What timeouts do to crunch time
February, 15, 2012
By Henry Abbott
Jeremy Lin's shot heard 'round the world came on an unusual crunch-time play: it was late in the game, it was for all the cookies, and yet it was not preceded by a timeout.
That was certainly exciting, wasn't it? It's also exactly what I have been pining for: continuous crunch time play.
Mike D'Antoni was all smiles afterward, saying Lin's "too good to even call a timeout. That makes it easy for a coach that's able just to trust the point guard."
The coach explained that he didn't want to give the defense an opportunity to get set.
I suspect this is brilliant strategy, and no small part of the reason the play worked. Jose Calderon gave Lin a ton of room. After a timeout, it's possible he wouldn't have even been guarding Lin. It's possible the Knicks would have had a hard time inbounding. It's possible Lin would have been trapped, or subject to some other scheme.
But instead, the Raptors had to play the players they had on the floor, with whatever direction could be screamed from the sidelines.
This time, it didn't work.
Last week, when we launched HoopIdea, I complained about the amount of dead ball time at the end of close games. David Thorpe argued on the NBA Today podcast that coaches needed those dead balls to put their imprint on the game, and to raise the quality of play.
I have seen all kinds of data about how teams perform in crunch time. At the end of close games, offenses generally perform poorly. Which makes me wonder what those coaches are really accomplishing. Teams score well all game long. Then they get the special coaching secret sauce and can't buy a bucket. How useful could that coaching insight really be?
I'm intentionally being a bit dense on that topic. I know coaches can draw up good plays. I'm just saying it's sure not magic sauce, 'cause teams have a hell of a time scoring.
My best guess is that, in fact, coaches are really effective in timeouts -- only both coaches, and the one without the ball even more.
That's right, I'm theorizing timeouts are best used by the team that is banned from calling them: the team on defense.
Timeouts stop the ball and set the defense. They also allow substitutions and real contingency planning.
In general, in any sport, the offense has the advantage of initiative. The defense has to react to whatever the offense decides to do. The offense has a knowledge advantage, because the defense is ignorant as to what the offense is going to do.
If nobody knows what's happening, even a feeble surprise can carry the day. For instance, I'm right-handed. But a few times in my life, for fun, when playing with new players, I have started a game with a ton of left-handed dribbling. Then when I want to make a really strong move, sometimes I have been able to get defenders to bite on fakes to my (in reality weak) left hand, while finishing easily to the right.
Once everyone knows I'm a rightie, though, it's fewer surprises all around and there advantage: defense.
The same goes for most knowledge -- the more everybody reads the scouting report, the less valuable the element of surprise, and the better the defense. Offensive coaches can't use a timeout to make people better shooters, or to make them jump higher. But they really can use that time to decide how five players will work together to cope with the most likely scenarios. If they run a high pick-and-roll with less than five seconds left, we'll trap the ball-handler. If they isolate, we'll double on the dribble. That kind of stuff drastically limits open shots, and really works.
Freewheeling and chaotic plays, like Lin's last night, very often end in buckets, especially in a league where everyone can score (I'm saying that with confidence today because Ben Wallace just hit a 3). A weak point anywhere in the defense can quickly become a bucket.
So it makes sense that the team that would benefit the most from a coach's instilling order would be the team without the ball.
(All you quant experts out there, please let me know if you have ideas about how to put my theory to the test. Does defense improve out of timeouts?)