Late game questions: Who is the Thunder’s most reliable option?

Just after the third game of the season, a heartbreaking loss to the Minnesota Timberwolves in a game that frankly, the Thunder didn’t deserve to win, a question arose. Who among OKC’s three main offensive options is the best when it comes to late game shooting? Russell Westbrook earned a reputation as a clutch killer last year, and Carmelo Anthony has some history as well. Today we hope to clarify who has really been best in the clutch.

Now I’ll preface this with one note: I don’t really believe in “clutch.” The notion that some players are better in late game situations is more an issue of small sample size than actual substantial improvements. I do however think some players tend to wilt a little more under pressure than others which may detract from performance.

As well, the NBA uses a strange definition of “clutch”: Any game within 5 points with 5 or fewer minutes remaining. To me that’s not clutch that’s simply a relatively close late game. So for our purposes today, I’ve looked at both the NBA definition of clutch and something I’d consider more fitting: A shot to tie or take the lead in the last 24 seconds (that is to say under one full shot clock possession) of the 4th quarter or overtime. For our purposes the last three years are used, which was both a cut off for sample size as well as performance given Paul George’s injury in 2015. Noted are each player’s made and attempted field goals, effective field goal percentage (which takes into account both 2 and 3 point field goals), and the number of his makes which were assisted by a teammate. The final figure is simply used to show how much of the offense the player created on his own versus was set up to take.

2015:

NBA clutch: 5 minutes or less, within 5 points:

1. PG: 1/2, .500 eFG, 1 assisted
2. Russ: 46/120, .392 eFG, 7 assisted
3. Melo: 23/54, .481 eFG, 5 assisted

Dom’s clutch: 24 seconds or less, shot to tie or take lead:

1. PG: 1/2, .500 eFG percentage, 1 assisted
2. Russ: 1/9, .111 eFG percentage, 1 assisted
3. Melo: 1/4, .375 eFG percentage, 0 assisted

In 2015, we’re working from a pretty limited sample with Paul George due to him missing the vast majority of the season. His percentages are obviously best, but the sample is so small it’s tough to draw anything meaningful from it. Now looking to Russ and Melo, we see early on that Melo was more reliable both within the final 5 minutes and the final 24 seconds, besting Westbrook in percentages in both categories. However, he was set up on a much higher relative amount of his late game shots which factors into his higher efficiency. He has traditionally been a better spot up shooter, so this isn’t too surprising and could continue in OKC.

2016:

NBA clutch: 5 minutes or less, within 5 points:

1. PG: 44/118, .432 eFG, 18 assisted
2. Russ: 42/108, .417 eFG, 3 assisted
3. Melo: 25/69, .406 efG, 7 assisted

Dom’s clutch: 24 seconds or less, shot to tie or take lead:

1. PG: 1/10, .100 eFG, 0 assisted
2. Russ: 1/11, .136 eFT, 0 assisted
3. Melo: 2/10, .250 eFG, 0 assisted

Ugh. Those final 24 seconds numbers are just ugly in every way imaginable. To be fair, these are generally tough shots which are well defended to win games, and are small sample sizes, but none of these three shot it well. Moving the the 5 minute barrier, we see pretty similar shooting from the three players with Westbrook and George at considerably higher volume. As well, Russ was creating his own shot much more than the others in these situations, which isn’t too shocking given how much creating he does in general compared to the others.

2017:

NBA clutch: 5 minutes or less, within 5 points:

1. PG: 50/105, .529 eFG, 17 assisted
2. Russ: 36/82, .497 eFG, 5 assisted
3. Melo: 39/90, .478 efG, 15 assisted

Dom’s Clutch: 24 seconds or less, shot to tie or take lead:

1. PG: 1/5, .200 eFG, 1 assisted
2. Russ: 8/27. .370 eFG, 2 assisted
3. Melo: 6/16, .375 eFG, 2 assisted

Finally we see in the most recent year that all three were capable of hitting shots late game, and Russ at a rather high volume in the final 24 seconds in hitting 8 shots to tie or take the lead. Melo also did so at a pretty solid shooting percentage. In the final five minutes we have George shooting best and at the highest volume with Russ and Melo still at a reasonable volume but a fair bit lower efficiency.

***

As a whole, we can see here there is some variability year to year, but in general Westbrook has had to create the most of the three. None of them are great in the final moments but George has generally been most efficient. Given his defensive aptitude, I’d personally feel best with him and Westbrook on court late game. That said, I fully expect Anthony to remain there due to both reputation and ego concerns with the roster.