We’ve all been in it, that conversation where you’re debating, “If you have one guy to take the shot to win the game, who would it be?” And that usually dissolves within seconds into a Kobe Bryant debate. So how do you resolve these discussions? And how do you know who you’d pick to take the shot?
Before we start getting down to who did what, let’s take a look at some of the raw facts because frankly, we need to adjust our understanding of what “good” is when we’re talking about last-shot situations.
Even how you define such things can be problematic. I like to go with the thinking that any shot with 24 seconds or less remaining on the clock to tie or win is one with the game on the line. There’s no assurance that you will get the ball back. Sure, it’s possible you will, but it’s also possible that you don’t get another chance.
Ergo, there can be multiple “last shots” in a single game, but that’s not an issue to me. Making a last-second shot isn’t any less clutch if it’s followed up by an even more clutch one. The mentality of what went into making the shot isn’t retroactively retracted by what follows.
So for example, this shot by Reggie Miller was still pure hero, even if he did regret the bow before it was all over:
Using the data from the Play Index+ at Basketball-Reference.com, I looked at the 573 shots with the game on the line. The home team took 286 of them and the road team 287, so it’s a nearly perfect split. What’s surprising is that visitors shot an effective field goal percentage of 35.9 percent, which was better than the hosts’ 32.2 percent.
In all, just 167 of the shots found the net, and 56 of those were from deep. The NBA, as a whole, shot 34.0 percent in such situations. It’s darned hard to score those points, and there are good reasons why. Typically, the defense is set. It’s coming out of a timeout, so the best defenders are on the court. The result is that the looks aren’t as good and come from less-than-ideal distances.
Look how last-shot opportunities stacked up by both time and distance:
This illustrates that teams go for the last shot, but not why. There’s a bit of a misunderstanding about why teams and players go with “hero ball” in such situations. Isolation plays are in fact lower percentage plays, but there are good reasons to run them. The primary thing to remember about why is that when you score with time left, it gives your opponent a chance to score, too.
For example, say you run a play where your chances of scoring are higher — like around 50 percent. But you give your opponents the ball, and they get a chance to score now. And your chances of stopping them are the league average of 70 percent. That means your chance of winning the game are about 35 percent (.7 *.5).
Alternatively, if you keep the ball, take a last-second shot, and reduce your chances of scoring to 40 percent, you actually have a better chance of winning the game. In other words, denying your opponent a chance at a shot has value too, and isolation plays are the best way to control the clock while reducing the chances of a turnover.
Ergo, slightly over half of the shots (298) came with three seconds or less on the clock, and the effective field goal percentage on those was just 26.84. The team attempting the shot won 101 of those games (33.9 percent). Teams that took the shot with four or more seconds won 103 of 275 contests or 37.5 percent of the time, in spite of shooting an effective field goal percentage of 41.8 percent — significantly better.
So the difference in winning percentage is close enough that it may depend on who you have to take the shot and what kind of defense you have. So, let’s start with that and take a look at how each team did, both in terms of both scoring the shot (top chart) and defending the shot (bottom chart).
In both cases, the number of shots is on the vertical axis, the points scored is on the horizontal access and the larger the team icon, the greater the effective field goal percentage scored by/against the team. So remember, in the bottom chart, smaller is better.
First notice the Memphis Grizzlies in the top chart. Their 60.87 effective field goal percentage led the NBA. The Oklahoma City Thunder, at 4.76 percent were the worst. Memphis outscored the Thunder 28-2 in these situations, in spite of holding a shots advantage of just 23-21. If you’re looking for a “team ball > hero ball” justification out of this, that’s it.
Then take a look at the Boston Celtics, because that’s such an extraordinary case. They attempted 38 “last shots” and were far from efficient on those, with an effective 36.8 percentage.
But look at what their opponents did. They only attempted 16 shots. That means the Celtics took 70.4 percent of the shots with less than 24 second on the clock.
Now look at the Portland Trail Blazers, who had opponents attempt 23 shots with the shot clock off while only attempting 11 themselves. That means they only got 32.3 percent of the chances to score down the stretch.
Most people would rather have the ball twice as much rather than the opponent getting the ball twice as much. This shows there’s something to the whole “clock management” part of the argument.
Alright, now this is where the real fun begins. Are you ready? Here’s every attempt to tie or win a game with the shot clock off last season. Every. Single. One. They’re plotted by team, player, time and distance, as well as makes (green) and misses (red). Clicking on the team will let you filter by that. Hovering over a player will highlight all his shots, as well as the details of that specific shot. Or, if you want to pull up just his shots, type all or part of his name in the box. Have fun. Play with it. That’s what it’s there for.
The main thing that you’ll notice is that a lot of concepts get blown. That’s a lot to do with things ranging from small sample sizes to the player’s skills to who’s around the player. The main thing to take from this is that hard/fast rankings are hard to take from this. “OK, OK,” you say, “I got it! Now just shut up and tell me who’s the best.” Here’s one final shot that’ll do that, but bear in mind the caveats that have preceded it.
Last season, at least statistically, the best man with the game on the line was Anthony Davis. Not only was he the player who notched the most points (13), he was also the one with the highest effective field goal percentage (92.86—are you kidding me?!?!). Danny Green had the next best rate, scoring nine points on five attempts. James Harden had the next best point total with 12. Kemba Walker took the most shots (15) but only scored 11 points on those shots.
Russell Westbrook’s 1-of-9 was the worst rate of anyone who made a shot. Rudy Gay took the most shots without a make, with seven. And Kobe? He was 2-of-10. Lakers not affiliated with poisonous African snakes had an effective field goal percentage of 26.9 percent, which means their shots were 35 percent more likely to go in than Lakers of the lethal reptilian variety. So make of that what you will.
 It’s worth stating that since this doesn’t include free throws, that could be throwing off some of the percentages a bit.