Sunday, December 15, 2013

DRS - A Day 3 view after a couple of glasses of Sauvignon

It's Midnight just before Start of Play on Day 3, so what else other than some Ramblings on DRS on a Saturday night after a couple of glasses of Sauvignon? Yes I know it's almost the definition of "sad", but there you are.

As far as the technical equipment is concerned, it’s much more complicated than I originally thought. It involves consideration of a very complex technical issue (a rough, rotating sphere spinning and moving through a varying and viscous atmosphere) which has a requirement for a very accurate and speedy element of measurement at the heart of it, followed by a calculated prediction based on those measurements.

It also includes an awareness (I was going to say an understanding but I’m not sure anyone out there really “understands” it!) of how an umpire’s mind works when that mind is out in the middle of a field for 8 hours a day, watching a ball being bowled 540 times. Add in the stress, add in the heat, add in the umpteen outside influences which can and must influence any human mind and you can only be surprised at just how good they actually are. As an aside, it does make you wonder how Jim Laker’s 19 for 90 figures would have changed if DRS had been invented 50 years ago!

On a different tack and on a much more ethereal wavelength, it needs to recognise the subtle and nuanced issue of “The Spirit of Cricket” -  a thing which means very different things to different people. Is the umpire still to be seen as God on the pitch or, now that the facility has become available via technology to improve, support or question his decision making capability, should this new capability be admitted (and to what degree?) within the Laws to override or influence his original decision?

To take the technical issue first, we have to accept that the DRS system is inherently inaccurate. I don’t think anyone disputes that but, that is not the point. You have to compare the accuracy of both the umpire and the box of technical tricks which now sits alongside him at the same time to make an informed judgement. The real issue for HotSpot etc. here is “Is its inherent accuracy sufficient to offer an improved decision?” As an aside, it has always fascinated me that our Indian friends seem to have rejected the system because it’s not completely 100% fool proof. At the end of every Away Series they all get onto an aeroplane (the same aeroplane even!) and fly home. Now I have a degree in Aeronautical Engineering, and I know only too well that there is no such thing as a 100% safe aircraft. The Indians seem to want to play hardball when they are just playing a game, and offer themselves up to whichever God they follow (I was going to say "take a flyer", but that wouldn't have been very tasteful) when their lives depend on it. Weird.

My understanding is that the ICC statistics show that Elite panel umpires get something around 92% of their decisions correct, and that this is improved to something like 96% when DRS is involved. On the face of it, this doesn’t seem like much of an improvement.

But turn the numbers the other way round. On their own, umpires therefore get 8% of their decisions wrong, a figure reduced to 4% if they use DRS. That is an improvement of 50%! If you offered any engineer a system which halved the error rate on whatever he was developing, he’d bite your hand off. And yet here we are scratching each other’s eyes out as to whether it improves the situation.

Actually, I think the issue is greater than that. Not all umpire decisions are the same. Some are easy and some are very marginal. The 92% figure presumably includes both sorts of decisions. For the “hitting middle stump a foot off the ground” decision, the success rate is probably 100%, and DRS would never be at variance with an umpire’s decision. As the ball starts to swing, or move nearer the leg or off stump, the error rate increases. So the probability is that for the “difficult” decisions (a hairsbreadth away from the outside of the leg stump bail) I suspect the accuracy of most umpires’ judgment starts to resemble tossing a coin. If this is the area which becomes contentious where discussions about DRS are concerned, then the reality is that we’re probably understating the improvement we get when DRS is used – and probably by quite a lot.

One of the points you make in your article is important. However much it improves, it will still leave an area where errors still persist, and the consequent question of what we should do about it. The boffins will argue that the system will continue to improve, and it will, but always leaving a smaller element of error. We have to decide if we live with that or not. Perfection is not attainable.

The good thing about a system like DRS is that it’s not subject to the vagaries of an umpire’s mind. Its internal algorithms and controls rules are the same for every ball. Unlike an umpire, who is in the final analysis a human being, the system doesn’t look at a small diversion on the sightscreen at the wrong time. It isn’t deflected by a hilarious timely sledge from one of the 22 wits on the field. It’s on all the time, unvaryingly, unblinkingly. And it doesn’t have to look at the back of the bowler’s foot ½ second before he needs to make a potentially match changing decision some 20 yards away. And it doesn’t suffer from stress like an umpire must, particularly at pivotal moments in a match.

I am personally sold on it, as a system. Anything which improves the quality of a decision is good in my book.

No, the real issue here is different. It’s down to how we use it.

·      - Should the players have a say, or should the decisions be solely the responsibility of the umpire, with no claim for redress? One referral? Two referrals? Top up after 80 overs? What logic gave birth to that one?

·      - Should the off-field umpire be constantly talking to the onfield umpires to tell them what DRS is saying, so the onfield umpires always know before they announce a marginal decision what the guys up in the box are going to say, so reversals of the decision disappear?

·     -  Do we have the right qualities in our umpires, or is there a different set of skills required by a Third Umpire? Many of the Twitter blow-ups are rooted firmly in this area in my view.

·     -  How important is the “flow” and the speed of the game? DRS is a real interrupter in the continuity of play, but then so are drinks breaks, and players going off the field to pick up a packet of Polos. And those are far more easily solved. Real time Snicko must surely be a significant addition to the technical armoury here.

·     -  Should the Umpire still be seen as the ultimate decision maker? Surely, that was OK when there was nothing to help him, but life has, whether we like it or not, changed for ever. Pandora’s box opened a few years ago and DRS flew out. It cannot be closed, at least with DRS back inside it. So the new systems will always be there, and the TV companies will be showing it. I can’t believe the umpires want to have all their errors, however unintentional they are, shown on every big screen and TV in the land, and not have the power to correct them on the field of play.

So let’s live with it, and simply accept that there will still be errors. To me setting a set of rules which are something like –

·     -  No mark on HotSpot, the ball didn’t touch the bat. Sometimes that will be wrong, but – Live with it.

·     -  No Snicko perturbation, the ball didn’t touch the bat. The fact that the audio and visual signals are not quite in sync is surely down to internal processing lags and perhaps the fact that sound travels slightly slower than light. Sometimes that will be wrong, but – Live with it.

·     -  Hawkeye is a bit more difficult. Its predictive element is, I suspect far more accurate than an umpire’s judgement, at the margins. So personally, I’d remove the Umpire’s call caveat. If either all of the ball or an agreed small amount of the ball is predicted to hit the wicket, then as far as I’m concerned, I’d give it out. The fact that there’s an Umpire’s call involved seems to be irrelevant. At the margin, I think he’s just guessing, and the maths done by Hawkeye must be more reliable. When you’re doing a landing at a fogbound airport, I’d much rather rely on the box of tricks in the cockpit and on the ground, rather than on a highly stressed pilot’s eyesight. Even though the Automatic landing system is only reliable to 1 in 10,000,000.

Anyway, Day 3 starts in a couple of hours, so an hour of sleep beckons! Let’s hope the sleep deprivation will be worth it.

Roger Cable


Update at 12.00 Sunday. Well - it wasn't! Tired and seriously disappointed. What on earth has happened to our team? Win or Lose - that’s what happens, but the way it’s all unravelled is really, really depressing.