This seems like a more meaningless use of statistics than most. It completely ignores the quality of the defense. A team with a really good defense might be defending for most of the game, while the other team attacks and racks up a lot of points, 0.1 of a point at a time. The result would probably be either a 0-0 tie or a 1-0 win, both of which would statistically be unexpected.
2
With 27 shots, though only 7 on goal, I'm sure this metric favored Germany over England. For the first hour, most of the time, Germany WAS the better team. But the longer the game went, the more the quality shifted to England. And, of course, the metric, I'm sure, has no allowance for the three hand ball shouts in the area committed by Germany ( and one by England). Nor can the metric allow for a referee who consistently blows corner v goal kick calls all game.
3
1. Everybody already knows the better team (or player, in individual sports) doesn't always win. There's even a name for it: "upset."
2. This xG statistic sheds light on nothing - it seems invented only to give its proponents an opportunity to be surprised when the team with the higher xG doesn't win. And when that doesn't happen, it's called...see 1. above.
2. This xG statistic sheds light on nothing - it seems invented only to give its proponents an opportunity to be surprised when the team with the higher xG doesn't win. And when that doesn't happen, it's called...see 1. above.
3
Anyone who has watched even a few soccer games realizes that in the lowest scoring major sport, fluke goals; goals off glaring errors (giveaways or fouls in the box); and goals on brilliant individual plays determine the winner more than they would in other sports.
3
Doesn't this measure completely ignore the difference in skills of the two goalies? If so it seems rather useless
7
"Advanced Soccer Statistic Shows Better Team Doesn't Always Win."
Why is the NY Times publishing yet another "Dog Bites Man" story?
Soccer is quite simple. You score more goals than your opponent and guess what?:
You win the game!! Duh!
All the other stats are commentary!!
Why is the NY Times publishing yet another "Dog Bites Man" story?
Soccer is quite simple. You score more goals than your opponent and guess what?:
You win the game!! Duh!
All the other stats are commentary!!
4
Unfortunately, many of the Comments criticizing the xG stat are oblivious to this that Mr. Mather writes in the article:
"Analysts at Opta, a sports statistics service, studied thousands of shots and used logistic regression analysis to determine how likely each was to go in."
Logistic regression produces the stat based on many factors, such as just who is taking the shot.
Any statistical model is driven by probability, and so those who seek some sort of deterministic outcome from such are apparently mixing up stat with math.
"Analysts at Opta, a sports statistics service, studied thousands of shots and used logistic regression analysis to determine how likely each was to go in."
Logistic regression produces the stat based on many factors, such as just who is taking the shot.
Any statistical model is driven by probability, and so those who seek some sort of deterministic outcome from such are apparently mixing up stat with math.
2
Keep your statistics away from my beautiful game. Statistics are cold, heartless, for baseball and other such endeavors. We need them not.
4
This is a perfect example proving that LUCK is such an overwhelming factor in soccer marches and why soccer is ridiculously over rated as a legitimate sport. Soccer's success is built on hype more than anything else.
2
I can think of many examples in baseball (and hockey and basketball) where the "better team" has lost best of seven game series, not just a single knockout. If you don't like soccer, and don't respect it, that you take the time to comment on it marks you as a troll.
4
Actually, more skilled the players are, less room there is for luck. Otherwise, it would be quite puzzling why there are dominant players, teams, and nations in football. The fact that there is fortune involved even at the highest calibre of football only goes to show how overwhelmingly difficult it is to control a round rolling ball with your feet (certainly more difficult than, for instance, holding or throwing an oblong ball). The requisite skill seemingly requires early training: the students of the Barcelona youth academy, for example, are as young as six years old, I believe.
3
You know for what team this metric almost never works? How about Barcelona, who thoroughly dominate the ball while taking relatively few shots. Of course, four Champions League triumphs in the last decade, including being the only team to *ever* do the treble twice is a statistic better than goals or expected goals. For in soccer, goals on the pitch facilitate the greater goals, winning trophies.
Without knowing how the metric applied, I can point to a bunch of ?Barcelona matches where I'll bet the metric shows the better team did not prevail:
2006 CL Final, where numerical advantage does not translate into dominance
2008 CL semifinal, especially the home leg to Man United, parking the bus.
2009 CL semifinal leg at Chelsea, the infamous Tom Henning Ovrebo match, where .barca hung on, we're the beneficiaries of shoddy refereeing, and went through on Iniesta's late tying goal. Turnabout fair play as Chel$ki had parked the bus at Camp Nou in a nil draw that was totally one sided,
2010 CL semifinal leg in Milan. What is the metric for a blown offside or a PK not given when Sneijder took down Alves from behind?
2012 CL Semifinal, both legs. Chel$ki scored in three of four injury time periods and at no other time.
Without knowing how the metric applied, I can point to a bunch of ?Barcelona matches where I'll bet the metric shows the better team did not prevail:
2006 CL Final, where numerical advantage does not translate into dominance
2008 CL semifinal, especially the home leg to Man United, parking the bus.
2009 CL semifinal leg at Chelsea, the infamous Tom Henning Ovrebo match, where .barca hung on, we're the beneficiaries of shoddy refereeing, and went through on Iniesta's late tying goal. Turnabout fair play as Chel$ki had parked the bus at Camp Nou in a nil draw that was totally one sided,
2010 CL semifinal leg in Milan. What is the metric for a blown offside or a PK not given when Sneijder took down Alves from behind?
2012 CL Semifinal, both legs. Chel$ki scored in three of four injury time periods and at no other time.
1
Well, I don't know about "almost never" ... the Manita, Wembley 2011, Berlin 2015, those 2 half-years where Messi scored a total of 91 goals, routs over minnows and not so minnows. Maybe my memory's getting fuzzy but the impression I remember is that they were queueing up in strafing formation against Man Utd in 2011 before letting up. And that the 5-0 could have been worse but for a mercy call by the ref (not sending off Carvalho for a DOGSO handball)
I'll bet the round of 16 match a few years ago @ Arsenal, in the first leg, when Arsenal came back with two late goals to beat Barca on a late Arshavin goal will also show the "superior team" losing. Of course, the return leg didn't go so well...sigh.
1
Revenge of the Nerds.
There is a place for metrics as an adjunct to a sport but this article strays into the land of metrics inflating itself to the level of pompous ass when it asserts that England “deserved to win”.
No it didn’t. It lost and it deserved to lose.
I would not argue if the argument is that England dominated play and while that is interesting, it is also so what? The world of sports has always been littered with teams that dominated play BUT -- failed to capitalize, failed to seal the deal, failed to WIN.
You know, missed the object of the exercise.
And I also think our friends the statisticians have overlooked another vital point.
So okay, sure – some shots are better than others, but some players are better than others. So if you assign a value to position on the pitch from which the shot is struck, should you not also assign a value to the talent of the player who struck it?
Or do we say that Messi is just another guy and the three he put in the back of the net were interesting, but all from bad angles --so sorry the other fellas win, because they had more attempts from better positions even though none went in?
Do we say the basketball team with the best shot selection deserved to win? The golfer with the shortest total length of putts deserved to win?
No, we say, as Mr. Hickman of the ’62 Mets said all too frequently that “we snatched defeat from the jaws of victory”.
Yes, statistics are for losers. So study the tape England -- and find some finishers.
No it didn’t. It lost and it deserved to lose.
I would not argue if the argument is that England dominated play and while that is interesting, it is also so what? The world of sports has always been littered with teams that dominated play BUT -- failed to capitalize, failed to seal the deal, failed to WIN.
You know, missed the object of the exercise.
And I also think our friends the statisticians have overlooked another vital point.
So okay, sure – some shots are better than others, but some players are better than others. So if you assign a value to position on the pitch from which the shot is struck, should you not also assign a value to the talent of the player who struck it?
Or do we say that Messi is just another guy and the three he put in the back of the net were interesting, but all from bad angles --so sorry the other fellas win, because they had more attempts from better positions even though none went in?
Do we say the basketball team with the best shot selection deserved to win? The golfer with the shortest total length of putts deserved to win?
No, we say, as Mr. Hickman of the ’62 Mets said all too frequently that “we snatched defeat from the jaws of victory”.
Yes, statistics are for losers. So study the tape England -- and find some finishers.
2
@Rollo, I think your argument better applies to Germany beating France in the QF, when, refereeing aside, the French were conspicuously profligate in front of net (Necib in the first minute and Thiney in the 118th most consequentially).
On the other hand, England forced three tremendous saves out of the Japanese keeper who had done nothing earlier inthe tournament to suggest she had those saves in her. In the second half, England not only got into the right positions, their shots were dangerous and on target, not right down the middle. They were effectively beaten by the Japanese keeper.
On the other hand, England forced three tremendous saves out of the Japanese keeper who had done nothing earlier inthe tournament to suggest she had those saves in her. In the second half, England not only got into the right positions, their shots were dangerous and on target, not right down the middle. They were effectively beaten by the Japanese keeper.
1
Well, I'd argue that my argument applies equally well to each.
Stats are trailing indicators trying to quantify and explain the
at times unexplainable. They are not predictors of what must happen or who deserves to win.
My argument is that the score is by far the best and most beautiful expression of who deserves to win.
Obviously, the Japanese keeper had the saves in her so when England watched the tape maybe they just tipped the ole cap.
Wow, where did that come from? Or maybe, they found some things they could have done differently whilst still tipping the cap.
The mystery to me is this vague and strange notion of fairness, as if doing the most things right, dominating play, entitles you to come out on top. It doesn't work in life, so why should it work in sports?
That great baseball writer Roger Angell had it, I think, when he said -- there are not just two teams -- there are two teams and then there is the game. The game is the wildcard -- sometimes you do everything right and it comes out wrong. That's the game.
Or as Branch Rickey had it, you can be good, or you can be lucky and if you have to pick one, pick lucky,
Stats are trailing indicators trying to quantify and explain the
at times unexplainable. They are not predictors of what must happen or who deserves to win.
My argument is that the score is by far the best and most beautiful expression of who deserves to win.
Obviously, the Japanese keeper had the saves in her so when England watched the tape maybe they just tipped the ole cap.
Wow, where did that come from? Or maybe, they found some things they could have done differently whilst still tipping the cap.
The mystery to me is this vague and strange notion of fairness, as if doing the most things right, dominating play, entitles you to come out on top. It doesn't work in life, so why should it work in sports?
That great baseball writer Roger Angell had it, I think, when he said -- there are not just two teams -- there are two teams and then there is the game. The game is the wildcard -- sometimes you do everything right and it comes out wrong. That's the game.
Or as Branch Rickey had it, you can be good, or you can be lucky and if you have to pick one, pick lucky,
1
This headline was written either with tongue firmly in cheek, or by someone with absolutely no familiarity with the game and its history.
5
Thanks for the looking at it the math way.
However math does not take into account a forward breathing down your back inside the 18.
The British defender tried two sweep away the threat with her foot. An inch higher off the crossbar we would have been in into penalty kicks.
Math is a cold comfort.
However math does not take into account a forward breathing down your back inside the 18.
The British defender tried two sweep away the threat with her foot. An inch higher off the crossbar we would have been in into penalty kicks.
Math is a cold comfort.
3
@ G.P. — Well, “an inch higher off the crossbar,” and then we (likely) would have been into half an hour of extra time, before we’d know if spot kicks were necessary.
— Brian
— Brian
2
This article doesn’t clearly state whether xG stats have been shown generally — let’s say, over the course of a league season — to reflect pretty closely the goals teams actually score.
If so, then in matches where the results diverge from the xG stats, those stats may help us understand how an anomalous outcome came to pass.
On the other hand, if a long-term compilation of xG stats shows a consistent divergence of the majority of actual results from the results “expected” on the basis of the statistics, then that might help illuminate our understanding of how the game actually works — that is, what is •truly• important in producing a winning performance.
— Brian
If so, then in matches where the results diverge from the xG stats, those stats may help us understand how an anomalous outcome came to pass.
On the other hand, if a long-term compilation of xG stats shows a consistent divergence of the majority of actual results from the results “expected” on the basis of the statistics, then that might help illuminate our understanding of how the game actually works — that is, what is •truly• important in producing a winning performance.
— Brian
3
I agree about the Germany-France game. France was the better team. Have you seen the stats on the England-Canada game? England was NOT the better team. I saw it live and that seemed clear. On the own goal didn't anyone notice that had the ball not gone in and bounced back it was an easy tap in by Japanese strikers waiting and ready almost at the goal line. It's great that England as a nation got behind their team but it is clear that even commentators or "experts" are not nearly as "objective" as they believe they are. That's sports. There's just a lot of luck and luck isn't always with your favorite team.
2
Come on NYT, you can plug these metrics into any sport and get the same result.
The NYT spends energy on pieces lile this and sending Jere Longmann to the Canadian arctic to do a story on soccer, a good piece by the way, yet the NYT completely snubs Copa America.
I am econsidering my subscription very carefully. Almost $40/month for my package and this is the soccer coverage the NYT delivers. Disappointing.
The NYT spends energy on pieces lile this and sending Jere Longmann to the Canadian arctic to do a story on soccer, a good piece by the way, yet the NYT completely snubs Copa America.
I am econsidering my subscription very carefully. Almost $40/month for my package and this is the soccer coverage the NYT delivers. Disappointing.
3
The NYT sent a reporter to Chile for two weeks. He wrote about five (excellent) pieces from the Copa America. So it is incorrect -- and grossly unfair -- to say the NYT "completely snubs" the Copa America.<br/><br/>Your threatening to cancel your subscription in every comment does not make your charge accurate.
Really? Send me the links.
I remember one feature about Santiago stadium and one about the Mapuches.
Where's the indepth coverage on matches. It does not exist.
Most of them have been blurbs from Reuters or AP.
You do know the difference bewteen wire stories on games and reports on matches, right?
The NYT has been remiss in it's coverage of Copa matches.
Anyone can see that.
I remember one feature about Santiago stadium and one about the Mapuches.
Where's the indepth coverage on matches. It does not exist.
Most of them have been blurbs from Reuters or AP.
You do know the difference bewteen wire stories on games and reports on matches, right?
The NYT has been remiss in it's coverage of Copa matches.
Anyone can see that.
2
Yet another sideshow curiosity of a statistical metric. I do like this statement, though:
" is actually a better measure of a team’s performance than goals because goals can be subject to so many random factors."
And systematic factors. In this Cup, that would be uniformly abysmal refereeing. 3 examples:
France-Germany: the imaginary handball called against France but not against Germany. France left their shooting boots at home and wasted a hatful of chances, but even so, without that gift PK Germany were going home. As it was the better team was sent home.
In the US-Germany SF Johnston should have been sent off for clear DOGSO and the Morgan foul was never a PK. The better team won -- but not clear they would be the better team playing a man down.
In England-Japan neither PK was warranted, though it's arguable the England one was a makeup. England had the better chances, but they also committed one fatal mental error. If you're going to score xG you should also score cB against it (critical blunders, in this case switching off before the whistle sounded).
Way too many instances. Hoping for a cleanly reffed Final, but hardly expecting it.
" is actually a better measure of a team’s performance than goals because goals can be subject to so many random factors."
And systematic factors. In this Cup, that would be uniformly abysmal refereeing. 3 examples:
France-Germany: the imaginary handball called against France but not against Germany. France left their shooting boots at home and wasted a hatful of chances, but even so, without that gift PK Germany were going home. As it was the better team was sent home.
In the US-Germany SF Johnston should have been sent off for clear DOGSO and the Morgan foul was never a PK. The better team won -- but not clear they would be the better team playing a man down.
In England-Japan neither PK was warranted, though it's arguable the England one was a makeup. England had the better chances, but they also committed one fatal mental error. If you're going to score xG you should also score cB against it (critical blunders, in this case switching off before the whistle sounded).
Way too many instances. Hoping for a cleanly reffed Final, but hardly expecting it.
4
The Morgan foul was absolutely a PK. FIFA Questions and Answers on Laws of the Game state that a foul begun outside the penalty area but continued into the penalty area result in a PK. And the foul clearly continued into the penalty area.
Unfortunately, none of the soccer analysts that I heard pointed this out and it has become accepted truth in the US that if a foul is begun outside the penalty area then it is does not result in a PK. But this is simply not true.
I am not saying that mistakes weren't made but that was not one of them.
Unfortunately, none of the soccer analysts that I heard pointed this out and it has become accepted truth in the US that if a foul is begun outside the penalty area then it is does not result in a PK. But this is simply not true.
I am not saying that mistakes weren't made but that was not one of them.
4
Expat: the Faux Sports refereeing expert, doddering Dr. Joe Machnick, clearly stated that the only foul that can start outside the box and continue inside for a penalty is holding, which was not the nature of Krahn's foul, which was obstruction. And just because Morgan went airborne into the area does NOT mean that the foul continued into the box.
There is no metric for refereeing error. I would only be interested in this metric if I saw it applied to every match of the tournament. Other games where the metric doesn't necessarily reflect the result? USWNT-Australia, Australia-Brazil and Australia-Japan, just to name three. And to give but two results, one where the metric fit the result, and the other where it didn't tells one nothing at all.
There is no metric for refereeing error. I would only be interested in this metric if I saw it applied to every match of the tournament. Other games where the metric doesn't necessarily reflect the result? USWNT-Australia, Australia-Brazil and Australia-Japan, just to name three. And to give but two results, one where the metric fit the result, and the other where it didn't tells one nothing at all.
3
Machnick got several things wrong but he's right about this. A foul may start outside the box but it has to be a foul within the box to be a penalty. There's no rule against contact in the box, no rule about being airborne. The foul was obstruction and that took place outside box. Not a penalty.
1
If one believes in statistical analysis in soccer, which I don't, it would be interesting to look at France-Germany. I'd that the better team lost that one, too. One doesn't need a statistic to prove what to a longtime fan is empirically obvious.
Victor, for kicks and giggles, try running some Champions League matches through this analysis. Start with, say, the 2012 Champions League Final...
Victor, for kicks and giggles, try running some Champions League matches through this analysis. Start with, say, the 2012 Champions League Final...
2
Not sure how xG is calculated, exactly, but if it includes kicking lumps out of the other team Chelsea might still be deserving winners on that metric ...
2