[Albion] Offside?

[Blues Guitarist]

View attachment 169244
Whilst I would've loved it to have been onside, it's clear that the line is drawn from Dunk's sleeve, as is Tarkowski's, the frame they chose is with Gross' foot touching the ball (at the moment the ball is played - as per law 11, Why would an image before the ball was touched matter at all?).

There is overwhelming evidence that it is offside. Look it was a great finish, and a joy to watch live, but for me, there are hundreds of more subjective issues to do with refereeing rather than rubbishing this one. It's just a shame because it's so tight and such a quality strike.

Taking all of this into account, why did it take 4 minutes (2 from when the VAR check screen went up in ground), to get to what is ultimately an objective decision, aided by technology?

I did ask for an image BEFORE the ball is kicked. Showing the wrong frame and trying to guess from there is wrong.

 

[Albion Andy]

In recent weeks I have become fascinated by posters who are suddenly making their first post YEARS after initially signing up to the forum.
Almost nine years @Albion Andy - welcome!
Just out of curiosity: why now? A lot has happened in nine years - why has a controversial decision in a run-of-the-mill match at Goodison Park suddenly caused you to hit that "Post reply" button?

Oh no particular reason, sometimes people just do stuff - no great mystery I'm afraid. Maybe I was more drunk than normal today.

 

[Uh_huh_him]

So Kyle Walker, at full speed, was clocked as running at 37kmph last season. That's about 10 metres per second. So at full pelt the furthest a player travelled in the Premier League in the whole of last season is 20cm in a VAR frame (50fps).

Firstly, we all know Dunk is considerably slower than Kyle Walker's top speed. Secondly, this isn't Dunk's top speed at all, he's probably, due to the context of the situation (trying to time his run, watching the flight of the ball etc.) going at considerably less than half of the top possible speed. Dunk and Tarkowski are both moving in the same direction, but even at wildly different speeds it is hard to imagine that between frames you'd be seeing anything close to a movement of more than 7cm between frames and the difference between the two would be considerably smaller.

Sorry to be such a nerd and pedant, but factually it is so unlikely that there is any discernible difference in Dunk and Tarkowski's position between the frame shown and the previous one.

You are missing the point.
Giving an offside position down to millimeteres, based on conjecture and not an absolute position, is where the issue lies.

We accept that the image that was measured, was not the precise point the incident occurred.
We also accept that Goodison does not have great camera positions.
The cameras are calibrated, but it's hard enough to pinpoint whether a ball is in or out of play from those sort of angles, let alone identifying the exact position of Tarkowski and Dunk's armpit!

There are multiple factors at play, that are not factored into the decision, which then delivers a black and white decision.
There is no tolerance allowed for the limitations inherent in the process.

You are saying, it's highly unlikely that the difference would have kept Dunk onside, but you don't know that.
The simple answer is nobody knows whether he was offside or not.

PGMOL need to step away from the idea that they can be certain about all offside calls.

 

[Hamilton]

Another question is, even if Dunk was technically offside going by the laws as they currently stand, should the laws be changed such that he technically wouldn't be?

I think they should.

Technically, yes.

I'd also argue that if VAR does not have a good enough angle to overrule an on-field decision then VAR has no business disallowing a goal. Michael Oliver arguably overstepped his remit. He should have said that he did not have an angle that would clearly suggest that the onfield ref was wrong.

Again, our officials are failing to communicate properly with one another and transparently.

They desperately need someone from the RFU to come in and start getting a hold of how we train top flight refs and how they communicate with one another. The fact that the head of the world body for refs said that he was against mic'ing up refs because the conversations people would hear were chaotic sent shivvers of worry through my spine. Get Nigel Owens in there!

 

[brighton_tom]

Not quite sure how you have reached that conclusion. Look at his kicking leg in relation to the front of the ball.

If his foot was still touching the ball, then his leg would have been shown blurred to the same degree as the ball.

It isn't, because presuambly his leg was moving slower than the ball - which is what usually happens when you kick a ball.

View attachment 169253

Doesn’t really matter too much where the ball is positioned along that smudge. The reason for the smudge is because it’s a still frame taken from a moving video image, so you’re seeing the distance the ball has moved from one frame of video to another blended into one still image. The cameras will most likely be running at 25 frames per second, so debating where the ball is along that smudge will only effect the offside call by how much Dunk can move within 1/25th of a second, which I imagine is not that much.

 

[Uh_huh_him]

Doesn’t really matter too much where the ball is positioned along that smudge. The reason for the smudge is because it’s a still frame taken from a moving video image, so you’re seeing the distance the ball has moved from one frame of video to another blended into one still image. The cameras will most likely be running at 25 frames per second, so debating where the ball is along that smudge will only effect the offside call by how much Dunk can move within 1/25th of a second, which I imagine is not that much.

We come back to the idea over and over again, between people who think the difference could be enough and those who think it isn't.
The reason the goal was ruled out is because PGMOL claim the technology delivers an absolute truth.

Clearly it doesn't.
These types of decisions are subjective.

The position of Lewis Dunk's armpit is subjective.
The position of Tarkowski's armpit is subjective
Where those armpits were positioned 1/25th of a second ago, is subjective.

But OFFSIDE IS OFFSIDE

 

[brighton_tom]

We come back to the idea over and over again, between people who think the difference could be enough and those who think it isn't.
The reason the goal was ruled out is because PGMOL claim the technology delivers an absolute truth.

Clearly it doesn't.
These types of decisions are subjective.

The position of Lewis Dunk's armpit is subjective.
The position of Tarkowski's armpit is subjective
Where those armpits were positioned 1/25th of a second ago, is subjective.

But OFFSIDE IS OFFSIDE

Too many variables for them to go into such microscopic analysis of an offside by inches, the technology just isn’t available to be THAT accurate.

 

[South Stand Bonfire]

I mean, even from that pixelated image, it's clear to me that both of those lines use the same point of the arm (the sleeve) as the point of reference for both players.

I respect your opinion, but please, in that image it is not clear! The ball is blurred and the line appears to have been taken to Dunk’s elbow/arms . it may be offside but if VAR is to overturn the on field decision, it needs to show a clear and obvious error, which it hasn’t done yet.

 

[Deleted member 37369]

We come back to the idea over and over again, between people who think the difference could be enough and those who think it isn't.
The reason the goal was ruled out is because PGMOL claim the technology delivers an absolute truth.

Clearly it doesn't.
These types of decisions are subjective.

The position of Lewis Dunk's armpit is subjective.
The position of Tarkowski's armpit is subjective
Where those armpits were positioned 1/25th of a second ago, is subjective.

But OFFSIDE IS OFFSIDE

Agreed ... but the other thing I'm not sure about with VAR is when the players are level. Level is onside - but as far as I'm aware, the two lines they draw can never be in the same position. So they draw the first line in blue ... and then these mm judgements have to fall one way on/off ... green/red.

It's almost 4 years since VAR was used in English football for the first time when VAR decided that Murray hadn't used his hand to score the winner against Palace in the FA Cup. Whereas some decisions will still be subjective, I can't believe we haven't got a better process to ensure offside decisions are correct. And the fact some of these offside decisions are taking 4 minutes or so is crazy. If they can't draw the lines to get a decision in 2 minutes (or less) just give the goal!

 

[Uh_huh_him]

Too many variables for them to go into such microscopic analysis of an offside by inches, the technology just isn’t available to be THAT accurate.

Which is why they need to move away from this idea that the lines are a precise way of measuring offside.

 

[Beanstalk]

You are missing the point.
Giving an offside position down to millimeteres, based on conjecture and not an absolute position, is where the issue lies.

We accept that the image that was measured, was not the precise point the incident occurred.
We also accept that Goodison does not have great camera positions.
The cameras are calibrated, but it's hard enough to pinpoint whether a ball is in or out of play from those sort of angles, let alone identifying the exact position of Tarkowski and Dunk's armpit!

There are multiple factors at play, that are not factored into the decision, which then delivers a black and white decision.
There is no tolerance allowed for the limitations inherent in the process.

You are saying, it's highly unlikely that the difference would have kept Dunk onside, but you don't know that.
The simple answer is nobody knows whether he was offside or not.

PGMOL need to step away from the idea that they can be certain about all offside calls.

I'm not missing the point, and I agree with most of what you're saying but all I'm saying is that there are two very different questions in conversation here. 1) Under the current rules, was Dunk offside? 2) Are the current rules fit for purpose?

In the current system we see the definitions given by IFAB as offside being a factual decision. Based on the technology that the Premier League clubs use they are "certain" that that was offside because there is currently no subjectivity in the offside law currently (except for interfering of course). I'm all for getting rid of the PGMOL, but it's the clubs who decided in this case that they didn't want the certainty with semi-automated offsides. What we're left with is a system that gives an answer based on lots of angles and technology that gives more certainty than an assistant referee can statistically give.

My pedantic comment was in response to someone saying that it was unlikely that in a fiftieth of a second Dunk and Tarkowski wouldn't have changed position. My point was, bizarrely, the same as yours. We can't know for sure, but mathematically it is very unlikely that in the frame before there is a big enough change for it to then be onside. That if we take the decision making process in good faith, and that the VAR chose the best frame available to them, then it is likely, regardless of which frame was chosen, the decision would've likely been the same, and we'd probably all have still disagreed with it.

There are a load of other questions as you and others have said that come from refereeing like this (what is level? when is the ball actually kicked? where does the playing part of the body start and end exactly? Is the current system's lack of flexibility against the spirit of the game?) but that is not the same question as asking if the referees came to the right decision based on the evidence available to them and how they've been asked to interpret and implement the current offside laws.

 

[Hotchilidog]

I respect your opinion, but please, in that image it is not clear! The ball is blurred and the line appears to have been taken to Dunk’s elbow/arms . it may be offside but if VAR is to overturn the on field decision, it needs to show a clear and obvious error, which it hasn’t done yet.

Absolutely spot on. The on field decision was goal. The technology did not conclusively show that this was an error, the goal should have stood.

In the NFL (where the refs equally awful) the replay officials will only change a decision when there is very clear evidence that there was a mistake, they set the bar very high to overturn a call. It does appear that the bar is set much lower in the EPL.

 

[Uh_huh_him]

I'm not missing the point, and I agree with most of what you're saying but all I'm saying is that there are two very different questions in conversation here. 1) Under the current rules, was Dunk offside? 2) Are the current rules fit for purpose?

In the current system we see the definitions given by IFAB as offside being a factual decision. Based on the technology that the Premier League clubs use they are "certain" that that was offside because there is currently no subjectivity in the offside law currently (except for interfering of course). I'm all for getting rid of the PGMOL, but it's the clubs who decided in this case that they didn't want the certainty with semi-automated offsides. What we're left with is a system that gives an answer based on lots of angles and technology that gives more certainty than an assistant referee can statistically give.

My pedantic comment was in response to someone saying that it was unlikely that in a fiftieth of a second Dunk and Tarkowski wouldn't have changed position. My point was, bizarrely, the same as yours. We can't know for sure, but mathematically it is very unlikely that in the frame before there is a big enough change for it to then be onside. That if we take the decision making process in good faith, and that the VAR chose the best frame available to them, then it is likely, regardless of which frame was chosen, the decision would've likely been the same, and we'd probably all have still disagreed with it.

There are a load of other questions as you and others have said that come from refereeing like this (what is level? when is the ball actually kicked? where does the playing part of the body start and end exactly? Is the current system's lack of flexibility against the spirit of the game?) but that is not the same question as asking if the referees came to the right decision based on the evidence available to them and how they've been asked to interpret and implement the current offside laws.

Fair enough.

Hopefully IFAB will have a proper review of the Laws of the game.
They weren't written with VAR in mind and they really need to be reassessed in light of the experiences of the past few seasons.

 

[Deleted member 37369]

I'm not missing the point, and I agree with most of what you're saying but all I'm saying is that there are two very different questions in conversation here. 1) Under the current rules, was Dunk offside? 2) Are the current rules fit for purpose?

In the current system we see the definitions given by IFAB as offside being a factual decision. Based on the technology that the Premier League clubs use they are "certain" that that was offside because there is currently no subjectivity in the offside law currently (except for interfering of course). I'm all for getting rid of the PGMOL, but it's the clubs who decided in this case that they didn't want the certainty with semi-automated offsides. What we're left with is a system that gives an answer based on lots of angles and technology that gives more certainty than an assistant referee can statistically give.

My pedantic comment was in response to someone saying that it was unlikely that in a fiftieth of a second Dunk and Tarkowski wouldn't have changed position. My point was, bizarrely, the same as yours. We can't know for sure, but mathematically it is very unlikely that in the frame before there is a big enough change for it to then be onside. That if we take the decision making process in good faith, and that the VAR chose the best frame available to them, then it is likely, regardless of which frame was chosen, the decision would've likely been the same, and we'd probably all have still disagreed with it.

There are a load of other questions as you and others have said that come from refereeing like this (what is level? when is the ball actually kicked? where does the playing part of the body start and end exactly? Is the current system's lack of flexibility against the spirit of the game?) but that is not the same question as asking if the referees came to the right decision based on the evidence available to them and how they've been asked to interpret and implement the current offside laws.

Good explanation And the bit I've bolded is quite key in this, I think.

 

[trueblue]

It's an aside but something I mentioned on a different thread. While refs are micro-managing offside decisions like this, you've got ludicrous situations like the first Nicolas Jackson goal last night. Ball is played through to a winger, who's onside. The central striker is miles off but disregarded because the ball doesn't go to him. However, he now has a 10 yard head start on all the defenders sprinting towards goal and is therefore able to take his shot in acres of space, completely unchallenged. That's EXACTLY what the offside law should be stopping - he's scored the bloody goal so it can't be irrelevant! Yet we're now bogged down in layers of complexity for every decision and endless talk of different phases of play, active and inactive, yadda, yadda, yadda....

 

[McTavish]

Doesn’t really matter too much where the ball is positioned along that smudge. The reason for the smudge is because it’s a still frame taken from a moving video image, so you’re seeing the distance the ball has moved from one frame of video to another blended into one still image. The cameras will most likely be running at 25 frames per second, so debating where the ball is along that smudge will only effect the offside call by how much Dunk can move within 1/25th of a second, which I imagine is not that much.

You would imagine wrong.

Dunk can probably sprint 100m in 12 seconds or so. That is 8m/s or 33cm every 1/25th of a second. So even if he is going at a fifth of his top speed he is moving over 6cm in 1/25th of a scecond. If the last defender is moving the other way at a similar leisurely fashion you are up to 13cm difference between the relative positions of the attacker and the defender.

When the offside decisions are decided by a couple of cms, 1/25th second is plenty of time for everything to change.

And that is before you decide whether the touch of the ball by the teammate of the offside player should be decided when he first makes contact with the ball or when it leaves his foot. Another couple of centimetres right there.

 

[brighton_tom]

You would imagine wrong.

Dunk can probably sprint 100m in 12 seconds or so. That is 8m/s or 33cm every 1/25th of a second. So even if he is going at a fifth of his top speed he is moving over 6cm in 1/25th of a scecond. If the last defender is moving the other way at a similar leisurely fashion you are up to 13cm difference between the relative positions of the attacker and the defender.

When the offside decisions are decided by a couple of cms, 1/25th second is plenty of time for everything to change.

And that is before you decide whether the touch of the ball by the teammate of the offside player should be decided when he first makes contact with the ball or when it leaves his foot. Another couple of centimetres right there.

You’re right, but the wider point is that 13cm is lots of movement IF they can accurately call offsides by such small amounts, but the technology simply isn’t available to do so. The fact they are calling these tight ones as offside without properly being able to do so is a farce. With the technology available they need a greater distance than 13cm between dunk & tarkowski to give an assured decision.

 

[SeagullsoverLondon]

I'm not missing the point, and I agree with most of what you're saying but all I'm saying is that there are two very different questions in conversation here. 1) Under the current rules, was Dunk offside? 2) Are the current rules fit for purpose?

In the current system we see the definitions given by IFAB as offside being a factual decision. Based on the technology that the Premier League clubs use they are "certain" that that was offside because there is currently no subjectivity in the offside law currently (except for interfering of course). I'm all for getting rid of the PGMOL, but it's the clubs who decided in this case that they didn't want the certainty with semi-automated offsides. What we're left with is a system that gives an answer based on lots of angles and technology that gives more certainty than an assistant referee can statistically give.

My pedantic comment was in response to someone saying that it was unlikely that in a fiftieth of a second Dunk and Tarkowski wouldn't have changed position. My point was, bizarrely, the same as yours. We can't know for sure, but mathematically it is very unlikely that in the frame before there is a big enough change for it to then be onside. That if we take the decision making process in good faith, and that the VAR chose the best frame available to them, then it is likely, regardless of which frame was chosen, the decision would've likely been the same, and we'd probably all have still disagreed with it.

There are a load of other questions as you and others have said that come from refereeing like this (what is level? when is the ball actually kicked? where does the playing part of the body start and end exactly? Is the current system's lack of flexibility against the spirit of the game?) but that is not the same question as asking if the referees came to the right decision based on the evidence available to them and how they've been asked to interpret and implement the current offside laws.

Agree with most of what you say. And of course that means that in any particular game and for each particular incident if you ran a proper statistical analysis to create the error range, you would have to have different thickness of lines to show on or offside for each incident.

That's why I would go back to a system where the lino flags if he/she thinks it's offside.
The technology takes a look, but uses thicker lines tan presently to allow for statistical error. If it is clear that the decision is wrong, overrule, if not stick with linesman's call.
It works ok in cricket and is accepted.

 

[Beanstalk]

Agree with most of what you say. And of course that means that in any particular game and for each particular incident if you ran a proper statistical analysis to create the error range, you would have to have different thickness of lines to show on or offside for each incident.

That's why I would go back to a system where the lino flags if he/she thinks it's offside.
The technology takes a look, but uses thicker lines tan presently to allow for statistical error. If it is clear that the decision is wrong, overrule, if not stick with linesman's call.
It works ok in cricket and is accepted.

Completely agree with point one, but fall the other way with the second paragraph.

Assistant referees are making unbelievably close calls, that are unbelievably difficult to get right in the moment. I think we should be leaning hard into semi-automated offsides. It is able to conclusively judge the moment that the ball is kicked, it is able to conclusively work out positions of players and deliver an offside decision within seconds, not minutes, and finally, it is able to create an image that clearly shows the crowd whether it was offside or onside. I understand that a lot of people want to get rid of VAR completely but that isn't going to happen - there are too many people and too much money involved to just scrap it. I cannot fathom why the clubs thought this wishy washy, neither here nor there approach was better than semi-automated.

 

[ac gull]

VAR was meant for "clear and obvious" errors

It's being used for totally different pedantic purposes - and just proving the human being in the VAR room is often as bad at getting decisions wrong as the ref on the pitch was always quite capable of - whilst the least well informed are those at the match