***Official*** DRS discussion thread

[vcs]

Because it makes the line clear.

Yeah, that's a good point.

 

[Jarquis]

Then every single marginal decision would be referred?

What's the problem with that? Not all of them would be marginal, a batsman could refer one thinking it hit him a bit high to find out it's taking out middle half way up. Hawkeye's shown that players get it wrong.

Why? I've always been a "benefit of doubt to the batsman" guy.

Batsmen get enough in their favour + what Jono said.

 

[vic_orthdox]

I'm generally comfortable with hawkeye, my issue are the ones where it's more like the ball is literally just clipping, like less than 10% (lets say) is hitting the stumps or it almost looks like the ball is right next to the stumps but then the bails come off.

 

[keeper]

I just want to ask everyone that why there is so much blind faith in hawk-eye, cricket is probably the only sport where hawk-eye is used in a predictive fashion whereas everywhere else its used as graphical representation of a real event

Good question, there may be an interesting psychology at play whereby people trust a graphic given by what is perceived as an authoritive body. Certainly you hear very little about the inherant margins of error in such technology - if it's clipping it's assumed to be clipping.

I'm very much in a minority in that I could live without UDRS. Human error in officiating is as much part of the game for me as errors in other areas of play and very much part of the fabric and drama of the game. Why would certain Indian players or certain England players, for example, complain about umpiring standards whern they can't field or run between the wickets much beyond schoolboy level? And umpiring standards haven't dropped - they just come under more scrutiny and attract higher levels of expectation.

I also miss the incredibly dramatic moment of a massive LBW appeal followed by the raising of the finger. This is now a qualified dramatic moment.

Having said all that, I'm not a complete luddite and I wouldn't get rid of it now. Think there's a good balance between the authority of the on-field umpires, taking time out of the game and the need to get rid of the worst howlers. I certainly wouldn't expand it further though and think on-field call is a good idea - then the burden of proof lies with UDRS to overturn a decision, a good thing I think.

Couple of final points before anyone who actually reads this nods off. First, it's use must be consistent across series. It's absurd that any one country can veto it and arrogant to do so for an away series. Second, we know that the science dictates a false impression will be generated when assessing whether cayches carry. Use in this context has to be severely restricted.

Good debate this by the way, obviously a more intelligent forum.

 

[Top_Cat]

Because it makes the line clear.

haha, lawyers/students just can't deal with uncertainty.

Drawing a line like that, when there's clearly statistical limitations with the hardware, sacrifices accuracy for efficacy, tbh. No less arbitrary than 'if 50% of the ball is hitting, it's out', given, but I wouldn't agree with trading one arbitrary distinction for another. Just need to prediction to improve which will happen.

 

[Neil Pickup]

The Hawkeye prediction is much more accurate than half a stump, though, isn't it? I thought this margin of error was introduced to cater for those who were concerned about the accuracy of the prediction, and ensure the discretion of the technology wasn't given too much weight.

On the subject of DRS, here is some of the raw footage from that Ajmal/Tendulkar leg before shout.

 

[Howe_zat]

The Hawkeye prediction is much more accurate than half a stump, though, isn't it? I thought this margin of error was introduced to cater for those who were concerned about the accuracy of the prediction, and ensure the discretion of the technology wasn't given too much weight.

On the subject of DRS, here is some of the raw footage from that Ajmal/Tendulkar leg before shout.

Thanks for that. I'm still yet to see any claim of Hawkeye being clearly wrong that's accompanied by an image, or hardly ever something more than "I reckon".

 

[Neil Pickup]

Thanks for that. I'm still yet to see any claim of Hawkeye being clearly wrong that's accompanied by an image, or hardly ever something more than "I reckon".

Hmm, I wouldn't go that far. I remember a very early Hawkeye clip showing that a delivery that had clean bowled somebody (Kumble, IIRC) was missing the stumps.

 

[Jarquis]

Didn't the same happen with Warne vs Strauss?

 

[Neil Pickup]

That just broke the system because it turned too much, didn't it?

This just didn't work - IR Bell b Kumble 38 (I know, Bell chucking a start away, shock horror...)

Here's the match thread (post 3029). Other highlights in this game include Munaf' Patel's debut, and Jono getting all offended because I suggested Pathan bowled pies.

 

[Furball]

Didn't the same happen with Warne vs Strauss?

At the Oval Warne was getting stupid amounts of turn bowling round the wicket to left handers - one of his deliveries to Trescothick (iirc) broke HawkEye because it wasn't calibrated to handle that much spin.

YouTube - Shane Warne Big Turn

 

[Top_Cat]

The Strauss/Warne ball was 6 years ago too!

Basically, there's a proportion of the population which will never be comfortable with Hawkeye because there will always be errors in any predictive maths, no matter how comprehensive the model. They want a line which says 'out' or 'not out' so, in the absence of that, will either decide on some arbitrary criteria which allows that or reject it out-of-hand.

That said, I doubt the maths of the prediction mechanism is that complex. Seems like a fairly simple modelling problem. The big issue (and where the big $ would have been spent) would be translating/approximating video footage into animations. Would be related to the quality of the footage and where the largest source of measurement error is, I'd guess.

 

[Fusion]

Hawk-Eye dismisses doubts over Tendulkar lbw.

The company has published a detailed report of the Tendulkar referral on its website.

The above report is fascinating to read.

 

[TumTum]

At the Oval Warne was getting stupid amounts of turn bowling round the wicket to left handers - one of his deliveries to Trescothick (iirc) broke HawkEye because it wasn't calibrated to handle that much spin.

Epic! Warne 1 - Hawkeye 0

 

[Top_Cat]

Hawk-Eye dismisses doubts over Tendulkar lbw.

The company has published a detailed report of the Tendulkar referral on its website.

The above report is fascinating to read.

There is, I think, a huge error in that report and it makes their position look worse when, in reality, they're even more right. On the second-last page, second pic, why take the line/direction to show the case where the balls were hitting middle/leg stump from the point of impact? Wouldn't you take it from where the ball pitched? It doesn't make logical sense unless they reckon there's a possibility the ball moved in the air after impact or after pitching. This, of course, does happen over a long enough distance but not enough in under 2m to make a ball which was missing leg hit it surely?

That is, of course, assuming they didn't do that and just overlaid the colours from the point of impact in putting this out quickly.

Showing the ball's path from the point it pitched would make their case stronger, really. It'd show how much further to the left the ball would have had to pitch to hit the stumps assuming no other changes.

 

[KiWiNiNjA]

They are just showing the predicted part, that's all.
There would be no need to show projected paths from where the ball pitched, because the line from where the ball pitched until the point of impact actually happened. What they are showing is that, in order for the ball to have hit middle and leg or leg, the ball would have had to swing back in. In other words, that's not going to happen.

Basically, they don't feel the need to show other paths from where the ball pitched because they are 100% confident in their representation of the actual path that the ball took.

Anyways, it's amazing that there is so much blind faith in human-eye.

 

[pup11]

Because it makes the line clear.

The only problem with that is that most balls that hit a batsman on the pads is shown to be clipping the stumps, so if we start giving those out then you would see games getting over in a jiffy.

I have heard the commentators say this often and now I too agree with this that players have actually started to use UDRS as a tactical ploy where they are actually trying to manipulate technology to work around an on-field umpire's decision.

As I have said often UDRS' main job is not to give the players the power to question an umpire's decision but to stop bad decisions from being made, now that is something that can easily be achieved if the 3rd umpire is given the chance to correct an on-field umpire when he makes a mistake and that is something that would spare the umpires from the embarrassment of their decision being constantly challenged by the players.

 

[Spark]

TBH the umpire's call thing is fine. What I'm not so sure about is that if a ball is predicted to miss the stumps by five millimetres then it's overturned, that's a bit iffy.

 

[KiWiNiNjA]

The only problem with that is that most balls that hit a batsman on the pads is shown to be clipping the stumps, so if we start giving those out then you would see games getting over in a jiffy.

I have heard the commentators say this often and now I too agree with this that players have actually started to use UDRS as a tactical ploy where they are actually trying to manipulate technology to work around an on-field umpire's decision.

As I have said often UDRS' main job is not to give the players the power to question an umpire's decision but to stop bad decisions from being made, now that is something that can easily be achieved if the 3rd umpire is given the chance to correct an on-field umpire when he makes a mistake and that is something that would spare the umpires from the embarrassment of their decision being constantly challenged by the players.

How is this wrong? And why is it even a tactical ploy? And why is tactics a bad word, even?

A batsman gets given out LBW, asks for a review, and it's shown to be missing the stumps. The decision is reversed and the batsman continues on. It may not be a howler, but an incorrect decision has still been overturned.

If the LBW is marginal i.e just clipping the stumps, then smart players will not ask for a review. Yes, there are the Watsons who review everything, but in most instances you will see the batsman talk to his partner and then walk off. If the player does review it then he wastes a review.

What is wrong with that? It's either out or you waste a precious review.
I think it was the NZ-RSA match where there was no reviews at all. Yes, umpires made correct decision, but there were also plenty of close calls that could have been reviewed but weren't. I think the players can get some credit as well. They weren't trying to get one over on the umpires. Maybe, just maybe, players want to use UDRS to overturn wrong decisions?

Manipulating technology? Pffft, they aren't jedi ffs.
Oh, and I'm so over the "oh noes, umpires are getting embarrassed" crap.

 

[KiWiNiNjA]

TBH the umpire's call thing is fine. What I'm not so sure about is that if a ball is predicted to miss the stumps by five millimetres then it's overturned, that's a bit iffy.

Benefit of the doubt going to the batsman and all that jazz, methinks.