There have been multiple games recently in which a team could hold the ball for the final shot in a tie game. Both the Celtics in game 1 against Miami and the Raptors in game 6 against Boston shot a pull-up 3 in these spots. These were heavily criticized by lots of people. I think that criticism comes from a place of not understanding the reality of both how hard it is to score in these spots and what the proper strategy should be in these spots. The biggest mistake you can make in this situation is to not score and leave the opponent a chance to win the game in regulation. As we have seen in these playoffs even leaving 0.5 seconds left on the clock is enough for your opponent to get a shot off and win.
The most important things to do in these possessions are to leave 0:00 on the clock after the shot and to get a relatively clean shot off. Holding the ball and taking a pull-up 3 accomplishes both of these goals. And while it may seem like the shot is not a good shot, given the context of the situation it’s not so bad. There have been ~1100 possessions in the NBA over the last 20 seasons that started with 10+ seconds remaining in a tie game where the team with the ball can hold for the last shot. The FG% on shots in the final 2 seconds of these possessions is ~25%. All of a sudden a pull-up 3 from Jayson Tatum doesn’t look so bad anymore. Sure you can nitpick that he could have gotten a little closer, but Jayson Tatum shot 43% on 3s off 3+ dribbles this season.
I’ve seen people say they want to see players drive to the basket or run some sort of play. I get it, doing those can result in you getting a better shot. But the risk/reward calculations here are different than other possessions over the course of the game. You aren’t trying to maximize the number of points you score per possession, you are trying to maximize your chances of winning the game. Allowing your opponent to get the ball back with time on the clock, especially if the game is still tied (and you are more likely to miss than make in these spots - when a team can hold for last shot in a tie game but shoot with over 5 seconds left the FG% is 38%), is a huge mistake and can’t really be made up for unless you can get an open layup. It is much harder to time things so that you get a shot with the ball in the air with 0:00 on the clock when you drive to the basket or run a play, so even if you do score, which is far from guaranteed (the league average on contested shots within 10 feet of the basket was around 54% this season, and on the final possession it is probably even lower) you still give your opponent a chance to tie or win the game. As we saw in game 3 of the Boston Toronto series, even when you do score, it does not guarantee a win if there is still time left on the clock.
Any time you pass the ball or drive into traffic you also increase your chances of turning the ball over. The defense may collapse on you and you may not even be able to get a shot off. Teams have turned the ball over on ~9% of these possessions in the last 20 seasons and not gotten a shot off on another 3-4% of possessions. Last week we saw Kemba Walker drive to the basket and clearly get fouled at the end of regulation in game 6 against Toronto, but no foul was called. It’s hard not to notice that refs tend to swallow the whistle in these spots. This further disincentivizes driving to the basket. A shooting foul has been called on ~5% of possessions when the game is tied and the team can hold for the final shot. This makes scoring on a drive harder than it is at other points in the game.
Some people may say you want to shoot earlier to give yourself a chance at an offensive rebound putback. Well a missed shot is more likely to be rebounded by the defense so it also gives them a chance to get a rebound, call timeout to advance the ball and get a shot off at the buzzer.
I should note that if the other team has no timeouts and can’t advance the ball the arguments for doing something other than holding the ball and shooting a pull-up have more merit. If a team can’t call timeout to advance the ball then it becomes a much smaller mistake to leave time on the clock to get a shot off because their chances of getting off a decent shot drop (this is a reason why teams should be saving their final timeout in case they need to advance the ball in the final few seconds, but that’s a rant for another day).
Holding the ball and shooting a pull-up 3 takes away almost all of the risk. It’s really hard to turn the ball over and you have far more control over when you take the shot. Sure it may increase your points per possession to do something else, but I’m not certain it increases your win probability. It may not seem like a good shot on the surface, but given the context of the situation, waiting until the final 2 seconds and having a good shooter pull up for 3, while it may not be the best you can possibly do, is a perfectly fine shot in a tie game.