Big Brother is watching Boris Becker: How many statistics can tennis fans stomach? Computers now tell us more about a match than we ever need to know, says Susan Watts

IF YOU visit Wimbledon this week, you may notice the occasional pair of spectators hunched over what looks like a sawn-off computer keyboard. One will be tapping away furiously, the other glued to the match.

Ten such couples are roving Wimbledon this year. They are not quite the secret tennis police, but they are part of a Big Brother-style operation that is logging the form of this year's players. The 'data entry' teams are the only visible and human link in a vast web of information technology ticking away beneath Wimbledon's strawberries-and-cream gentility.

The statistics these couples are collecting are fed into a hidden network of more than 70 computer systems. As the match proceeds, they plot how the players win each point. Often the computer knows who did what before the umpires have had time to make their own notes.

The aim is to bring the most accurate, up-to-date match statistics possible to players, officials, spectators and the press at the tournament. But this service is also tailored to television commentators. For those who like to moan about the inanity of the tournament's coverage, it may sound like a good idea to put more information at presenters' fingertips - but it could make matters worse.

The more data available, the worse may be the quality of the commentary. In America, for instance, presenters scream non-stop statistics to viewers, who apparently love it. British coverage has resisted such extremes, but slowly we are moving in the American direction. Every year, viewers are bombarded with more data: how many matches a player has won on grass against their current opponent in the past six years, in how many sets and with how many aces; that sort of thing.

Thanks to the small army of data gatherers, the statistics come in live. At any point during a match, the television and radio commentators using the system can trace point-by-point details, analyse trends in a player's game and tell viewers exactly which shots are swinging the match.

IBM has pulled all the disparate sources of information together into a 'multimedia' system that makes full- motion video, tables of statistics and computer graphics available on a single computer screen. And it is accessible even to computer illiterates: a few prods at the touch-sensitive screen and up pops the information.

The system provides a live feed in the centre of the screen, taken directly from cameras on court. This is surrounded by a menu listing related services. Presenters can delve among past match data (stored in Florida but downloaded to a local 'store' and refreshed a couple of times a day), summon up the day's order of play, or order a player's mini-profile. The graphics that spring up at the bottom of our television screens, recording double faults or successful first serves, come from the same network.

IBM is the official supplier of information technology to the All England Lawn Tennis Club. John Taylor, project manager for the Wimbledon system, says most club members, players and commentators have embraced it, whereas there was controversy over the most visible piece of technology to arrive at Wimbledon in recent years, the Cyclops, or 'magic eye', system that watches the lines on court.

None of IBM's 50 permanent staff working on the project, and 20 casual staff collecting raw match data, is less than a county-level tennis player. They need to be able to distinguish immediately between a volley and a smash, and to discern the exact type of stroke being played.

Is it all worth it? There are really only two sets of match statistics worth knowing. At Wimbledon, the grass surface means the server has a tremendous advantage, so it is useful to know: one, the number of break points a player sets up; and, two, how many they convert.

Last year, Jeremy Bates had a sensational Wimbledon. The IBM system picked up trends in his game that otherwise would have taken hours to spot, and only then by scrabbling through the mass of data collected by umpires. The reason for Bates's sparkling performance, the analysis showed, was that he saved a far higher than average number of break points against him.

The other useful statistic is the percentage of first serves a player gets in. Boris Becker, for example, has a big first serve and powerful forehand, so the percentage of first serves he goes on to win is a strong clue as to how well his game is working.

Almost every match is won by the player who makes the fewest errors, not by the one with the jazziest passing shots. According to Charles Arthur, former associate editor of Tennis magazine, the untrained viewer will find it hard to keep tabs on this. 'They will remember the passing shot at full speed and full stretch where the player did a somersault and came up standing, but they will forget when the same player hit three returns into the stands.'

Part of what made Dan Maskell great was his ability to track the underlying trends of a match and highlight the scores only at key points. A good commentator is tuned in to the ebb and flow of a match; live data should reinforce their observations.

Statistics alone can be a dry alternative to a commentator's understanding of the human aspects of the game. Mr Taylor cites concern over the public's obsession with one of the statistics the system provides: speed of serve. 'People are getting hung up on high-speed serves, when the Lawn Tennis Association wants to de- emphasise the power game,' he says. Speed of serve tells viewers little about its effectiveness; a slow serve that is well placed or surprises the opponent often wins the point.

Another danger is that access to a vast data bank may tempt presenters to chatter when viewers would rather they shut up; good commentators know when to keep quiet. With the stereo sound on modern televisions, it is possible to separate commentary from the ambient noise of the game and feed it into one channel only. Viewers could then cut out the televised commentary altogether while retaining the atmosphere.

Digital television, when it eventually becomes available, will give the audience even greater control. Viewers may even be able to call up their own version of IBM's Wimbledon system; the company is hoping to strike just such a deal with a firm that specialises in interactive television systems. If the system became truly interactive, viewers could even decide from which camera they want to watch key points.

Technology clearly has a role at Wimbledon, but statistics will never be able to tell the commentator how well a player will perform when the mental pressure is high. Nerves may cause players to hit comparatively easy shots into the net, and a good commentator knows almost instinctively who is gaining the mental edge.

In last year's final, Goran Ivanisevic was serving at 4-5 down to Andre Agassi in the fifth set, and had to hold his serve to stay in the match. The commentary had highlighted Ivanisevic's great serve and how few people had broken it, yet he hit two double faults to trail 0-30. He recovered; but on match point, when Agassi hit what should have been an easy return, Ivanisevic put the ball into the net and lost the match.

Wimbledon will always be about how players - and commentators - peform in unique situations. But all the technology in the world will not create another Dan Maskell.

(Photographs omitted)