Shot location - are we doing it wrong?

[andersj]

American sports are probably far ahead of football in terms of analytics and big data. The picture above illustrates how big data has changed how NBA-teams approach the most important part of the game; scoring points.

When I first saw it I was wondering how scientic Premier League teams are when they approach where they want to finish/take shots. Is there a lesson to be learned here for football teams?

As most of our fans will aknowledge, Man Utd have been quite trigger happy this season. I was still surprised to see how trigger happy we have been compared to other teams. Peter McKeever on Twitter had made a few brilliant illustrations.

He has made one illustration where he compares Man Utd and Liverpools median shot distance and percentage of long range shot each season since 2008.

Tweet

— Twitter API (@user) date

In general, you have to look at it a couple of times as it is moving quite fast. But the trend is clear. We take a bigger share of our shots from a distance, while Liverpool has moved in the other direction.

That should not come as a big surprise. A better team will be able to create better scoring chances, and better scoring chances will be closer to goal. We have lacked creativity and had difficulty breaking down lower block teams. Consequently, we take shots from a greater distance and have bigger share of long distance shots than Man Utd anno 08-13 and Liverpool today.

But how do we compare to other PL-teams? And this came as a suprise to me.

Tweet

— Twitter API (@user) date

The illustration speaks for itself.

It looks like a deliberate strategy to shot frequently from a distance. And it pose an important question; do we shot frequently from a longer distance because we cant break down teams or do we struggle to break down teams because we prefer shooting from a longer distance? And if taking shots from a distance is a deliberate strategy, is it a good one?

Even if we have a few players who are very good at shooting from outside the box (Rashford, Pogba and Bruno) I dont think this approach makes much sense. Our players need in game practise «playing» in the final third to become better at breaking down a lower block. We will not get that if we shot to frequently. It will be interesting to see how this develops next season and as we get better players.

 

[Pexbo]

Doesn’t Steph Curry take a huge amount of credit for this? Also it’s the same amount of goals inside or outside the box so there’s no value in focusing on one area or the other.

 

[Samid]

That's a cool chart. Curry and GSW revolutionised the game.

 

[andersj]

Doesn’t Steph Curry take a huge amount of credit for this? Also it’s the same amount of goals inside or outside the box so there’s no value in focusing on one area or the other.

I’m sorry, what?

 

[Web of Bissaka]

Interesting thread.

Tweet

— Twitter API (@user) date

The illustration speaks for itself.

Top 3 this season are among the 5 lowest i.e. closest distance shooter teams.

 

[AltiUn]

I think having someone be able to pop one in from distance every now and again isn't necessarily a bad thing, but if you becomne reliant on that then I think it shows issues with the system because you're always going to be much less likely to score from random pot shots. It's telling that the top 3 are all near the bottom of the long shot table, it shows they've got systems capable of crerating consistent chances, something we've lacked for a long time. It's why I've had renewed optimism of late because we seem to be creating chances much more easily and seem to be able to work it into the box better than we used to.

In our defence though, with Pogba being out all season we've not had anyone in midfield with enough quality on the ball to help us work it into the box, just players like Lingard and Pereira (who I suspect is one of the biggest long shot culprits alongside Fred). It wasn't until Fernandes came in that we finally had a bit of quality in CAM.

 

[hungrywing]

Doesn’t Steph Curry take a huge amount of credit for this? Also it’s the same amount of goals inside or outside the box so there’s no value in focusing on one area or the other.

No. It's actually kind of the other way around.

Long story short, data showed it was better to take three point shots. (If you take one hundred shots per game and shoot 40% from three and 50% from two, you get 120 points vs 100 points) Around ten-ish years ago NBA teams started to be aware of it and it took varying amounts of time for them to embrace it due to the long-held idea that the three was a 'weaker/worse' shot attempt.

 

[Pexbo]

I’m sorry, what?

The first graphic is showing how NBA has changed and my post was focusing on that. I've not followed NBA closely for years and years but as far as I am aware Steph Curry came along and started hitting 3's like no tomorrow and other teams followed suit recognising just how profitable it is.

My point is that it's a daft comparison to football where it doesn't matter if you hit a 40 yard pinger or bundle it in of your arse from 3 yards, you still register one goal for your side. The last graphic is absolutely pointless too as there is no correlation between the chart and positions in the table.

 

[buckooo1978]

I think it's obvious why we shoot from distance

ineffective passing and movement in attack, with Lingard or Pereira in the creative hub limited our chance creation. Our most incisive midfielder Pogba was out all season and McTominay/Matic don't put up close to the same numbers in terms of key passes

the contrast was there to see with the arrival of Bruno - I'd like to see a compare and contrast in terms of data - pre-Bruno and Bruno - he does love a shot from distance also

 

[Grib]

American sports are probably far ahead of football in terms of analytics and big data. The picture above illustrates how big data has changed how NBA-teams approach the most important part of the game; scoring points.

When I first saw it I was wondering how scientic Premier League teams are when they approach where they want to finish/take shots. Is there a lesson to be learned here for football teams?

As most of our fans will aknowledge, Man Utd have been quite trigger happy this season. I was still surprised to see how trigger happy we have been compared to other teams. Peter McKeever on Twitter had made a few brilliant illustrations.

He has made one illustration where he compares Man Utd and Liverpools median shot distance and percentage of long range shot each season since 2008.

Tweet

— Twitter API (@user) date

In general, you have to look at it a couple of times as it is moving quite fast. But the trend is clear. We take a bigger share of our shots from a distance, while Liverpool has moved in the other direction.

That should not come as a big surprise. A better team will be able to create better scoring chances, and better scoring chances will be closer to goal. We have lacked creativity and had difficulty breaking down lower block teams. Consequently, we take shots from a greater distance and have bigger share of long distance shots than Man Utd anno 08-13 and Liverpool today.

But how do we compare to other PL-teams? And this came as a suprise to me.

Tweet

— Twitter API (@user) date

The illustration speaks for itself.

It looks like a deliberate strategy to shot frequently from a distance. And it pose an important question; do we shot frequently from a longer distance because we cant break down teams or do we struggle to break down teams because we prefer shooting from a longer distance? And if taking shots from a distance is a deliberate strategy, is it a good one?

Even if we have a few players who are very good at shooting from outside the box (Rashford, Pogba and Bruno) I dont think this approach makes much sense. Our players need in game practise «playing» in the final third to become better at breaking down a lower block. We will not get that if we shot to frequently. It will be interesting to see how this develops next season and as we get better players.

That nba chart makes me sad

 

[VorZakone]

That nba chart makes me sad

Why?

 

[Grib]

Why?

Knew the game changed, but not to that extreme

It's like a meta in a video game now, if you don't use this this and this with this then you can't really win

 

[calodo2003]

That nba chart makes me sad

So true, very little nuance in scoring in the game anymore, but I would never have expected to see it in this extreme.

 

[paraguayo]

NBA tended towards long shots because they are... worth more points.

There is not parallel to that in football, every distance goal is worth the same as far as I can tell.

 

[andersj]

The first graphic is showing how NBA has changed and my post was focusing on that. I've not followed NBA closely for years and years but as far as I am aware Steph Curry came along and started hitting 3's like no tomorrow and other teams followed suit recognising just how profitable it is.

My point is that it's a daft comparison to football where it doesn't matter if you hit a 40 yard pinger or bundle it in of your arse from 3 yards, you still register one goal for your side. The last graphic is absolutely pointless too as there is no correlation between the chart and positions in the table.

But that is not the only point of the graph. It is also where they shoot from to get three or two points, not only that they shoot more to get three points. If they shoot to get two points, they do it from the right angle.

 

[Pagh Wraith]

StatsBomb are doing a lot of great work in that regard and a lot of professional teams are cooperating with them already. As has been said, football is really a bit behind other sports when it comes to data analysis which means there is lot of room for improvement in a lot of aspects of the game. And I suspect set-pieces is going to be a huge one. Watching even Premier League sides waste free kick opportunities is tragic at times.

They are currently offering an Introduction to Football Analysis course on their website for £50 which I'm definitely going to take in the coming days if I find the time.

 

[Chipper]

Saw some early thoughts on this topic a few years back from one of the football stats/analytics blogs. Can't remember which one, think one of the people had worked with NASA if that means anything to anyone. It's not really much different from xg really I suppose.

PS I don't watch basketball but that graphic makes me sad too. Always thought it was boring and too samey, and most of the basketball I've ever watched was before that change. Now it looks even more uniform.

 

[Nick7]

Knew the game changed, but not to that extreme

It's like a meta in a video game now, if you don't use this this and this with this then you can't really win

To be fair, isn't that the point of management in competitive sports? why take a lower % shot when you can take a higher % shot? why take a long two when stepping back and making a shot of similar difficulty can get you an extra point.

 

[RedCurry]

It’s interesting analysis. This might offer some explanation why we didn’t face major challenge from likes of Man City for Bruno.

While I don’t want us to stop taking long range shots altogether, I want us to move towards taking only higher probability long range shots. For example, shooting from close range but a very tight angle might be much lower probability than bang center from outside of the box. Then you have clearly better long range shooters in the game and we should normalize the stats for those players too.

As far as United are concerned, we have been too trigger happy even when the best option would have been to make that extra pass or take an extra touch.

 

[Kasper]

The fact that Newcastle had registered shots at 50+ meters made me laugh

Interesting stats for sure. Obviously doesn't say much in isolation but interesting to see how City and Liverpool are focusing at shots no further out than ~22 meters.

 

[GifLord]

Didn't they change some rules in the NBA in the late 00s to make up for the lack of low scoring matches?

 

[NinjaZombie]

The fact that Newcastle had registered shots at 50+ meters made me laugh

Interesting stats for sure. Obviously doesn't say much in isolation but interesting to see how City and Liverpool are focusing at shots no further out than ~22 meters.

Liverpool and City are better than us at delivering crosses into the box, better at getting into the box with quick incisive passing and are better than us at getting to the byline and cutting balls in, resulting in them getting shots nearer to goal.

I've been watching a lot of our old games and the most glaring difference between us back then and now is our full backs were so much better than our current ones at the attacking aspects of the game. Rafael, Evra, Gaz, Silvestre and Irwin knew what they were doing in the last third of the pitch.

 

[AR87]

The first graphic is showing how NBA has changed and my post was focusing on that. I've not followed NBA closely for years and years but as far as I am aware Steph Curry came along and started hitting 3's like no tomorrow and other teams followed suit recognising just how profitable it is.

Team were trending that way well before GSW/Curry, but their extreme success definitely confirmed the value of such an extreme approach, although funny enough they took a ton of mid-range shots over the last couple of years.

 

[Snafu17]

I mean the closer you're able to get a clean shoot at goal the easier is to score, obviously. And you can do that by being creative, precise, having good movement and ability to stretch teams ie being all around competent in the final third. The stats pretty much only show that Liverpool has been better than United offensively.

It's a bit different with basketball where a shoot has a different value depending on the distance.

 

[Wednesday at Stoke]

The NBA shot locations are directly comparable because the pace (number of possessions) is relatively uniform and the shot clock mandates that each team has roughly the same number of shooting opportunities, if you account for any rebounding differential. In football, this is not true and possession is never uniform even from game to game. So its kind of naive to compare shooting distances without normalizing for possession etc.

 

[Gasolin]

That's a cool chart. Curry and GSW revolutionised the game.

In my personal opinion, they made the game boring with those 3 pts, no more attempt to get into the racket, but yeah, point wise, sure.

 

[Davìd Moyéz]

Wouldn't mind that chart if Beckham and Scholes were playing but I'd be interested to know how many of Fred's long range attempts are on target...

 

[Gasolin]

I’m sorry, what?

In football. In basket ball, shooting outside of the racket may add you 1 pt, but in football, it really is the same goal.

 

[hmchan]

I mean the closer you're able to get a clean shoot at goal the easier is to score, obviously. And you can do that by being creative, precise, having good movement and ability to stretch teams ie being all around competent in the final third. The stats pretty much only show that Liverpool has been better than United offensively.

It's a bit different with basketball where a shoot has a different value depending on the distance.

Stats are sometimes misleading. It's dangerous to jump to conclusion based on one or two parameters. In terms of xG, Man City and Chelsea are well ahead of us. I agree with the rest of your post though.

 

[harms]

Well, the infamous xG is basically about the same thing — different zones have different goalscoring coefficient, and we know that it's a statistic that coaches actually use today (although, like with every statistic, it's only used as an addition to an actual analysis and not as a basis for it, it's the line that regular fans often overstep when talking about xG and similar stats). Pep has been installing the same ideas since his very first season at Barca — try to create the best goalscoring opportunity instead of taking a shot from a far, which has a very little success rate. It's definitely something that top teams today try to implement, although if you have someone like peak Cristiano or peak Bale, it may be better to let them shoot, as truly exceptional players can prove general stats wrong. Funnily enough, I think it's actually Messi, (and not Cristiano or any other long distance specialist), who has the best success rate from his outside the box shots over the last couple of years, even though many of those were free kicks.

By the way, City's heatmaps also often look like artificial symmetrical objects (on the left is the one against Chelsea in 2018):

Thankfully, football has more variables than basketball, it's fascinating to read about tactical changes in basketball but when the game is basically gets hacked more and more, it takes away the magic.

 

[UnrelatedPsuedo]

The best teams get closer to goal more often. We are not very good. We don’t create that many chances inside the box.