it works, right? Just say the winner of the Euros is the #1 candidate to win the next Euros because that's the way it works. *** I know many of you may not care that much about the men's tournaments and the men's rankings. I'm a much bigger fan of the women's game, too. But it may be helpful to be aware of how superior the women's rankings still are. Like if you're talking about the women's ranking and someone says, "FIFA rankings are garbage." The women's system can't produce a ranking like the US men at #10, something which doesn't pass the smell test. (The main exception being the possibility that someday a good team from Africa will be grossly underrated. But that's for reasons a rating system really can't be blamed for and for which there's no obvious logical solution)
About two more days of peace and calm before the Canadians begin rampaging once they find out Canada didn't become #1 after winning the Olympics Gold Medal...???!!!
This would only happen if @hotjam2 was Canadian! (And he wasn't currently banned from the "Women's International" boards).
The reward for winning a shootout reflect that a team can win shootouts, just as the reward for a win reflects that a team can win. It mostly reflects that a team can move on in a tournament, since that’s the only place shootouts happen. Teams that can’t move on aren’t as ‘ good’. some teams are better at it, you know. Look at the England track record. It’s actually exactly what a rating system should do.
England is the exception not the rule. Penalty shootouts take on the characteristic of a coin flip. Literally. As studies show, winning the coin flip statistically improves a team's chances of winning the shootout. Which is probably why FIFA's rankings have always scored a game which goes to PKs as an official tie for the purpose of the women's rankings before (with the exception of a championship match). And which is probably why they still score it that way for England — and all the other teams which lose a PK shootout. I'm putting this in the category of arguing for the sake of arguing.
There is a correct way to decide which side is right in this argument. Take the ratings with positive shootout results assigned a value within the ratings system and evaluate them for accuracy in relation to game results. Then, do the same thing for ratings that treat shootouts as ties. The one that is more accurate in relation to game results is the better rating system.
If penalty shootouts take on the characteristic of a coin flip, then the ranking will reflect a situation of a tie over time, won’t it?
https://www.fifa.com/fifa-world-ranking/women/news/fifa-women-ranking-august Buoyed by their Olympic gold, Canada (6, +2) are now sixth, while Germany (3, -1) have dropped a place to third and France (5, -2) two places to fifth. Netherlands (4) are non-movers at fourth. Korea DPR (9, +2) are back in the top ten, and Spain (10, +2) climb into it for the first time ever. Both sides have profited from falls for Japan (13, -3), who won just one match – against Chile (37) – at Tokyo 2020, and Australia (11, -2), whose quarter-final defeat of Great Britain was not taken into account when calculating the points.
I really wish Zambia had managed to beat China or draw Brazil; just needed one more goal in either match. They would've deserved the boost - It's criminal how underrated CAF sides are. I can't wait for the expanded World Cup playoffs. And how about how that final match was calculated?
I have not followed this thread, sorry, but I did just read on another board that North Korea moved from 11th to 9th in the FIFA ranking--despite not having played a match in 2 and a half years! Higher ranking than Spain and Australia, among others! Uh, oh: Something's gotta be wrong.
Well, if by "wrong" you mean that their inactivity should be temporarily removing them from the rankings altogether, then you'd mostly be right... It's insane that FIFA increased their inactivity window from 18mo to 4yrs recently. But in terms of teams moving up or down despite not playing between rankings in general: nah, that's normal, as long as teams near them gained or lost points to account for the shift. In this case, Australia and Japan lost a fair amount of points over the Olympics, letting N.Korea (and Spain) rise into the top 10 despite not playing since earlier this year.
A full four years cycle is mad for sure. Artificially keeping North Korea in the ranking is 100% a political move.