Why do fans complain about crosses?

[moly99]

It's been nearly a year since David Moyes left Manchester United and I still have to listen to their fans complain about his crossing-heavy strategy. I can understand why they dislike hoofball (long passes up the field) because they are very low percentage passes: it's the soccer equivalent of a team taking long 2 point shots in basketball. But crosses actually seem to lead to quite a lot of goals if the team has a decent striker to punch them in the net and they often lead to favorable second chances if the cross isn't immediately successful.
 
So . . . why do fans dislike crossing?

 

[Nick Kaufman]

It's been nearly a year since David Moyes left Manchester United and I still have to listen to their fans complain about his crossing-heavy strategy. I can understand why they dislike hoofball (long passes up the field) because they are very low percentage passes: it's the soccer equivalent of a team taking long 2 point shots in basketball. But crosses actually seem to lead to quite a lot of goals if the team has a decent striker to punch them in the net and they often lead to favorable second chances if the cross isn't immediately successful.
 
So . . . why do fans dislike crossing?

I like crosses, but I don't think they are as efficient as you do. It s the same with corners. I always thought corners presented a good chance, but it s more like one in a hundred that lead to a goal. I be also noticed that modern possession oriented sides like Barcelona avoid both crosses and long shots away from the goal. I think they be looked at he numbers and come to the conclusion, that the best way to score is to keep possession until you crate a head to head with the goalkeeper.

 

[DJnVa]

Crossing is inefficient: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2225728
 

However, crossing from an open play is hugely inefficient, only 1 open cross in 4.87 is an accurate cross and only 1 open cross out of 91.47 leads to a goal on average. Statistical evidence indeed confirms that games with smaller number of open crosses tend to lead to more goals. When we estimate the impact of open crossing on scoring of the individual teams, we conclude that the net effect of crossing is negative for all the teams. This is also true for the German Bundesliga and the recent World Cup 2014. The quality of the attacking team is the major explanatory factor on the number of such missed scoring opportunities, stronger teams are more negatively influenced than weaker teams. Stronger teams have more options how to score and open play crossing seems as one of the suboptimal ways of a goal creation. Interestingly, the actual conversion of open crosses to goals plays only a minor role for explaining the impact of open crossing on goals
 

 

[moly99]

Thanks. That's exactly what I was looking for.

 

[Warning Track Speed]

Furthering the point on corners, The Numbers Game does a nice job debunking their effectiveness. I loaned the book out so can't cite at the moment, but long-derided short corners are more effective. I thought the book was an interesting read.

 

[Infield Infidel]

I've seen some criticism about The Numbers Game stuff about corner kicks http://www.7amkickoff.com/2013/of-course-corners-matter-why-the-numbers-game-got-it-wrong/
 

The stats here are simple. Last season (2012-13) in the Premier League all 20 teams took 10559 shots and there were 1063 goals scored. That doesn’t exclude penalties (there were 68 goals scored on penalties), corners (126 goals), throw ins (7), or dead ball situations (62). Just simply dividing the number of shots taken by the number of goals scored gives us a conversion rate of almost exactly 10%. Just taking out penalties and corners and the conversion rate drops to 9% last season. Meanwhile, taking the total number of shots from corners (805) and the number of goals scored from those shots we see that corners are converted at a 15.65% rate.
 

Sally and Anderson are correct, if you look at the total number of corners taken there is little correlation between that number and goal difference. But that’s because not all corners are equal. A team down one goal in the 90th minute is going to try to get a shot on goal from any corners that they win, even Barcelona and Spain will try that. A team up a goal in the 90th minute is going to take a short corner and try to run the clock out. I would go even further and say that shots from corners are only ever taken when the attacking team needs a goal: either in the case of a tie game that you are trying to win or if the attacking team is behind.

 
I think that last sentence overstates things a little, teams usually go for goals from corners early even if they have a lead, but later in the game they aren't going to take that chance nearly as much

 

[DLew On Roids]

Yeah, I'd say that in the Premier League, the percentage of corners that don't directly attempt to create a chance is so low that it can be assumed away.

At any rate, a 15.6% conversion rate on shots from corners ignores the first half of the process--getting a shot from the corner.

Edit: I posted above before ever reading the 7am Kickoff post, and now that I have, I'm taking it with even more of a grain of salt. A thorough analysis would have listed conversion rates of corners into shots and conversion rates of shots into shots on target and goals. That post just cherry-picks stats into a hot take--Chelsea scores at this rate on shots from corners, Man U converts corners into shots at this rate, Man U scores a lot of their goals from corners, QED MOFOS! Blech. I'm skeptical that the net benefit of corners is zero, but that wasn't the counter-argument.

 

[SoxFanInCali]

Personally, as a Liverpool fan, I hate crosses because Stewart Downing.

 

[Infield Infidel]

Yeah, I'd say that in the Premier League, the percentage of corners that don't directly attempt to create a chance is so low that it can be assumed away.

At any rate, a 15.6% conversion rate on shots from corners ignores the first half of the process--getting a shot from the corner.

Edit: I posted above before ever reading the 7am Kickoff post, and now that I have, I'm taking it with even more of a grain of salt. A thorough analysis would have listed conversion rates of corners into shots and conversion rates of shots into shots on target and goals. That post just cherry-picks stats into a hot take--Chelsea scores at this rate on shots from corners, Man U converts corners into shots at this rate, Man U scores a lot of their goals from corners, QED MOFOS! Blech. I'm skeptical that the net benefit of corners is zero, but that wasn't the counter-argument.

 
First, from soccerbynumbers in 2011 20.5% of corners resulted his shots within three touches, and there were 5.52 corners per game. And just seeing goals/shot go down in 2012/13 after taking out penalties and corners, there has to be value there.
 
I think you have to take this team by team, just saying "on average" doesn't mean as much because some teams are much better than others. It depends on how good you and your opponent are. Like the graphs from soccerbynumbers here http://www.soccerbythenumbers.com/2011/05/why-goal-value-of-corners-is-almost.html . (The numbers are from a different season)

 

 
What I'm seeing is that the better sides had much fewer shots per corner. My guess is two reasons: these better teams are better in open play than off corners, and late in matches they have the lead (there were only 5.52 total team-corners per game, so one late in a match will skew the numbers a bit). Whereas the lesser sides aren't as good in open play so shots from corners are a better avenue to get goals, and act as an equalizer of sorts. So if you're an elite team, don't go for shots from corners against crappier sides, but if you are a lesser side send it in, since there's a higher chance of scoring off a corner than straight up. 

 

[Infield Infidel]

I guess my gripe isn't in their research, it's in how they characterize it. Instead of saying "More corners negligibly correlates to more goals" or more simply, "More corners doesn't mean more goals" they say "The Goal Value of Corners: Zero". I think that's misleading. They are measuring correlation, but giving the result in aggregate. When I see "Goal Value" I'm thinking of stats that take discrete things, give them a value, and then total them up. Goal Value sounds like they gave a value to all kinds of stuff in soccer and found that corners weren't worth anything. That's not at all what they did. What they should do is find the value of all passes in the final third and compare them to corners

 

[DLew On Roids]

Maybe the first goal of Inside the Corner Flag could be to establish a comprehensive match modeling system.

You know, low-hanging fruit.