Friday, 6 April 2012

Do Umpires Dream Of Electric Sheep?

Last week the writer Jon Ronson had a run-in with a group of university lecturers who had created a twitter account very similar to Ronson's own, and used an avatar made from his picture.

It wasn't a fake account as such, although there was some disagreement over what it actually was. Ronson called it a spambot, the lecturers said it was an infomorph. What it did was to source information from Ronson's wikipedia entry and his geographical area of London, use it to invent a persona for itself, and begin spewing out tweets. Some of it was nonsense, some of it was, by default, funny. Ronson quite rightly felt put upon, and made a little film about his confrontation with the spambot builders (This was funny too - Ronson: 'It tweeted 'I like time and cock'! Lecturer: 'And do you?'; Ronson: 'Yes! I mean no!')

The infomorph's future application seemed to be as a kind of outrider of the virtual horizons, sent ahead to a particular location to pull in information and relay it back. As such it had certain random, human qualities. By exercising those, it could infuriate as well as serve. I thought of it when David Hopps at cricinfo referred to 'the philosophy' behind the DRS, because the philosophy is what it's about, and that philosophy is becoming obscured by the DRS' more immediate effects.

In essence the DRS exists to enhance human judgment rather than replace it. It's not the technology itself that has reshaped the game, but the humans using it. Umpires have given more decisions because DRS has shown them that they're right to do so. In turn, it has changed the way bowlers bowl and batsmen bat, how fields are set, how long matches take. What the technology has revealed is how accurate humans can be.

Emboldened, our human accuracy is now questioning the judgment of the machines. This is a good thing, and the predictive element of the ball-tracking system, patently wrong in some cases, will either improve or become something else.

When DRS began we expected it to produce absolutes, because that is what technology does: it has lumpen computing power beyond our reach. Yet, perhaps surprisingly, that is not what has happened. Instead, it has required interpretation; it has posed questions, demanded more from us.

At the heart of DRS has been a paradox of the Schrodinger's Cat kind, that a player can be both in and out to the same ball depending on the umpire's initial judgment. This is something that wasn't foreseen, and it is sometimes hard to defend. For a long time, this blog was against it. An ambiguity like this one runs against the existing concept of technology, which is there to give us an answer, not to throw the question back.

Were this paradox to be applied strictly to the first stated aim of the DRS - to overturn blatantly bad decisions - then there would be no paradox, because it only occurs on marginal calls. But the 'obvious howler' position has never been workable. Ironically, you don't even really need DRS to resolve those, just a TV replay. Instead, we quickly reverted to the usual human requirement for technology: be certain for us.

Now we've discovered that in cricket, rather marvellously, not even technology can be certain, and this demands, as David Hopps suggested, a philosophy for the way we use it. One of the most accurate measurements that DRS has given us is how good the umpires are - right way over ninety per cent of the time. That is the building block for the technology to use in its next phase.

DRS needs to be regarded as an augment to, and an enhancement of, umpiring. For it to become a more organic part of that process, it can't remain a tactical part of the game. The deeply flawed logic that each side has two successful appeals per innings not only blurs the lines between players and umpires, it politicises decision-making and mitigates against fairness (for example when a player left without review is wrongly given out or a fielding side has no remedy when pushing for a win). We're now in the ludicrous position where certain captains are praised for the way they work a system instigated to increase equality, and umpires have been hacked off at the knees by their colleagues for giving perfectly defendable decisions.

Instead we should regard the three umpires and the DRS as a revolving team, and remove the term 'review' from the acronym. The solution is simple enough: the umpires rotate throughout the game, each taking a turn handling the technology on a session-by-session basis. As an integrated unit, each decision can be made using technology where appropriate. The on-field umpire may give an immediate judgment, or he may use the third man before offering it. It's a small shift, but one that removes the notion of the technology bringing an element of challenge to the process.

This way, the DRS - or DS as it might be renamed - fulfills its philosophical aim as an enhancement, while acknowledging the obvious glory of the game; that whether playing or umpiring, it is a fully human endeavour, sometimes random, occasionally maddening, always evolving.

12 comments:

Pavilionopinions said...

Cracking stuff as always. The tactical element is def becoming ever more brazen with players fishing for info off the umps before deciding to go upstairs - was he outside the line, was he playing a shot - when none whatsoever is meant to be given under the present ICC guidelines. Saying 'going down' must be force of habit to some umps trying to placate convinced bowlers, but even that looks dicey under the present regs. Perhaps there is some clarification of these issues that I've missed, but I think a lot of people have too, umps included.

Cricket Life Forum said...

No, i dont think they Dream Of Electric Sheep LOL

John Halliwell said...

At the risk of being termed reactionary, I would be happiest if only on-field umpires faced with doubt were able to ask the third umpire to look at the evidence before giving a decision. A new signal for asking for this guidance would have to be devised (I prefer a straight-armed two fingered gesture, but again this is the reactionary in me). In order to ensure that captains, bowlers, coaches, cricket boards, commentators, crowds, et al, were not able to argue the toss over the eventual decision, Hawk-Eye, Hot-spot, and Snickometer evidence would come under the thirty year secrecy rule, be locked away in a fireproof Giles Clarke and buried under the Lords pavillion, and therefore denied to the world’s broadcasters (except the BBC who would never show it anyway having completely forgotten what cricket is). Sky Sports would go berserk over this and would, therefore, be offered special dispensation allowing Sir Ian to use the utterly shredded, completely knackered, expression: “He’s a big unit” up to three times every session.

I realise that my suggestion lacks the OB’s common-sensical, practical, and beautifully argued way forward, but with a proven 90% plus success rate, can we please return full powers to the umpires. I have to go now as I’ve reached an inspirational part of my library book: ‘Luddites - Victims of a Non-DR System? ’ Given out for encouraging smashing of machinery when all they said was: “Isn’t machinery smashing?”

Russ said...

OB, did people really expect absolutes? Perhaps I'm more familiar with the science underlying the technologies, but it seems remarkable that we ever thought we could get a precise answer from the technology, when every other technology (from runouts with gaps between frames, to replays of low catches) has failed to do so.

I've argued for some time that the aim should be to make the technology instantaneous and automated. Where an answer can be obtained with ~90-95% certainty, that should be conveyed to the central umpire before they make a decision, to augment their decision process.

The players ought to get no reviews, though the umpires could if they feel the camera can shed light on an unusual incident. If an umpire errs given a dozen inconclusive replays then so be it; at least we won't waste time waiting for decisions. At the moment we have the opposite: an inconsistent and marginal improvement in decision making, and a much slower game.

Russ said...

OB, did people really expect absolutes? Perhaps I'm more familiar with the science underlying the technologies, but it seems remarkable that we ever thought we could get a precise answer from the technology, when every other technology (from runouts with gaps between frames, to replays of low catches) has failed to do so.

I've argued for some time that the aim should be to make the technology instantaneous and automated. Where an answer can be obtained with ~90-95% certainty, that should be conveyed to the central umpire before they make a decision, to augment their decision process.

The players ought to get no reviews, though the umpires could if they feel the camera can shed light on an unusual incident. If an umpire errs given a dozen inconclusive replays then so be it; at least we won't waste time waiting for decisions. At the moment we have the opposite: an inconsistent and marginal improvement in decision making, and a much slower game.

Tim Newman said...

At the moment we have the opposite: an inconsistent and marginal improvement in decision making, and a much slower game.

Slower game? Erm, did you see Ed Cowan batting yesterday?!

diogenes said...

Just 2 points:

1/ in the case of run-outs and stumpings, the decision-making has improved immeasurably as a result of technology. Certainly, in the case of run-outs, I think umpires were right only about 10% of the time before they started to use camera evidence.

2/ the review system has increased the number of lbws obtained by spinners to front-foot shots. I suggest that only Bill Alley of the old school would ever have dared to give out some of the decisions that the spinners got over the last winter. Was there ever a case in Australia of a batsman given out lbw sweeping? I can recall Ian Chappell casting doubt on more than a few such decisions during Ashes tours in England.

The technology has given the umpires the confidence to start giving these decisions, which I believe is to the benefit of the game. There should be no reason why playing a spinner with a pad should be effectively risk-free for a batsman.

I would prefer the system not to be used on appeal by the mplayers but more along the lines it is used in rugby union. If the umpire would like confirmation of specific points, then he/she can ask to "go upstairs". And the decision should take into account the uncertainties inherent in any scientific or quasi-scientific measurement.

livescore said...

Umpires play a vital rule on every match and after entrance of technology in umpiring system it is more secure then before however there is more need to respect umpire decisions by every cricket player.

MM said...

http://www.cricketyorkshire.com/headlines/alan-igglesden/
Just came across this. Dear old iggy. Thought it may be of interest to you OB.

The Old Batsman said...

Cheers, think there is growing opinion that the umpires should consult the third before giving a decision rather than making it a gladiatorial moment of challenge.

Maybe not the two fingers, John, but the old Roman thumb waggling between up and down...

cost per head said...

that is a very good and interesting question and I have to say that I do not have the answer yet, but if I find it, I will come and tell you!

James Knight said...

The problem with leaving the technology to be used solely at the umpires' disgression is that they end up using it as they do with run outs and stumpings. Yes, the use of replays has massively improved decision-making there, but they now review everything upstairs. In the Champions Trophy even batsmen in by two or three yards were having their fate decided by the third umpire.

The risk is that with LBW's then falling into the same bracket we'd be having referrals at increasingly regular intervals because the umpire is scared of cocking up when he could use the technology instead.