DRS , good, bad or could be better

First topic message reminder :

I love using technology to help the game- however what i hate is this strange use of DRS, that more than half the time reviews go back to the original umpires decison. its immaterial whether the unpire calls decison out or not out, his decision stays

It doesnt make the best sense- cant we just have clear guidlines as to what is in or out after review- ie half the ball has to hit the stumps means its out, if it doesnt its in!

In the past (in tennis) I have seen estimates of Hawkeye accuracy of +/- 5 mm.

Nore Staat wrote:In the past (in tennis) I have seen estimates of Hawkeye accuracy of +/- 5 mm.

well thats good to know- and if hawkeye used that - kp would have stayed in

100% in the context Oakey, in that it has to be clearly the wrong decision for it to be overturned. KPs wasn't wrong so move on. Is there really any point in complaining about it? What will it change? England have performed poorly again and all the KP situation is doing is covering up the mess our batsmen have made of the bowlers hard work.

hampo171 wrote:100% in the context Oakey, in that it has to be clearly the wrong decision for it to be overturned. KPs wasn't wrong so move on. Is there really any point in complaining about it? What will it change? England have performed poorly again and all the KP situation is doing is covering up the mess our batsmen have made of the bowlers hard work.

i am not complaining about decision in reagrds to england- i am discussing drs, thats what forums are for.

do you honestly think because i like to discuss the laws of the game- i am masking my annoyance from englands batting- blimey

CF & Mistir, I'm not going to get the red pen out just yet (as I've been taking part in the debate), but can you BOTH please drop the personal attacks, they're unnecessary.

Mystiri is broadly arguing that in cases where Hawkeye suggests that a ball is just clipping the stumps the batsman should be given the benefit of the doubt; as given Hawkeye's margin for error in those cases there is a decent chance that the ball would have missed.

By way of an example, supposing in the case where a batsman reviews an out decision by the on-field umpire and (these numbers are hypothetical) Hawkeye is showing a ball clipping 2mm of the stump, while there is a 5mm margin for error. In this hypothetical case there is around a 1 in 3 chance that the ball was missing the stumps (remember there's a 50% chance the ball would have hit even more of the stump). Given cricket's tradition of giving the batsman the benefit of the doubt, I'd prefer to be at least 80% confident it was out.


As an aside, I've just found this blog http://cricketingview.blogspot.com/2010/07/reporting-on-hawkeye-ii.html which suggests that Hawkeye recommended showing an "uncertainty-ellipse" around the ball (as I suggested up-thread - nice to see an expert agreeing with me ) but broadcasters weren't keen:

We have proposed on a number of occasions the idea of presenting an “uncertainty ellipse” around the ball, but all broadcasters have rejected the idea for a number of reasons:

1) Almost all commentators are ex cricketers, and generally not that scientifically bright. They would have no ability to explain what an uncertainty ellipse is
2) Broadcasters prefer a “definitive” where was the ball going – some commentators are then good at interpreting that information saying something like: “Hawk-Eye shows it just clipping leg stump, so a good decision to give the benefit to the batsman”
3) If the uncertainty ellipse is supposed to reflect Hawk-Eye error, then the ellipse would be so small around the ball that you would not see it around the ball

cheers kiwi,

i brought my conversation's from my own thread on to this one. my posts are obviously abit confusing to some..

We dont know what the probability of that ball missing was, we dont have access to the full data and probability calculations.
We do know what the most likely outcome was, that it would have hit the stumps.

Purely from a a laws perspective it was right to have been given out. From the information we have it most likely wouldve hit the stumps.
Both the umpire and hawkeye agree, yet we say ity shouldnt be out?

That comes down to an old convention, lbw should only be given when youre sure its out.
Trouble with that was it was massively difficult to apply evenly. Frankly some umpires seemed to believe that lbw exists, doubly so when Tendulkar was involved. Dickie Bird is one who springs to mind as a law unto himself, regarding it as impossible for certain bowlers to get lbw calls.
Hawkeye has challenged that by showing just how bad some of these judgements were, and that a lot of things that people called "thats not an lbw" would clearly have hit the stumps, and therefor should have been given.
This case is less clear cut. We have hawkeye saying theres an element of doubt, trouble is we dont know how much doubt that is. So whats an acceptable level of doubt? If hawkeye said it were 70% likely to have hit the stumps would you be satisfied? 90% ?
You have to draw the line somewhere, no calculation of this type can ever be 100% certain....what if the ball had hit the stumps but caused the bails to jump up and lad back in place?

The application of hawkeye in this case made the decision no worse. It upheld that the umpire was most likely correct in his assertion that the ball wouldve gone on to hit the stumps. I get the objection is that theres still an element of doubt but its hard to say KP was hard done by or savage the umpire for being "highly confident" it wouldve struck, he was only marginally wrong in that it was only "most probable."

I really dont get how we can overturn a call based on evidence showing it was most likely correct, that would be far more absurd that basing a call on the balance of evidence or just leaving it full stop to the umpire.

The DRS consistency or inconsistency needs to be debated. I think a batter has to be given out if any part of the ball is shown to be touching the stumps, regardless of whatever been the onfield umpire's call. That ways, there won't be all these rather meaningless debates. Its time the bowlers get a fair share in the game.
Unless this is done, just do away with the ball tracking, and use sneecko, and yes, hot spot.

The DRS consistency or inconsistency needs to be debated. I think a batter has to be given out if any part of the ball is shown to be touching the stumps, regardless of whatever been the onfield umpire's call. That ways, there won't be all these rather meaningless debates. Its time the bowlers get a fair share in the game.
Unless this is done, just do away with the ball tracking, and use sneecko, and yes, hot spot.

"
I really dont get how we can overturn a call based on evidence showing it was most likely correct, that would be far more absurd that basing a call on the balance of evidence or just leaving it full stop to the umpire"

it has to be one or the other.

if hawkeye is proved to be accurate to a good degree of certainty- we should use that with a margin of error (the elipse seems like a good idea)

if hawkeye is proved to be to inaccurate then in all hoensty i dont see the point in using it at all.

Would anyone be upset if I moved the last few DRS-specific posts over to the DRS thread? I think they add to the overall debate there, and I actually thought I was posting on that one not this in one of mine.

"The DRS consistency or inconsistency needs to be debated. I think a batter has to be given out if any part of the ball is shown to be touching the stumps, regardless of whatever been the onfield umpire's call. That ways, there won't be all these rather meaningless debates. Its time the bowlers get a fair share in the game.
Unless this is done, just do away with the ball tracking, and use sneecko, and yes, hot spot. "

i totally agree however i think there should be a margin of error appplied- if 5 mm is proved then as long as more than 5mm of the ball hits then its out, if not then its not out.

mystiroakey wrote:"The DRS consistency or inconsistency needs to be debated. I think a batter has to be given out if any part of the ball is shown to be touching the stumps, regardless of whatever been the onfield umpire's call. That ways, there won't be all these rather meaningless debates. Its time the bowlers get a fair share in the game.
Unless this is done, just do away with the ball tracking, and use sneecko, and yes, hot spot. "

i totally agree however i think there should be a margin of error appplied- if 5 mm is proved then as long as more than 5mm of the ball hits then its out, if not then its not out.

Just to clarify the 5mm rule, when they say there is a "5mm margin for error" that generally means that 95% of the time (or 19 balls out of 20) the ball will go through within 5mm of where Hawkeye's put it. They test that by recording a lot of deliveries with no batsman "in the way" and having predict from various points in the flight, and comparing with the actual delivery.

so in reality the margin of error is 5%, however the 95% correct decisions are only accurate to 5mm variance.

when we take that into perepective .Any batsman getting out to a delivery by 2.2 mm could certainly feel hard done by.

especially when we consider that a bowler could take a whole bail on a not out review and the batter could stay in

mystiroakey wrote:atherton makes a very good point on the commentary- drs is changing the game in a fundemental way

This, I feel, is the important point. DRS is going to breed a new generation of batsmen who play in a different way to combat the new challenges they face, for example playing with the pad first to a spinner is no longer an option. Right now it seems like bowlers have an advantage but tbh I think it's because all the batsmen on show (today especially) are out of form and making a lot of mistakes. I don't think today would have gone much differently had DRS not been available.

personally i hope drs changes before the game changes completly

what has happend is that we have limited some bad calls, but not all. we have done this by switching benefit of doubt to the bowlers rather than the batsmen.

i am all for limiting human error,and making the game fairer- however i hope its not just gonna stop here- surely we can keep improving the tech and decision making in all sports.

"I don't think today would have gone much differently had DRS not been available."

drs did reverse some decsions today as the umpire was off his game- so it made it fairer , and it gave the game more wickets.

My point was really that all of the batsmen looked so nervous and out of form they would have got themselves out sooner or later anyway.

oh i see- yeah i reakon your right- they may have got a few more. bowlers were to good out there, batters to rubbish

I really do like the DRS, particularly for lbws. I sometimes think that commentators/ex-players like to hark back to a day when umpires had to give the 'benefit of the doubt to the batsman' due to a lack of certainty. However, a quick look at pre-2005ish umpiring in Tests, particularly in the sub-continent (SL v Eng in 2000-01 was particularly shocking) would reveal that stardards increase significantly with a 'benefit of the doubt to the umpire during a review' system. I have no qualms with Taufel giving KP out today, for example, on a judgment that the ball would have hit the stumps - a decision justified by Hawk Eye. This is because it encourages more correct decisions in the true sense (i.e. not is there no doubt at all that the ball would have hit the stumps but simply would the ball have hit the stumps, providing that the other criteria are filled)

mystiroakey wrote:
what has happend is that we have limited some bad calls, but not all. we have done this by switching benefit of doubt to the bowlers rather than the batsmen.

No we haven't and it is disingeneous to argue that we have. We have arguably moved the benefit of doubt to the umpires.

The fact is the exact margin for error is unkown and variable. We have seen examples of ball-tracking technology which have just been plain wrong (Phil Hughes in Sri Lanka for one, that ball turned!). I actually think DRS works fairly well as it is now, in that if the ball-tracking shows that the umpire got it totally wrong it is overturned, whereas if it shows the umpire was in someway justified to give the decision as he did then his decision is upheld.

But I repeat my argument: Pietersen (and indeed any batsman who have been wrongly given out, except in some circumstances) wasn't "unlucky" except that the umpire (possibly in Pietersen's case) made an error immediately after he did. Why always blame the umpire/the system, rather than the batsman who missed the ball in the first place (I assume Pietersen didn't deliberately miss the ball, if he did then it was certainly an error of judgement as it was too close to leave)?

Would it not just make sense when using Haweye to scale the stumps down by say 2mm width ways and 2mm height (or whatever an appropriate measure of inaccuracy is) and then say if the ball is hitting the stumps it is out and if it is missing then it is not out. Surely this would take away the "umpire's call" decisions which causes the main controversy whilst still keeping the main power with them.

And I'm not sure if I buy this theory that it is turning the game too much in the bowler's corner, you can't say that after one series. Especially after a series where both sides, but mainly England, have batted very poorly at times and not played spin well at all. That is more to blame than than DRS for all the LBW's.

Mike Selig wrote:

mystiroakey wrote:
what has happend is that we have limited some bad calls, but not all. we have done this by switching benefit of doubt to the bowlers rather than the batsmen.

No we haven't and it is disingeneous to argue that we have. We have arguably moved the benefit of doubt to the umpires.

The fact is the exact margin for error is unkown and variable. We have seen examples of ball-tracking technology which have just been plain wrong (Phil Hughes in Sri Lanka for one, that ball turned!). I actually think DRS works fairly well as it is now, in that if the ball-tracking shows that the umpire got it totally wrong it is overturned, whereas if it shows the umpire was in someway justified to give the decision as he did then his decision is upheld.

But I repeat my argument: Pietersen (and indeed any batsman who have been wrongly given out, except in some circumstances) wasn't "unlucky" except that the umpire (possibly in Pietersen's case) made an error immediately after he did. Why always blame the umpire/the system, rather than the batsman who missed the ball in the first place (I assume Pietersen didn't deliberately miss the ball, if he did then it was certainly an error of judgement as it was too close to leave)?

yes i have mentioned that the benifit of the doubt has gone the umpires way.

however we are now talking about the game as it stands- the bowlers have the upper hand in this scenerio- it would be foolish to suggest otherwise based on how recent games have gone under the drs as it stands.

however i do think drs works better than without it.

but i am not happy that this is the end of the road for decision making- a lot of improvement is still needed

"Would it not just make sense when using Haweye to scale the stumps down by say 2mm width ways and 2mm height (or whatever an appropriate measure of inaccuracy is) and then say if the ball is hitting the stumps it is out and if it is missing then it is not out. Surely this would take away the "umpire's call" decisions which causes the main controversy whilst still keeping the main power with them."

yep totally- i think that would be a much better situation- we take out umpire error completly- we are only left with hawkeye error, which from my point of view is less inaccurate and more consitant

"But I repeat my argument: Pietersen (and indeed any batsman who have been wrongly given out, except in some circumstances) wasn't "unlucky" except that the umpire (possibly in Pietersen's case) made an error immediately after he did. Why always blame the umpire/the system, rather than the batsman who missed the ball in the first place (I assume Pietersen didn't deliberately miss the ball, if he did then it was certainly an error of judgement as it was too close to leave)?"

its not really about blaming the umpire, blaming the system as it stands yes, umpires are human, huamns arnt perfect, but then neither are systems.

However the point is simple-if we all had your attitude then we would never have had drs as it stands today, because we would all accept that systems/umpires are fine as there are, and wouldnt try to make things fairer or better.

we constantly try and improve - if we didnt it would be a less evolved world.

we can say that kp was unlucky, we can also say that kp missed the ball- thats all abit immaterial to the debate of improving the drs system.

kps out decsion is a good example of the problems drs faces today.

JDizzle wrote:Would it not just make sense when using Haweye to scale the stumps down by say 2mm width ways and 2mm height (or whatever an appropriate measure of inaccuracy is) and then say if the ball is hitting the stumps it is out and if it is missing then it is not out. Surely this would take away the "umpire's call" decisions which causes the main controversy whilst still keeping the main power with them.
And I'm not sure if I buy this theory that it is turning the game too much in the bowler's corner, you can't say that after one series. Especially after a series where both sides, but mainly England, have batted very poorly at times and not played spin well at all. That is more to blame than than DRS for all the LBW's.

mystiroakey wrote:so in reality the margin of error is 5%, however the 95% correct decisions are only accurate to 5mm variance.

when we take that into perepective .Any batsman getting out to a delivery by 2.2 mm could certainly feel hard done by.

especially when we consider that a bowler could take a whole bail on a not out review and the batter could stay in

But the bowler who wasnt given the wicket would feel more hard done by if it werent given, since it was 95% likely a correct out. As much as theres a chance that the ball was 2mm further away from the stumps theres an equal chance it was 2mm more into the stumps. But in all probability we know it wouldve hit the stumps and therefor under the intention of the lbw law should be out.
The law is there to stop you defending the stumps with your legs. Players have gotten away with it for years because umpires where scared to give decisions.

Is it really such a bad thing if players are forced to hit the ball with the bat and bowlers encouraged to bowl at the wickets? Isnt that the fundamental basic intention of the game?

Not that I think Mike needs any help but I thought the comment just made by David Lloyd very relevant:

''If you play with the bat, DRS is not in it.''

marcus trescothick said earlier on the SKY coverage that graeme thorpe, who is working as batting coach with the lions atm, has been getting the players to take nets against spin with no pad on their front leg to force them to play with the bat so they dont get a whack on the shin from the ball

mystiroakey wrote:

however we are now talking about the game as it stands- the bowlers have the upper hand in this scenerio- it would be foolish to suggest otherwise based on how recent games have gone under the drs as it stands.

Fine. In that case then label me foolish, because I don't see it.

The system isn't perfect and if someone comes up with a genuine improvement (actually I would like it if "out" decisions were still upheld even if the ball was only just missing the stumps, within the degree of accuracy as given by Hawk-eye, since we accept we don't overturn "not out" decisions unless we're sure the ball was hitting the stumps surely we should accept "out" decisions unless we're sure the ball's missing the stumps? Anyway I am somewhat playing devil's advocate) then I'm all for it. But arbitrarily deciding that you should overturn what could be a perfectly correct decision because you feel there's too much doubt isn't a real improvement.

My point was more general, and is borne out of frustration of fans (and some players I coach) moaning at umpires when the players made the original mistake.

why get bogged down with players mistakes.

its so immaterial to reffering- it only confusses matters, surely thats a discussion for another thread

the problem is clear, umpires are human and make mistakes.

every game we get tons of not out lbw calls- most arnt reviewed because the bowlers and fielders realise that the ball will not hit middle stump- but would hit or clip leg of off stump

the batter missed the ball- the ball would have struck the wicket- the player time and time stays in

then you get cases like peitersons, where it clips the stumps by 2.2 mm by hawkeyes reakoning- yes he missed the ball - so the heck what, but he is out, which in a way is fine- however what about the other 20 decisions that could have also been out

the system needs improving- just leave it to hawk eye with a margin of error applied- job done.

yes this will help bowlers out even more- however the game will be fair and consistant.