AGB Cricket

CRAZY EYE

Chris Rogers, writing about his one Test in Perth in 2008, appears a little skeptical about Hawk Eye:

Four magical, suffocating days in Perth

My mood wasn't made any better when standing next to the umpire who had given me out - Asad Rauf. He told me that if he'd had the benefit of the replay, he wouldn't have given me out. It felt like a punch to the chest. But I wasn't the first person to get a dodgy decision, and I won't be the last. Still, I respected him for telling me that, even after Hawk-Eye had estimated the ball was angling crazily back to clip the leg stump.

HACK HAWK DOWN

I don't want to say "I told you so":

Umpires ask ICC to look in to Hughes' dismissal

Simon Taufel, the senior international umpire, has referred Phil Hughes' second innings lbw dismissal in the first Test between Sri Lanka and Australia to the ICC as a serious question mark against the accuracy of Hawk-Eye, the ball tracking technology. Taufel and the officiating umpires Richard Kettleborough, Aleem Dar and Tony Hill have have also sent the relevant footage of the incident to the ICC's cricket operations department.

The problem with Hawk and late jag has been evident for years, so why has an official investigation taken so long to emerge?

THE TOPIC ON EVERYONE'S ELLIPSE

The AGB has maintained three things about Hawkeye: one) it makes mistakes when there is sharp late movement; two) commentators who treat Hawkeye as infallible get right up my bugle; and three) Hawkeye was chosen because it is a colourful toy for the coverage.

With that in mind, it has been interesting to follow Kartikeya's correspondence with Paul Hawkins of Hawkeye:

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.

And over at Wisdens, Ed Craig writes about a curious incident during the Headingley Test:

Hawk-Eye’s one mistake shows it’s accuracy

At Headingley, Hawk-Eye made a mistake.

Mohammed Amir was lbw to Shane Watson first ball in Pakistan’s first innings. The ball crashed into his pads, he was playing no shot and Rudi Koertzen did the rest. Hawk-Eye showed the ball clearly missing off stump.

The ball was not missing off stump, in reality, but at the point of impact was in line with middle-and-off – it was out – Rudi got it right. The super-slow motion replays and Hotspot confirmed this as a befuddled Paul Allot and Michael Holding tried to make sense of Hawk-Eye, without ever accusing it of being fallible.

I don't particularly mind Hawkeye. The super slo-mo combined with the blue line down the pitch that says whether the ball pitched in line with the stumps would probably work just as effectively, but is obviously not as sexy from a broadcast perspective. Hawk gives a pretty fair indication most of the time, it deploys simple visuals, the background ads tell me what's coming up next on Nine so I know precisely what not to watch and it's a reliable straight man for the radio commentators.

Just don't sell it to me as the last word on decisions.

Happily, even Nine have now started to treat it with the appropriate level of skepticism. Gone is "Hawkeye said it was out", which was rivalling Ian Chappell's "Read the score in the paper tomorrow" for aggravating rote commentary. In is a nod-and-a-wink that Hawk is not always right; well, an embarrassed silence anyway, and a giggle from Ian Healy. Channel Nine are not complete fools. There is no point boldly stating that Hawkeye says "out" when the viewing public can plainly see it is "not out", or verse vica.

TELL IT HAWKIN'

Worth a squizz:

Problems with Hawkeye

Would Hawkeye be as persuasive if it said that "the ball will most probably reach the stumps at this point, but could also reach there at any point within this range", instead of saying, as it says now, that the ball would reach the stumps exactly at this point? Probably not.

It is a mistake to confuse the mathematics of Hawkeye with the technology of Hawkeye. How Hawkeye is presented to us is a function of the fact that the people who produce it want to make money. The certainty claimed by Hawkeye is function not of it's mathematics, but of it's marketing. In this post, I have shown that Hawkeye's graphics are misleading, and also tried to point out, why, in my view, they have to be misleading in order for Hawkeye to be persuasive.

The AGB has long maintained that Hawkeye is a colourful addition to the coverage and a reasonable guide as to whether an elbee decision is in the ballpark, as long as the ball is not well pitched up and sharply moving to either leg or off, either off the pitch or through the air. But the hard sell from the makers was far less aggravating than the hard sell from the Channel Nine commentators, who, at the merest hint of an elbee, deferred to Hawk as if it was the be-all and end-all in technological accuracy.

Fortunately the Nine plonkers have reined in their reliance on Hawkeye. Too often they were caught out; embarrassed, even. "Hawkeye will show us exactly whether that was out" has been watered down to "Let's see what Hawkeye has to say." There was only so many times Tony Grieg could say "See, that was out" when the ball was clearly missing the stumps. Not all Nine viewers are blind. We know why Nine was so keen on touting the integrity of Hawk - they had invested a shed-load of wonga on it - but sooner or later they were going to have to inject a note of circumspection. That's if "note of circumspection" is defined as "uncomfortable silence". Meanwhile, the ABC, who watched Hawk second-hand like the viewers and who had no investment in the product, would treat Hawk with outright derision: "I want some of what the Hawkeye blokes are smoking."

LEG BEFORE BOWLED

A typically thought-provoking post from The Old Batsman:

The power of suggestion

In a few days time, we will farewell the first decade in the history of Test cricket that has seen more batsmen dismissed LBW than bowled.

This revelation would seem to be supported by an answer Daryl Harper gave last year to a question about Hawkeye:

HARPER'S BIZARRE

I probably do give more lbws because of Hawk-Eye than I did ten years ago. My personal feeling is that the ball hitting the leg stump a decade back would very rarely be given out lbw because of the feeling that the ball was sliding down leg. But now we've seen so many times from the replays that the ball hits leg stump very often, and we give it not out much more than [the ones hitting] off stump. I've subconsciously made that adjustment.

I wouldn't say I have become bolder, but more accurate. I normally store the Hawk-Eye situation, and use that bank of information along with little details like the bounce, the pitches, and use that to make the right decision in the field.

LLLOOKED OUT

Welsh boffins have investigated Hawk-Eye:

We CAN be serious: Cardiff University disputes Hawk-eye's Wimbledon line call

Ahead of Wimbledon fortnight (23 June to 6 July), researchers from Cardiff University are advising that sports decision aids such as the Hawk-Eye system should come with a ‘health’ warning attached.

Hawk-Eye, and similar officiating tools which are used to supplement or replace decision-making by umpires and referees, are becoming an increasingly common part of televised sports coverage. But the new technology has also attracted its critics.

(Thanks to Alistair A. for the link.)

Pity the article is concerned with tennis. I would have preferred to read what it says about cricket - where it matters and where it is sometimes way out - but you need to subscribe to read the whole study.

Hawk works well enough in tennis, anyway, and I'm inclined to agree with its inventor:

Keeping the Hawk-Eye on the ball

Paul Hawkins said: "Anyone living in the sane world would say Hawk-Eye is sufficiently accurate."

"No system is ever going to have zero margin of error but at some point you have to make a decision. What are you going to do - go up to the net and start having a debate or asking fans to phone in and vote?"

"But isn't it wonderful that we live in such a developed world where university professors can spend their time researching something that doesn't need to be researched. Do they really have nothing more important to do?"

The 3.6 mm margin for error and the 5 mm International Tennis Federation limit are reasonable; especially as the Cardiff academics claim "the real figure could be either higher or even lower than 3.6mm." That's hardly a debunking of the earth-is-round variety.

No, Hawk's disputed margins for error in tennis don't worry me, my concerns lie elsewhere.  First, Super Slo Mo would work better. Track the ball, zoom in, show where it lands. Perfect. But what administrator could resist a CGI ball zooming down the court, the ball's meteor trail, and the fans' sound-effects? The main reason tennis uses Hawk is not because it works better than any other method, but because it looks sexy. Second, Hawk has been sold to us by commentators as the last word in accuracy. At least now the cricket commentators - even those at Nine, Hawk's traditional spruikers - express skepticism, if not outright contempt.

Mind you, an inventor settling for "sufficiently accurate" and sneering "Do they really have nothing more important to do?" at university professors sounds about as clever as a driver sneering "Haven't you got any murderers to catch?" at a traffic cop.

HARPER'S BIZARRE

Which Hawk-Eye has Daryl Harper been looking at?

'The on-field umpire is in the best position'

Do you currently use Hawk-Eye graphics as part of your preparation process? And do you think umpires have become bolder in giving lbws thanks to exposure to Hawk-Eye graphics?

Bowden: No. It is just a gadget for commentators to use to explain to the viewers what might or could have happened, not what would happen. It's all subjective, and lbws will always be a matter of opinion. One thing that is definite is, there is no doubt the on-field umpire is in the best position and has the best view to make an lbw decision

Only statistics will show if the percentage of lbws given by umpires has increased overall since Hawk-Eye's introduction. One thing is for sure, Hawk-Eye shows a lot more balls hitting the stumps than I believe. This does not really persuade me to give more lbw decisions. It's never been proven that it is 100% accurate. I call it the "Guessereye". Leg-befores will always be open to debate. There is no final outcome or definite conclusion. It's in the opinion of the umpire - who is in the best position, and not Hawk-Eye, to make the final decision. This will always be the case.

Harper: I probably do give more lbws because of Hawk-Eye than I did ten years ago. My personal feeling is that the ball hitting the leg stump a decade back would very rarely be given out lbw because of the feeling that the ball was sliding down leg. But now we've seen so many times from the replays that the ball hits leg stump very often, and we give it not out much more than [the ones hitting] off stump. I've subconsciously made that adjustment.

I wouldn't say I have become bolder, but more accurate. I normally store the Hawk-Eye situation, and use that bank of information along with little details like the bounce, the pitches, and use that to make the right decision in the field.

HI, TECH

For a moment I thought the bit below read "allowing each team to lodge up to three challenges per innings". Fat lot of good that would have done in Sydney. Naturally, I read it again. Three "unsuccessful" challenges works for me. (Who will be the first team to call for more?)

Australia wary of ICC proposal 

ALREADY under unprecedented pressure, Test umpires may have their decisions challenged by players under an International Cricket Council proposal for expanded use of technology, which has drawn scepticism from Australian coach Tim Nielsen. 

Cricket Australia is reserving its judgement until director Mark Taylor, a member of the ICC's cricket committee, presents the detailed recommendations upon his return from Dubai. 

But Nielsen warned that such a system, allowing each team to lodge up to three unsuccessful challenges per innings, could slow down the game and create more intense conjecture about decisions.

Similarly, I'm not philosophically opposed to a predictless Hawkeye.

It is expected umpires would have at their disposal "Hawk-Eye" to judge whether a batsman has been struck in line with the stumps in leg-before-wicket decisions, stump microphones to determine whether there is an edge in caught behind appeals, and "Super Slow-Mo" for close bat-pad decisions. 

They could not, however, use Hawk-Eye to predict the trajectory of the ball in lbw calls. Nor could they use "Snicko" or the infra-red "Hot Spot" technology used in television broadcasts.

As we are all aware around these parts, the problem with Hawk, apart from it being a sly excuse for Nine to plug ACA, CSI and other upcoming programs, is its inability to accurately predict trajectory after the ball has bounced; especially when the spinners are on. But using it, for instance, to see if the ball has pitched outside leg is definitely a valid application. Not that you couldn't do that with super slow-mo or freeze-frame. I generally try to avoid getting between Nine and a good gadget.

I wonder why the umpires will be allowed to use stump microphones and won't be allowed to use Hot Spot. I'm not wild about Snicko.

Anyhoo, what's Tim Nielsen been smoking?

Nielsen warned that such a system... could create more intense conjecture about decisions.

A crack research team comprised of Isaac Newton, Louis Pasteur, James Cook, Hercule Poirot, Phileas Fogg and every single winner of The Price Is Right would struggle to unearth any more "intense conjecture about decisions" than existed last summer.

BIRD OF NAY

A little cricket then: Hawk-Eye.

Regular readers here at the AGB are a largely sensible bunch functioning with the majority of their faculties intact including almost fully operational bullshit detectors. (Notable exceptions are ... well, you know who.) We have been known to spot the odd rook and one of the regular rooks raised here is Hawk, cricket and tennis' toy du jour.

There is nothing Hawk can't reveal that can't be better shown by a blue pitch-mat between the stumps, combined with Super-Slo-Mo, high resolution footage. The ability of the viewer/umpire to accurately extrapolate the trajectory of a ball using the latter is patently superior to that of the former where frequent errors are revealed, especially when the ball moves late. (Troll through the archives for examples.) As I see it, the main reason Channel Nine in particular and pundits in general defer to Hawk is because: a) it gives sexy graphics to talk about rather than tactics; and b) it affords yet another opportunity for TV producers to plonk an advertisement into the background.

The blatant errors should, of themselves, give rise to healthy scepticism. And if they don't, Tony Grieg's embarrassed silence or Richie's famous "hmmmm" should at least set bells ringing. Scepticism, however, seems in short supply at CricInfo.

A threat or an advantage?

The technology: Six cameras, usually placed at long-on, long-off, third man, fine leg, and two square of the wicket, operate at 120 frames per second (the regular television cameras operate at 25). Three of those cameras are used to capture the path of the ball from each end, and the pictures from each of those cameras are fed into a central computer. Taking into account the sideways movement and the bounce of the ball, the computer then extrapolates this information to predict the path of the delivery. Because processing of the data happens as soon as it is fed in, the output is available almost instantaneously. Together with the pitch mat - the strip drawn through the middle of the pitch from one set of stumps to the other - Hawk-Eye accurately indicates where the ball hit the batsman (in line with the stumps or not) and whether or not the ball would have gone on to hit the stumps. [My boldening for emphasisness.]

In the words of Paul Hawkins, the brains behind the technology: "Hawk-Eye does not try to predict the path of the ball after the bounce. Instead, the Hawk-Eye cameras track the ball both before and after the bounce, so the correct trajectory of the ball leaving the wicket is determined. Hawk-Eye simply observes and then calculates the actual trajectory of the ball. Whether the cause of this trajectory was due to atmospheric conditions, the wicket, or the ball hitting the seam is irrelevant from a Hawk-Eye perspective. Hawk-Eye just tracks what happened - it does not try to predict nor to answer why it happened."

Accuracy: According to Hawkins, in most cases Hawk-Eye's output is accurate to within five millimetres in predicting the path of the ball. The accuracy levels are highest when the ball has travelled a fair distance after pitching, but even when the point of contact is very close to the pitch of the ball, the accuracy levels are still within 20mm. [More of my bolderation.]

A biggish article containing oodles of room for doubt, but other than a cursory nod to some slight inaccuracies which the article appears to concede are workable tolerances, there is not one acknowledgement of Hawk's many howlers. Surely even a CricInfo scribe can see when a ball is comfortably missing leg stump, for instance.

Why indeed, on the 13th of June, would CrookInfo launch into a long spiel detailing the merits of Hawk? "Hmmmm." Well, to pilfer a quote from noted American seam, swing, and more pointedly, spin bowler, Victor Kiam, "I liked it so much, I bought the ... thingo." This from June 14th:

Hawk-Eye bought by Wisden Group

Hawk-Eye, the global pioneer in ball tracking technology for sports broadcasting and officiating, has been bought by the Wisden Group, the owners of Cricinfo.

Mark Getty, a director of Wisden Group, said that the acquisition was a coup for the group which would strengthen its presence in the global cricket market and provide an entry into tennis and other major international sports.

(The Getty of Wisden. Thank you - no need to laugh.)

PAWK PYE

You may have noticed Hawk Eye was telling whoppers again. May? How frightfully civil of me. What I meant to write was "Hawk Eye is a stinking fiasco".

Darren Powell was given out LBW to McGrath but was clearly not out, with the ball clearly sliding low down the leg-side. Even Tony Grieg called it and was surprised by the decision. McGrath stiffled his appealed. Yet stupid Hawk showed the ball comfortably hitting leg-stump half way up.

It wouldn't be so bad if the Nine plonkers treated Hawk with due contempt, but the duplicitous dogs defer to it as final arbiter thus imbuing the fucker with an undeserved degree of accuracy. (This has to be a specific production directive. Contrary to appearances, the commentators aren't complete fools.) At least Heals and Mikey, who replaced Tone and Tubbs, shit-canned the decision, implicitly criticising Hawk.

Cricinfo's live commentary was also a load of arse:

McGrath to Powell, OUT: full length delivery on the middle and leg stump, Powell tries to work it away onside, gets hit on the pads right in front of the stumps.

PS: Can't remember which umpire fired Powell, but it was a shocker so let's just assume it was Bowden. Usual suspects, and all that.

RUBBISH

Hawkeye is a farce.

Every time there's an LBW appeal the Channel Nine commentators defer to it as the be-all-and-end-all when, in fact, every man and his blind dog can see different.

Many instances show the ball bouncing about a foot over the stumps when in reality it would be lucky to bounce over the batsman's shin.

It's ridiculous!

Channel Nine, cut the crap!

STEVE'S LEAVE

Tony Greig: "That was plumb, let's look at Hawk Eye. Well, I thought it was out."

HAWK EYES & PORK PIES

I get a strong desire to introduce Mr Foot to Mr TV - violently - when the Channel Nine commentators go with their Hawk Eye lie.

Matthew Hayden was NOT out. The ball turned from stump-line and would clearly have missed off stump. Yet the commentators went straight to the Hawk for confirmation that the ball would have hit off.

Lo and behold - with a "convenient" curve back - "The ball would have hit off." Rubbish!

PS: You just knew the only way Hayden was going to get out was via a dodgy decision.

PPS: Why the hell send Gilchrist in at four?!?

PPPS: This was supposed to post about six thirty, but the Drivel Warehouse was down. On the up side, I've watched The Natural, Men In Black II, The Dead Zone and now I'm watching Fargo.