PGMOL go it alone with pens?

[one]

The overall argument is that you don't want to give a goal or a goal scoring opportunity like a pen (and red cards too) from anything that is not clear cut. Careless is not black and white. It has a grey area and in above mentioned situations it has to be very close to clear.

It may seem like inconsistency when defensive free kicks have lower thresholds than attacking free kicks in a penalty area but it is actually consistent in applying the above concept.

An attacker strips the ball off a defender in a 'grey area' foul in the PA creating a GSO you very likely give a foul. A defender wins the ball in a similar way, you play on. This is consistent on both ends. Games should not be lost or won on uncertainty. That's the idea anyway.

It's not directly about what area of the field it happens, it's about the impact of the outcome of an uncertain decision which happens to be bigger when close to goal or in the PA.

 

[Anubis]

I dont envy the referee needing to explain after a 0-0 or 1 goal defeat that although there was a foul for a penalty, it was not deemed enough of a foul to award the pen.
Note, foul, not, contact.

Again One is perfect with the explaination above, consistency is awarding the same defensive foul to either side.

 

[RustyRef]

If I think there has been a push in the middle of the pitch I'm not too worried about giving it if I'm only 60 or 70% certain. Would I give it as a free kick if it was on the edge of the area, or a penalty if it was in it, with that level of certainty? No, not a chance and for potential game changing decisions I want to be much more certain, and I won't believe any referee who says they do otherwise.

There have been loads of complaints on this site about players buying penalties, yet when the authorities try to do something about it they get criticised. I don't get it.

 

[Big Cat]

If I think there has been a push in the middle of the pitch I'm not too worried about giving it if I'm only 60 or 70% certain. Would I give it as a free kick if it was on the edge of the area, or a penalty if it was in it, with that level of certainty? No, not a chance and for potential game changing decisions I want to be much more certain, and I won't believe any referee who says they do otherwise.

There have been loads of complaints on this site about players buying penalties, yet when the authorities try to do something about it they get criticised. I don't get it.

If the Refs weren't all working to the same set of turgid instructions in the first place, there wouldn't be anything to fix
I'm of a mind that it's not the EPL Refs that are collectively crap, it's the shared mandates they work to that are at fault. I don't believe we get to see the Referees' natural ability, because they're all clamouring over one another to be the same
So they'll all been giving soft pens for that reason... probably
Or they might not... but they'll all do the same, whatever that is

 

[Paul_10]

One thing that gets bought up is that the Raheem Sterling penelty won't be give in the PL because of this higher threshold but how do we know that for sure, and if a ref does give it, is VAR going to intervene because its too soft for a penelty to be given? It's very easy to say we won't give something that looks soft on countless replays but in the heat of the moment, judging it on live play and if it's a home team player having 10s of thousands of supporters screaming for a pen, I bet we will see pens like the Sterling one being given. It's the ones where a player forces the contact that needs to be stamped out of the game and that is where imo VAR should intervene.

I do worry the bar is going to be set too high and we are heading back where we were a few years ago where a subjective decision is not going to be overturned. Yes the amount of pens did increase last season but it was nothing too significant, it was not like the numbers we saw in Serie A in the season before last where referees were given every handball a penelty virtually.

 

[Big Cat]

One thing that gets bought up is that the Raheem Sterling penelty won't be give in the PL because of this higher threshold but how do we know that for sure, and if a ref does give it, is VAR going to intervene because its too soft for a penelty to be given? It's very easy to say we won't give something that looks soft on countless replays but in the heat of the moment, judging it on live play and if it's a home team player having 10s of thousands of supporters screaming for a pen, I bet we will see pens like the Sterling one being given. It's the ones where a player forces the contact that needs to be stamped out of the game and that is where imo VAR should intervene.

I do worry the bar is going to be set too high and we are heading back where we were a few years ago where a subjective decision is not going to be overturned. Yes the amount of pens did increase last season but it was nothing too significant, it was not like the numbers we saw in Serie A in the season before last where referees were given every handball a penelty virtually.

Interesting you think PKs haven't gone up that much
Over time, is this good for the game?

 

[GraemeS]

I don't think there's any real reason to think more or fewer penalties can be described as automatically either good or bad "for the game". We all know on a Saturday afternoon that if two teams decide to turn up and kick lumps out of each other, there's only so much a referee can do - and similarly in the PL, if teams are going to keep tripping each other in the area then blaming VAR for those actions is looking in the wrong place IMO.

Also intrigued to know what you credit for the big jump in that chart in 2005? In both % and raw number terms, going from ~60 penalties a seaon to ~95 is a bigger jump than the one season of data we have that suggests a jump from ~95 to 120. What caused that change? Was that good or bad for the game?

Anyway, I think Paul makes a very good point that people too often forget. VAR isn't a robot, it's a person and just like the people we send out on the pitch, that person is susceptible to human error. One of the incidents that has come up in discussions around new VAR this season is the narrow offside decision that preceeded Van Dijk being clattered by Pickford. Great, that offside wouldn't have been given now. But the red card could still have been given under old VAR protocols - that was simply a human error, and there's no reason to expect any more or less of that this season with broadly the same set of officials...

 

[Big Cat]

I don't think there's any real reason to think more or fewer penalties can be described as automatically either good or bad "for the game". We all know on a Saturday afternoon that if two teams decide to turn up and kick lumps out of each other, there's only so much a referee can do - and similarly in the PL, if teams are going to keep tripping each other in the area then blaming VAR for those actions is looking in the wrong place IMO.

Also intrigued to know what you credit for the big jump in that chart in 2005? In both % and raw number terms, going from ~60 penalties a seaon to ~95 is a bigger jump than the one season of data we have that suggests a jump from ~95 to 120. What caused that change? Was that good or bad for the game?

Anyway, I think Paul makes a very good point that people too often forget. VAR isn't a robot, it's a person and just like the people we send out on the pitch, that person is susceptible to human error. One of the incidents that has come up in discussions around new VAR this season is the narrow offside decision that preceeded Van Dijk being clattered by Pickford. Great, that offside wouldn't have been given now. But the red card could still have been given under old VAR protocols - that was simply a human error, and there's no reason to expect any more or less of that this season with broadly the same set of officials...

I just posted stats from Google because I was interested to see the trend. Make of it what you will, but I'm surmising that it reflects the level of cheating in the game and how VAR has sympathised with that cause. I only read the first two lines of your post, so forgive me if I missed anything . I just haven't got the will today
Whatever, I like to see goals scored by ability, rather than some unmerited re-refereeing intervention or 'act of God'

 

[Peter Grove]

I'm really not a fan of wide lines that a make a player onside; IMO, if we are adding gray areas for VAR, the answer should be that the call on the field stands. This is moving from correcting clear errors (well, we left that long ago) to reversing correct calls on OS because they were too close.

I actually think it's the correct way to do this. If the gap between the players is so tiny that it's within the margin of error related to the frame rate/speed of travel issue that we've discussed before, I feel the players should be considered as being in effect level - and level means onside.

If you just go with the on field decision and the AR has given the attacker as offside when, according to the technology being used, we've decided (and can see on the images) that he was functionally level, then you'd be allowing an incorrect decision to stand.

In the Netherlands, they allow the on field decision to stand but I think that's wrong, for the reasons I've given above.

 

[one]

This is going to be a long post and a proposal that I think would work well for VAR offside.

There are three uses of technology I know of for adjudication in sports for 'factual' decisions that have been well used and rarely ever disputed. GLT for football, Hawk Eye for LBW in cricket and hawk eye in tennis for ball in/out. These have 3 things in common:

• No intervention of human for drawing or placement. Its all 'computerised'.
• They all show a virtual replay to audience with zooming used to show the gap/distance
• They all use margin of error for automated decision.

I think we can use a similar concept for offside as well. So here is the process that should work.

1. A software uses multiple synched cameras to detect depth and placement. (similar to GLT)
2. The two best frames are used (the one before touch of ball and the one after) by the software to detect the exact timing for offside. This will be a time between the two frames.
3. A virtual 3D image of ball, second last defender and suspect attacker is built by the software at the exact moment of touch.
4. In this virtual image, the suspect attacker is moved away from the attacking goal line by the margin of error (I think this is 5mm for GLT). The margin of error is calculated the same way as GLT.
5. In the virtual image a line parallel to goal line is drawn by the software. The edge of this line touched the edge of the closest part of the second last defender, or ball if closer to goal. Keep in mind both the ball and second last defender are virtual images so they would have very sharp images just like what we see in GLT images.
6. If the attacker is behind the line then it is not offside.

The way this is sold the the public is simple and already trailed as mentioned before and works. In this case both the stadium and TV audience are shown hawk eye like replay of the 3D image.

1. This starts with the wide image freeze frame of actual broadcast. This is the closest of the two frames used for the virtual image.
2. The actual image is then morphed into the the virtual image
3. The virtual image is zoomed to the contact between the ball and the attacker who kicked to show the exact moment.
4. The image is zoomed out and then zoomed into a close up of second last defender (or ball) and the line, showing the edge of the line touching the furthest part of the second last defender. This view is from above (similar to GLT), meaning a vertical line is not required.
5. The image is zoomed out and then zoomed into the area where the attacker is. Again this view is from above with no vertical line required. If this is a close call the zoom is increased more and more (like GLT or tennis hawk eye) until a gap or an overlap between the line and the player is clear. This is possible because it is a virtual 3D image and all edges are sharp.

The whole replay should take less than 15 seconds and the entire process less that 30 seconds.

And i shall call this OLT (Offside Line Technology)

 

[Big Cat]

This is going to be a long post and a proposal that I think would work well for VAR offside.

There are three uses of technology I know of for adjudication in sports for 'factual' decisions that have been well used and rarely ever disputed. GLT for football, Hawk Eye for LBW in cricket and hawk eye in tennis for ball in/out. These have 3 things in common:

• No intervention of human for drawing or placement. Its all 'computerised'.
• They all show a virtual replay to audience with zooming used to show the gap/distance
• They all use margin of error for automated decision.

I think we can use a similar concept for offside as well. So here is the process that should work.

1. A software uses multiple synched cameras to detect depth and placement. (similar to GLT)
2. The two best frames are used (the one before touch of ball and the one after) by the software to detect the exact timing for offside. This will be a time between the two frames.
3. A virtual 3D image of ball, second last defender and suspect attacker is built by the software at the exact moment of touch.
4. In this virtual image, the suspect attacker is moved away from the attacking goal line by the margin of error (I think this is 5mm for GLT). The margin of error is calculated the same way as GLT.
5. In the virtual image a line parallel to goal line is drawn by the software. The edge of this line touched the edge of the closest part of the second last defender, or ball if closer to goal. Keep in mind both the ball and second last defender are virtual images so they would have very sharp images just like what we see in GLT images.
6. If the attacker is behind the line then it is not offside.

The way this is sold the the public is simple and already trailed as mentioned before and works. In this case both the stadium and TV audience are shown hawk eye like replay of the 3D image.

1. This starts with the wide image freeze frame of actual broadcast. This is the closest of the two frames used for the virtual image.
2. The actual image is then morphed into the the virtual image
3. The virtual image is zoomed to the contact between the ball and the attacker who kicked to show the exact moment.
4. The image is zoomed out and then zoomed into a close up of second last defender (or ball) and the line, showing the edge of the line touching the furthest part of the second last defender. This view is from above (similar to GLT), meaning a vertical line is not required.
5. The image is zoomed out and then zoomed into the area where the attacker is. Again this view is from above with no vertical line required. If this is a close call the zoom is increased more and more (like GLT or tennis hawk eye) until a gap or an overlap between the line and the player is clear. This is possible because it is a virtual 3D image and all edges are sharp.

The whole replay should take less than 15 seconds and the entire process less that 30 seconds.

And i shall call this OLT (Offside Line Technology)

I've always said that 'cartoonising' the offside would work for curious phycological reasons. But then the Broadcasters will always show the freeze frames now because that practice is well & truly established. The loss of spontaneous celebration when the ball hits the net is my biggest problem with VAR. If it wasn't for that, I'd probably accept the 'direction of travel' (to coin an irritating phrase!)