Greenlizard0 Weekend Football Thread ** spoilers ** [1st - 2nd January 2020]

We'll never know what the margin of error is because each case is different. I can't recall the exact speed but somebody calculated that if the defender standing still and the striker is sprinting directly straight then the margin of error is 15cm based on the cameras being used. Change the speed of the attacker, the angle of his run, whether the defender is moving in one direction or the other, the speed that he's moving and the margin of error is completely different. You cannot calculate and change the margin of error for every decisions so do you use the maximum possible margin of error, possibly 20cm+? In which case that will pretty much cover every marginal call.

The tech has to be used for every call (you don't know whether something falls within the margin of error until you've meansured the starting point) and the moment we revert back to the on-field decision just once, we're losing more than we're benefiting.
The margin of error is calculated to take that all in, it's basically a statistics and probability calculation, it's usually the point at which you are only 95% confident in the measurement (from what I can remember anyway, been years since I've done those sorts of calculations).
 
The margin of error is calculated to take that all in, it's basically a statistics and probability calculation, it's usually the point at which you are only 95% confident in the measurement (from what I can remember anyway, been years since I've done those sorts of calculations).
You can only work out that 95% confidence level if you know the speed of each player, the direction of their runs and so on. This is not something that can be built into the current system and I'd imagine would take an incredibly long time to calculate live. Therefore you have to build a set margin of error into the hawkeye technology and to cover every eventuality you'll have to make it so large that we'll not be making any calls with the tech, none of the remotely marginal ones at least. So again, it's maximum disruption, minimum benefit.
 
You can only work out that 95% confidence level if you know the speed of each player, the direction of their runs and so on. This is not something that can be built into the current system and I'd imagine would take an incredibly long time to calculate live. Therefore you have to build a set margin of error into the hawkeye technology and to cover every eventuality you'll have to make it so large that we'll not be making any calls with the tech, none of the remotely marginal ones at least. So again, it's maximum disruption, minimum benefit.
You don't, players can only run so fast and at 50fps or whatever they use they can't move that much, so that sets a bound on the inaccuracy associated with that and so on with anything else that impacts the accuracy. It would be a relatively easy calculation to work out upfront. Hawkeye will certainly have done it (I hope!) for the goal line technology and the line on the screen system. It's whether they were asked to do it for the overall offside system taking in the operator errors, or whether the PL did it themselves (I doubt!).
 
You don't, players can only run so fast and at 50fps or whatever they use they can't move that much, so that sets a bound on the inaccuracy associated with that and so on with anything else that impacts the accuracy. It would be a relatively easy calculation to work out upfront. Hawkeye will certainly have done it (I hope!) for the goal line technology and the line on the screen system. It's whether they were asked to do it for the overall offside system taking in the operator errors, or whether the PL did it themselves (I doubt!).
So as I said, you'll have to set the margin of error based on the worst case. If 15cm is based on an average sprinting speed and the defender standing still, what would the margin of error be for the maximum sprinting speed and the defender running in the opposite direction? 30cm+? That's all the remotely marginal decisions covered there.

We're just going around in circles now. I cannot see how you can introduce a margin of error and it be beneficial. There will be zero change to the time it takes to make these decisions and we'll be seeing fewer correct decisions.
 
So as I said, you'll have to set the margin of error based on the worst case. If 15cm is based on an average sprinting speed and the defender standing still, what would the margin of error be for the maximum sprinting speed and the defender running in the opposite direction? 30cm+? That's all the remotely marginal decisions covered there.

We're just going around in circles now. I cannot see how you can introduce a margin of error and it be beneficial. There will be zero change to the time it takes to make these decisions and we'll be seeing fewer correct decisions.
So the system could be inaccurate upto 15cm? Bloody hell that's not very good!

Edit: Possibly even more than that, that's just being out by 1 frame at 50fps, 2 frames could make it 30cm. It just re-inforces my view that it shouldn't be used to used with any confidence to say if someone is 50mm offside or not.
 
Last edited:
Mourinho booked for verbal abuse.

Lol.

Bet he's already looking forward to his multi-million pound cheque after being sacked and looking for the next mug club to give him a nice new fat contract.
 
How many of those 19 are Alli selling a dive? :p

About as many times that Southampton players should have been booked for kicking the ball away after the whistle has gone. Happened at least twice with not even a hint that the ref would even consider it.

They haven’t even been subtle, back of the stands jobs.
 
We can't even pass the ball 5 yards. Jose won't make 18m at this rate. He will demand a new squad and Levy will tell him to do one with the team in 15th position in the league.
 
10 points from Villa, Chelsea, Palace and Spurs is a bloody ludicrous return frankly, given the state of our play a month ago.
 
Back
Top Bottom