Some Thoughts on NBA End of Game Situations

3 things I think teams get wrong

Before anyone yells at me that there are teams that do get these things right and there are rare situations where these might be the right things to do, let me just say that I know that and I’m just speaking in general terms.

Not being aggressive enough seeking out 3s instead of 2s when down late.

I have written about this before here. I also created a tool to calculate the equivalent 2pt and 3pt percentages, in terms of impact on win probability, in late game spots. I won’t go into more detail than I previously have, but, for example, if you are down 4 with 20 seconds left and you don’t shoot a 3, unless that 2 is a wide open layup or dunk, you can’t really argue that you are trying your hardest to win the game.

Not saving your last timeout to advance the ball in the final seconds

My belief is that a timeout helps the defense more than the offense. I think the biggest benefit for the offense from calling a timeout is advancing the ball (depending on who is on the floor, being able to make substitutions could also benefit the offense). For that reason I think teams should be more willing to save their last timeout so they don’t end up in a situation where they are down 3 with 2 seconds left and inbounding from under their own basket.

Calling a timeout off a live ball rebound to prevent a transition/semi-transition opportunity

In these spots everyone already knows that you are probably just going to get the ball to your best player and either have him iso or have someone come set a screen for him. Why not just let the possession play out? Your opponent may have a weak defensive player they would take out if you called a timeout. There may be cross matchups that you can exploit that would go away if you called a timeout. Why not trust your players to make a play? The least efficient possessions in the NBA come off dead balls so why would you willingly put yourself in that spot with the game on the line?

One thing I think teams get right and fans get wrong

Holding the ball to take a pull-up 3 as time expires in a tie game

I wrote about this in more detail here and my thoughts on this haven’t changed.

Something I have been thinking about recently

Being more willing to give up 3s when down in the final few minutes

If are trailing late and are in a spot where you can’t really afford to give up any more points you should be fine with giving up open 3s, even to good shooters. An extreme example would be if you are down 2 with 30-35 seconds left. Your opponent will probably try to kill clock and get a shot off with around 10 seconds left. In this spot giving up 2 points or 3 points are pretty much the same thing, you have all but lost either way. In that situation this is a rare case in which field goal percentage is actually a meaningful stat. It is better to give up a 38% three in this spot than a 50% two, even though the three has a higher points per shot because what matters is the percentage of the time they miss. In practice this might be something that is harder to implement because it could require players trying to make plays that go against their instincts and the team’s defensive principles at all other points in the game.

End of game possessions are pretty different than other possessions. The focus shifts from trying to maximize your expected points per possession to trying to maximize your win probability and that should result in a shift in what shots are the best and worst shots to take.

See also