One would expect that all the tennis courts at the U.S. Open, the most attended tennis tournament in the known universe, are the same size. They are not. The courts are not level, either, at least not all of them — one need only look at them after a soaking rain to witness the uneven puddles that form.

“You have to measure the line and the height of the court,” Paul Hawkins, the inventor of the Hawk-Eye line calling system, said in a recent interview. “All courts are not flat. You have to measure the contour. The highest part of the court could be two centimeters higher.”

Small differences from court to court, perhaps as much as five centimeters between service boxes, are commonplace. Hawkins knows this only because his system of cameras and computers must have a precise map of the court in order to work properly. So Hawkins and his colleagues at Hawk-Eye Innovations Ltd. measure, as Hawkins learned to measure while he worked on his Ph.D. in artificial intelligence.

Hawkins isn’t trying to create a scandal by saying this, or to suggest that somehow the courts at the U.S. Open are any different from thousands of others around the world, including those at the other three Grand Slam events. The point is that what often seems like reality in tennis, which adores tradition and shies away from change, is very often not reality at all.

Still, many people do not believe him.

It’s been more than a year since the USTA began using instant replay, first in a series of smaller tournaments, and then at last year’s U.S. Open. This year, the Australian Open and Wimbledon used it, too. Players have accepted it without much of a fight, though if you corner them, almost all of them will admit a little skepticism. A few have been more vocal, particularly Marat Safin (he’s challenged more calls than anyone on the tour since Hawk-Eye was instituted, and has been successful 35% of the time) and Roger Federer, the implacable world no. 1 who came as close as he can come to throwing a fit during a tense Wimbledon final against Rafael Nadal.

It was 30–30 in the third game of the fourth set, and Federer seemed to be losing the vice grip he’d had on Wimbledon the last four years. Nadal hit a deep stroke that almost everyone — Federer, Nadal, the chair umpire, the crowd — saw as long. Hawk-Eye disagreed and called it good by a hair. Federer became flustered. He muttered. He asked the chair umpire to turn off Hawk-Eye. Nadal won the game and the set before Federer prevailed in the most taxing Grand Slam final of his career.

“The heart rate went up a bit, definitely,” Hawkins said when asked what it was like to watch the best player in the world, perhaps the best player ever, publicly challenge his brainchild.

Federer, who won his opening match at the U.S. Open on Monday, isn’t any happier about the system now that Wimbledon is in the past.

“Wrong question,” he said, flashing a grin, when asked about it. “No, I’m not happy. I will never be. Nobody spoke to me after what happened at Wimbledon. I don’t care honestly.”

Television commentators calling the match didn’t exactly race to Hawkins’s defense (they suggested the call had been botched), and Hawkins couldn’t immediately defend his technology from the charges, largely because his system’s high-speed cameras, which feed images into a computer that calculates the trajectory and landing point of a given ball, capture so many images that they can’t save them all — it’s a matter of disk space.

But after the tournament ended, he orchestrated a campaign to defend the system and that call. Here’s the brief version: Television cameras capture about a third as many frames a second as Hawk-Eye cameras, and television replays tend to show the ball as it leaves the ground after a bounce, when it is moving more slowly. However, a ball can graze a line and slide before it leaves the ground, so it might look long upon liftoff.

Hawkins can’t guarantee that the call was correct, as Hawk-Eye does have a range of plus or minus three millimeters. Hawkins claims the contested ball, by his calculations, was almost certainly “one millimeter in.” He added, “In all the tests we’ve done, we’ve never gotten a call wrong.”

Whether Hawkins is correct or not, I have sympathy for him, as he is essentially a man of science in a sport that trails other major sports when it comes to measurable data and apples-to-apples comparisons (the women’s tour still does not track basic statistics, except at the majors, like the number of aces its players hit from match to match).

Besides, it’s also clear that Hawk-Eye has done a great deal of good for tennis — fans enjoy it and some awfully bad calls have been overturned, all without slowing play — without any of the harm that people were worried about (the most attractive negative argument, for me, was that the technology would put an end to entertaining histrionics, but Federer’s moment at Wimbledon, and dozens by Safin and others, have proven that incorrect).

Since the introduction of Hawk-Eye last spring, the men on the tour have challenged 2,262 calls as of this month’s Masters event in Montreal. Hawk-Eye overruled 859 (39%) of those challenged calls. In James Blake’s first-round match yesterday, Hawk-Eye overturned an out call on a slow lob that just ticked the baseline — calling lines, even on slow shots, is difficult to do. (Overall data was not available for the women’s tour, but so far this year, 33% of challenges by the women were proven correct.)

If Major League Baseball announced that its umpires had made incorrect calls on balls and strikes 40% of the time, players, commentators, and owners would take up arms. Yet in tennis, where a single call is far more important than a single ball or a single strike, we shrug.

In 10 years or so, perhaps we’ll learn that Hawk-Eye had a fatal flaw. But until then, why not champion a tool that, from all the evidence we have so far, clearly makes the sport’s already skilled umpires that much better?

Mr. Perrotta is a senior editor at Tennis magazine. He can be reached at tperrotta@nysun.com.