moniker Posted February 22, 2015 Posted February 22, 2015 i have no problem with umpire's call but it should not be counted against number of reviews you can use. If you've used a review that should mean you're sure it is a howler. If it goes to umpire's call that means it was not a howler. You don't deserve to keep the referral after wasting everyone's time and trying to use the DRS as a tactical weapon.
Crookbond Posted February 22, 2015 Posted February 22, 2015 I haven't researched about this technology but if the error rate in the trajectory is +- 5 cm and if the system has been tested for so long, I reckon the average error rate could be around zero with an extreme of +-5 cms. We don't really know the distribution of the error. In other words, it could be that the likelihood of the ball remaining close the the projected path must be quite high. Hawkeye is also used in Tennis and is considered full proof in the absence of a better technology. There's no reason why we have to assume that the ball is somewhere around the extremes if the on-field umpire's decision has to be withheld The "small" difference is that in tennis, hawk eye does not predict the path of the ball.
Old guy Posted February 22, 2015 Posted February 22, 2015 If you've used a review that should mean you're sure it is a howler. If it goes to umpire's call that means it was not a howler. You don't deserve to keep the referral after wasting everyone's time and trying to use the DRS as a tactical weapon. when technology itself is not 100 percent sure how can you criticize captain for taking review ? how is it fair that same ball can have 2 completely different decisions but you still lose review ? sure if its not even umpire's call and you are completely wrong you should be penalized but by taking away reviews for umpire's call you are not stopping howlers but punishing captains for limitations of technology .
adi B Posted February 22, 2015 Posted February 22, 2015 I Don't mind the present rule at all.I think umpires call is fine because it's he who gave the decision in the first place and if the fielding team reviews for a 50-50 call rather than for a howler then they should better know that it'll be a massive gamble unless they are 100% sure that it's a howler
-Robin- Posted February 22, 2015 Posted February 22, 2015 i dont mind Umpire's call but losing a review because of umpire's call is ridiculous
MK55 Posted February 22, 2015 Posted February 22, 2015 The "small" difference is that in tennis' date=' hawk eye does not predict the path of the ball.[/quote'] Ya tennis isn't the right example. But the point I'm making is we don't need the umpire's intervention ( when we are using technology to correct a possible error made by them. We shouldn't be using the original decision as a starting point
putrevus Posted February 22, 2015 Posted February 22, 2015 Dont have problem with umpires call, if one side gets double whammy by these calls then it is major problem. They should define umpires call is either out or not out and side should not lose review.
CSK Fan Posted February 22, 2015 Posted February 22, 2015 Ya tennis isn't the right example. But the point I'm making is we don't need the umpire's intervention ( when we are using technology to correct a possible error made by them. We shouldn't be using the original decision as a starting point How many times do we need to repeat this? H/E themselves admit a +-5 error. So of course you need their original decision as a base point. If there was no error you would not need a reference point. Science (maths) often needs a starting point in such situations. And no, H/E is not to correct errors but to find out if an error has been made. Just because someone refers doesn't automatically mean the umpire has made an error
CSK Fan Posted February 22, 2015 Posted February 22, 2015 It should be not out always.Benefit of doubt should go to batsman.Out should be given only when its conclusively out Disagree. a lot of LBWs will be turned down on review then. As if the bowlers are not suffering enough
MK55 Posted February 22, 2015 Posted February 22, 2015 How many times do we need to repeat this? H/E themselves admit a +-5 error. So of course you need their original decision as a base point. If there was no error you would not need a reference point. Science (maths) often needs a starting point in such situations. And no, H/E is not to correct errors but to find out if an error has been made. Just because someone refers doesn't automatically mean the umpire has made an error The starting point can be the actual trajectory of the ball before impact and the previous database of deliveries that was used to build the algorithm. And Off course there will be an error. The point is we do not have to assume a maximum error just for the sake of protecting the on field decision. Science or maths often use statistics where they study the distribution of the error. If the error distribution is such that the max of +- 5 cm occurs only once in 100 outcomes then it makes no sense to give any significant weight to it. As I said no one really knows the technology and the algorithm that these guys use. So we cant comment how the error behaves with that +-5 cm.
moniker Posted February 22, 2015 Posted February 22, 2015 when technology itself is not 100 percent sure how can you criticize captain for taking review ? how is it fair that same ball can have 2 completely different decisions but you still lose review ? sure if its not even umpire's call and you are completely wrong you should be penalized but by taking away reviews for umpire's call you are not stopping howlers but punishing captains for limitations of technology . Technology is not 100% sure to provide a verdict on close calls. There's no problem with the current technology to prevent howlers. My point is captains should use the DRS to overturn howlers, not for close calls. Umpire's call itself means it was not a howler. There's no limitation with the current technology to overrule absolute howlers.
CSK Fan Posted February 22, 2015 Posted February 22, 2015 The starting point can be the actual trajectory of the ball before impact and the previous database of deliveries that was used to build the algorithm. And Off course there will be an error. The point is we do not have to assume a maximum error just for the sake of protecting the on field decision. Science or maths often use statistics where they study the distribution of the error. If the error distribution is such that the max of +- 5 cm occurs only once in 100 outcomes then it makes no sense to give any significant weight to it. As I said no one really knows the technology and the algorithm that these guys use. So we cant comment how the error behaves with that +-5 cm. We can't comment but surely the makers can? they are the ones quoting the error rate and surely as makers, they will be on the conservative side and not generous with the margins?
sourab10forever Posted February 22, 2015 Posted February 22, 2015 Agree. We need to consider the developers verdict as final. But have they mentioned that the common error rate is +-5 or is it the extream case?
akshayxyz Posted February 23, 2015 Posted February 23, 2015 I haven't researched about this technology but if the error rate in the trajectory is +- 5 cm and if the system has been tested for so long' date=' I reckon [b']the average error rate could be around zero with an extreme of +-5 cms. We don't really know the distribution of the error. In other words, it could be that the likelihood of the ball remaining close the the projected path must be quite high. Hawkeye is also used in Tennis and is considered full proof in the absence of a better technology. There's no reason why we have to assume that the ball is somewhere around the extremes if the on-field umpire's decision has to be withheld That is not how it works. Any such predictive technology system is going to have non-zero error. only the extent,variance, confidence etc. can be improved upon for a given scale/resolution. From my understanding of capabilities of current technology, it is already at the level where it is better (on an average) than human-eye in tasks of estimating predictive path in real-time.
MK55 Posted February 23, 2015 Posted February 23, 2015 We can't comment but surely the makers can? they are the ones quoting the error rate and surely as makers' date=' they will be on the conservative side and not generous with the margins?[/quote'] BTW from where have you come up with the +-5 cm data?? I did some search and in one article, they have claimed that the error is around 25mm in extreme cases and less than 15mm on average. 45mm is the protocol but it does not represent the error of the projection. The ICC could have agreed to it just to justify the on field decision which is what I'm arguing about. http://cricketingview.blogspot.in/2010/06/problems-with-hawkeye.html
MK55 Posted February 23, 2015 Posted February 23, 2015 That is not how it works. Any such predictive technology system is going to have non-zero error. only the extent,variance, confidence etc. can be improved upon for a given scale/resolution. From my understanding of capabilities of current technology, it is already at the level where it is better (on an average) than human-eye in tasks of estimating predictive path in real-time. Yes... I just assumed that the mean error must be much less than 5 cms and therefore went to the extreme case by saying that it must be very small...
Old guy Posted February 23, 2015 Posted February 23, 2015 Technology is not 100% sure to provide a verdict on close calls. There's no problem with the current technology to prevent howlers. My point is captains should use the DRS to overturn howlers, not for close calls. Umpire's call itself means it was not a howler. There's no limitation with the current technology to overrule absolute howlers. Makes sense ...my problem was with only 1 review and some time captain making mistake the entire review system becomes useless as from that point on you can't even use it for howlers and some teams like s***anistan were so scared to use that they did not even use it for howler . Also I will argue we after watching replay know something is bowler the players on field might not. Current howler is not defined by bad decision but what technology can accurately predict ...if it was more accurate the same system will have more howlers and less umpires call ...that's what I was trying to say .
Old guy Posted February 23, 2015 Posted February 23, 2015 Also if it's just for those totally swing decisions where batsman muddled the ball and was given LBW the third umpire can do it and we can just forget all review system...it will eliminate more howlers anyway .
Crookbond Posted February 24, 2015 Posted February 24, 2015 That is not how it works. Any such predictive technology system is going to have non-zero error. only the extent,variance, confidence etc. can be improved upon for a given scale/resolution. From my understanding of capabilities of current technology, it is already at the level where it is better (on an average) than human-eye in tasks of estimating predictive path in real-time. There is very little evidence to show that technology is better than humans for a cricket ball trajectory prediction.
NareshK Posted February 24, 2015 Posted February 24, 2015 You need at least 2 reviews per innings for a team even in an ODI. With only 1 review, if you make one bad review you have no chance of referring another bad call by the umpire which could prove crucial.
Recommended Posts
Create an account or sign in to comment
You need to be a member in order to leave a comment
Create an account
Sign up for a new account in our community. It's easy!
Register a new accountSign in
Already have an account? Sign in here.
Sign In Now