What's a few tenths among friends?

A couple of big-market teams had a close one last night, so today all kinds of people are fired up about tenths of seconds.

Matt Moore did a nice job of showing what happened ... how the Celtics' Paul Pierce made a shot with less than a second left, but probably more than the four tenths of a second the Knicks were allotted to respond.

The Knicks did respond, but with a that-would-have-been-fun Amare Stoudemire 3-pointer just slightly after the buzzer.

Knick fans are upset, because their team might have had a little more time, in a perfect world, or so it seems.

Here's where we encounter two long-term concerns about NBA crunch time.

The first is that instant replay can only be used when there are certain triggers, for instance to determine if a shot was a 2 or a 3, or if a player was fouled. Pierce hitting a jumper is no reason for the referees to huddle on the sidelines and tinker with the clock. To me, every fan at home can review that play, and I can't imagine why the NBA would want the referees to be in the minority of those who, with the game on the line, don't know what really happened.

A bigger concern is that even while Moore shows the clock was slow to stop in a crucial moment, the scoreboard operator nevertheless behaved perfectly.

The game clock shows time in tenths of seconds, you see, but it's still run by humans for whom such increments are too small to manage. Scoreboard operators say, and scientists agree, that a human who can see something like a made basket, and push the button to stop the clock within three tenths of a second is pretty quick. Two tenths is excellent. Four tenths happens. We're talking about the time it takes for a signal to get from a human's eye, to their brain, and back out to the finger that presses the button.

Meanwhile, one of the true crimes of running a scoreboard is to anticipate -- you don't want the scoreboard guy starting to stop the clock when Pierce's shot looks like it's going to be good. If it comes off the rim as a live ball, that clock needs to keep running.

So, the little delay that Moore points out ... that's going to keep happening until the robots take over.

One day, I'm sure, it really will all be managed by computers and measured accurately to the hundredths, as already happens in sprinting and swimming. But until then, I think we're just going to have to accept that there will be a few tenths lag time between when the clock is supposed to start and stop, and when it does. And maybe that doesn't matter very much: The delay after Pierce's made basket presumably approximates the delay between when Stoudemire catches the ball and the clock re-starts.

We go to the arena to cheer human performance, That little delay is just one more part of it, even if it can frustrate our sense of fair play in slow motion replay.