The man was safe, and I knew he was safe before the first replay was shown. When the ball beats the runner to the base by as much as Mike Lansing's relay beat Al Martin to the plate last night, and the umpire -- especially an umpire in perfect position, as Mark Wegner was -- still calls him safe, you know he's absolutely certain the tag was missed. And every look from every angle seemed to bear that out.

As for whether or not Jason Varitek should have been ejected, that's a different story. I didn't see any contact, Varitek insists there was no contact, and anything that might have happened was at most accidental, incidental and slight. You'd hope a major-league umpire would understand that if Varitek happened to brush him with his glove while arguing a call -- a call that he had to know was going to be controversial, since the ball was there so far ahead of the runner -- it should fall into the no-harm, no-foul category. Of course, you'd also hope that a major-league umpire would understand that a 77-mile-an-hour curveball that happened to break too far inside was no cause for a beanball warning, or that a game in which four batters were hit by pitches -- two in the head -- might warrant an ejection or two. We've come a long way toward gaining some quality control over the umpires in these post-Richie Phillips times, thankfully, but this last week shows that we still way have a ways to go.

But the most interesting aspect of that play was what happened after Varitek was tossed.

Tomo Ohka, who'd retired 14 of 18 batters prior to the ejection, gave up hits to four of the next five men he faced -- including two monstrous home runs -- before the Sox could hustle him out of there. What had been a 1-0 game when Varitek left the field was 5-1 when Ohka departed.

And all I could think was: Is the catcher that important? Is the difference between Varitek and Scott Hatteberg such that Ohka would go from cruise control to head-on crash in the blink of an eye? How could that be?

It couldn't be the pitch-calling; Williams/Kerrigan call the pitches. (Or at least I think they do. Certainly I see Varitek peering into the dugout before every delivery, watching Williams go through his touch-the-cheek/touch-the-nose/touch-the-other-cheek routine. I assume that's pitch-calling, since they do it even when no runners are on base.) It couldn't be communication; except in emergencies, the pitcher and catcher barely speak on the field. What is it? Just coincidence, maybe?

Neither Kerrigan nor Williams seem to think so. While they didn't shed a whole lot of light on the subject in their postgame comments, they did indicate that, well, there's a there there.

"You always want to maintain the flow with the pitcher and catcher," Kerrigan said. "You don't want to disrupt that. The pitcher has a feel for what the catcher is going to call, and the catcher has a feel for what is working for the pitcher."

"We don't need our catcher getting thrown out," Williams said. "(Ohka) had continuity going with the catcher there . . . He was pitching very well."

Williams also admitted that "Maybe (Ohka's meltdown) had nothing to do with anything", and Ohka himself said Varitek's disappearance "didn't have any affect". It could be something as simple as the fact that the 19th batter began the Mariners' third time around the order, and while Ohka might have been able to fool them once or twice he couldn't do it three times. It's one of the many things in baseball that seem unquantifiable.

Or is it? The Red Sox' ERA with Varitek catching this year is 2.94. With Hatteberg, it's 4.21.

The 2001 season is a ridiculously small sample size. What were the numbers in 2000?

Varitek: 3.96. Hatteberg: 5.02.

Now, obviously, we're getting into a very complex area in which I have neither the time nor the expertise to examine with any degree of sophistication. Keith Woolner has done some marvelous work on the subject of a catcher's influence on pitcher performance, and I recommend you to his original article on the subject, which appeared in the 1999 Baseball Prospectus, and his 2000 followup. His conclusion -- "if there's a true game-calling ability, it lies below the threshold of detection" -- is what I suspected going in, and I wasn't surprised he reached that conclusion. And it's something that, intuitively, we already believe. Certainly, Varitek's advantage over Hatteberg in the category of catcher's ERA could be nothing more than the fact that he catches Pedro Martinez 90% of the time . . . and if I caught Pedro Martinez, I'd lead the league in catcher's ERA.

I hate articles that go around in a circle, take you nowhere, and wind up right where they started. That, I'm afraid, is exactly what I've done here. But Ohka's collapse, and Kerrigan's and Williams's postgame talk of "continuity" and "flow" between pitcher and catcher, was too ripe a piece to leave hanging on the tree.

We study and we ponder and we discuss, and we think -- in that arrogant human way of ours -- that we have the answers, that we understand it all, that we know. And every so often, something comes along and shows us the awesome limits of not only our knowledge, but the knowledge of the professionals themselves. The difference is, at least the pros (some of them, anyway) are enlightened enough to admit it.

Baseball. You gotta love it.