# Statistical Analysis Request

So, I’m watching both of the games tonight, and in both of the games, a team has the ball down either 5 or 6 points with about half a minute left. In both instances, the announcers were stating “You don’t have to go for a three here,” and in both instances, the teams did not, in fact, go for threes.

Here’s my request. From my perspective, I think it is absolutely absurd to not go for three when you’re down five-six points with thirty seconds or less left (presuming your opponent can hit free throws with some expertise – my strategy might be different if I was playing, say, the Memphis Tigers).

Obviously, because two different sets of announcers think otherwise, and two other professional basketball coaches agreed with the announcers, some people who are very knowledgeable about basketball disagree with my position.

So my request is some statistical analysis on what is a better statistical move down five-six points with thirty seconds left. Does it make more sense to go for the quick two? Off the top of my head, I cannot think of a way to measure it, so I thought I’d toss it to you brilliant readers out there.

## 25 thoughts to “Statistical Analysis Request”

1. James says:

I think you have to go for the best shot. Draw up an open-ended play that involves multiple opportunities for defenses to leave a man open, and have that person take the open shot, from wherever he is. A 60% 2-pt FG is a better shot than a 35% 3-pt FG, especially if you can save 5 seconds. Also, have the point guard figure out if the defense is especially guarding the 3-pt range, and if so, it’ll be easy to get a quick 2. If not, try to get an open 3. Sorry I don’t have any statistical analysis

2. Milt says:

I think someone on the APBRmetrics boards did some research into this. Post a thread and I’m sure you’ll get an answer.

3. caleb says:

off the top of my head, I am more bewildered by the wasting of time… teams will swing the ball around for 10 seconds, to get a marginally better shot. The point of shooting a quick two, then fouling, is to extend the game into more posessions… and yet teams constantly waste the time.

(On a related note, the fact that 29 NBA coaches (Phil Jackson aside) haven’t realized that 2 for 1 is a good strategy at the end of a quarter, shows how pathetic math education must be in our schools….

4. Nick says:

Two for one is a good idea, it’s just the crappy shots flung up on the first shot probably turn lots of coaches off, since it means miss 1 then get it back with 5-8 secs left to have the opporuntiy to chuck up another rushed shot. Perhaps I’ve just been watching too much Jamal and Jason Kidd with their 40% over the last few years

5. caleb says:

re: 2 for 1 —

You are right that it freaks out coaches to see “bad” shots, but that’s their mistake — it’s a situation where a “bad” shot can actually be a good one.

THe bar is set very low — if the shot you get is even HALF as good as the “ideal” shot, the odds are in your favor.

This assumes a “true” 2 for 1 situation, meaning there will be at least a few seconds left for the 2nd posession (to find a meaningful shot, not a half-court fling) i.e. 26 seconds left in the quarter isn’t really a 2 for 1 opportunity; 35 seconds left definitely is.

6. Ricky says:

2 for 1 assumes that the other team has a very small chance of getting an offensive board on the other end of the floor, otherwise it’s like giving the other team two possessions. I’d say if a coach has a very strong rebounding team, it’s worth it. It’s definitely a risk though, certainly a more aggressive strategy than limiting your team to 1 shot so the other team is forced to run out the clock. The last team to shoot in a quarter actually has less of a chance for a put-back, if the shot comes close to the buzzer.

7. DS says:

1 week until the lottery… who’s excited?!

8. Ricky says:

I’d be more excited if we were looking at 14% instead of 7%.

9. caleb says:

of course, the team trying for the 2 for 1 also has a chance at an offensive rebound.

so, rebounding position (and personnel) are a factor in what makes a good shot… but the basic math is the same — I think it’s pretty easy to get a scoring opportunity that’s at least 50 or 60% as good as in a “standard” offensive set… you’d rather have two of those, than a single better opportunity.

10. big fella says:

i thought that i read we have something like an 18% chance of getting either the 1st or 2nd pick – which is what we need to get D Rose, right?

11. “the fact that 29 NBA coaches (Phil Jackson aside) havenâ€™t realized that 2 for 1 is a good strategy at the end of a quarter”

actually Popovich has been taking this one step further this postseason, fouling intentionally when the other team is holding for the last shot at the end of many quarters to get the ball back for their own last shot. I don’t think it’s been especially successful, certainly not this series.

12. caleb says:

You’re right, he has been doing it and it hasn’t worked… But it is a good percentage play, esp when fouling guys like ely or chandler.

Coincidence or not I’d call jax and pop the 2 best coaches in the league.

13. well, the one year Pop didn’t have Duncan in his career, he went 17-47. :)

more and more, I think that coaches really don’t matter too much, it’s all about your best player or two. Byron Scott got laughed out of NJ with the scuttlebutt being that he was a figurehead and Eddie Jordan did all the real coaching. Doc Rivers, Flip Saunders, these guys aren’t anything special. a bad coach can screw a team up, but if you send the right guys out there, I think quite a few of them are interchangeable.

Jerry Sloan I like a lot, though, he never seems to get mentioned enough as one of the very best.

14. W.C. says:

I think the decision is almost entirely dependent on the type of players you have and the defensive weaknesses of the team you are playing.

If you have a big man that score easily and quickly with a set play and all your 3 point shooters suck, is there really any question what you should do?

If you are a small team with a bunch of dead eye outside shooters and the opponent doesn’t defend the premiter well,is there really any question what you should do?

It also probably depends on what you think the other team thinks you are going to do. That way you can do the opposite and get an easier basket.

I realize this is not the answer people are looking for, but in my experience if you have to ask the question that means it’s a close enough decision to be driven by the specifics of the situation and not generalities.

15. caleb says:

Agree, impact of coach is dwarfed by impact of key players… But still there. Don’t think lakers would have #1 seed, be on verge of conf finals, if mike brown were coach.
Biggest role for coach is not xs and os, tho, but putting the right players on the floor. Look at IT…

16. Nick says:

I would agree that much of coaching is talent evaluation and putting the right players in the right roles. Though there was something about that Laker – Piston series that made you think Larry Brown somehow really did give them an edge.

17. caleb says:

An example on the other end of the spectrum would beletting yourself be the worst defensive team in the league while playing Renaldo Balkman for 8 minutes a game.

18. I expect to research this soon by (first, at least) modeling what teams have historically done to identify how a specific shot will change the probability of winning. Thus based on the situation you can get an idea as to what is the best choice, and/or explain why coaches instictively agree that going for 3 is a bad idea.

19. Hudson River says:

I think its tough to analyze a coach’s impact without actually being in the locker room, or in training camp. A coach can inspire a team or demoralize them, adapt to them or pound them with a system.

Theres no dispute there is a huge disparity between the Spurs offense and the Suns offense and I strongly doubt that the change in offensive efficiency relies solely on the players.

20. cwod says:

Mark Jackson says some dumb stuff.

21. Brian Cronin says:

I expect to research this soon by (first, at least) modeling what teams have historically done to identify how a specific shot will change the probability of winning. Thus based on the situation you can get an idea as to what is the best choice, and/or explain why coaches instictively agree that going for 3 is a bad idea.

Awesome, Ryan!

Drop me a line whenever you finish, and I’ll link to it/post it here!

22. Mike G says:

If I were coaching, down by 5 or 6, half a minute, I’d do this (until it works, or doesn’t):

1) Look for a 3; if the opponent has that covered, the 2 is a better bet.

2) Go aggressively for a steal, rather than for a deliberate foul. Even tell the refs, “We’re not trying to foul here”. You may have a 60% chance of a foul, 15% for a steal, 15% for something else good: turnover (by a player expecting a foul), offensive foul, missed call, etc.; and 10% that you’ll miss the guy with the ball and the opponent gets a breakaway, etc.
I like these odds better.

Then go back to #1.

The opponent is likely strategizing: Don’t give up a 3, and don’t foul in the act. Therefore, the uncontested (or lightly defended) 2 is often available.

To make up 5-6 pts in 30 sec., you need to have several things go right. They don’t have to be perfect, if the opponent plays imperfectly.

23. I’m looking forward to seeing some stats on this… I will check back soon…

24. Mike G says:

I don’t think you’ll find any coach that thinks “going for 3 is a bad idea.” You take what the defense gives, and ‘forcing a 3’, like forcing any shot, is a bad idea (exceptions granted).

It’s not all that unusual, in the normal course of a game, for a team to make up 5-6 points in 30 seconds. Maybe a study should just look at those intervals and see how it was done.