In my eye test this often ends up in us taking 2 bad shots vs a good one plus and-1 for the opposition
the 2 for 1 has got to be the most overrated thing in basketball. I see so many teams throw away a good possession throwing an incredibly low percentage shot, miss it, only to have the other team run good offense and score, then the other team isnt left with much time for the last shot, and just end up with another terrible attempt. run a good play, score, go down and set your defense and get a stop.
If I have to explain something sooo common knowledge like 2 for 1 vs just playing for 1 shot; than I'm certain clutchfans has turned into bizarro reddit.
If you can consistently hit at least one of the two attempts, then it is worth it. Do they have this kind of stats? To simplify it, let say, the percentage of hitting a normal shot is 50%. And the percentage of hitting a rushed shot to get a 2 for 1 is 33%. (All 2pt shots to simplify) So the percentage of missing a normal shot is 50%. The percentage of missing both rushed shot is 44.2%. So it is worth it. But I have no idea if my simplified estimate of the rushed shot percentage is correct.
Yessirrrr! Worth going 2 for 1 if we have 35+ seconds because we get at least 10 seconds the second possession to get off a decent shot. Definitely helps having 2 isolation maestros creating shots to take advantage of 2 for 1s.
I decided I wanted to see the actual numbers on this so I wrote a script to grab all the play by play data, came up with the following numbers for this season. I'm sure it's not perfect, but I didn't want to go through the play by play for every single game manually. Rockets gained: 258 Opponent gained: 187 Total opportunities: 136 Rockets points per: 1.897 Opponent points per: 1.375 We are only getting 1.9 points on two possessions, which means 0.95 points per possession in these 2 for 1 opportunities. That's really bad, so it confirms the "2 bad possessions" idea. Worth noting that we do not always get 2 possessions here, but I still counted those cases. Some other posters pointed out that we play really bad defense on the one possession we give to the other team. The stats confirm that, giving up 1.38 points on average is really bad. Note on the way I calculated it - there are some situations where we try to go 2 for 1, but the other team also gets 2 possessions. In these cases I counted all their points scored, so, in those cases it's a 2 for 2. I think this is the more accurate way to look at it, but as I said, that means it's not technically correct to say 1.38 points per possession. More like 1.38 points per situation. It's clear that we come out way ahead here, on average we get about half a point per 2 for 1, and we have 136 in 62 games, which is roughly 2 per game. So on average this strategy gives us 1 point per game. Devils advocate - you could still argue that it's not worth it. The benefit here is 0.522 points, so you could argue that "one good shot" for us would maybe be 1.4 PPP, and then we could hold the other team to 0.85 PPP on the following possession, given the short shot clock. I don't think those are realistic numbers, but if you still don't agree with this strategy, I think that's the case you would have to make.
First, very cool that you grabbed that data. Can I DM you later to discuss your script — nba stats api, gotchas, etc. I want to write some, too? Thx for doing this!! wrt your quote. Saying 1.9 confirms it’s bad, is not the way Morey and everyone in the NBA looks at it. They look at it as an extra possession, so any extra is gravy, meaning that’s a 1.9 single possession, by use of the strategy. Basically the strategy is: any time you score 2pts or more on the sum of the two possessions, it’s a *resounding* success When you score twice, it’s on the level of wicked unfair. It’s only when you fail on both possessions due to not running real offense is it a failure, which doesn’t occur much, on the average of 1.9 pts per use of the strategy.
Sure, I'd be happy to discuss it with you. I didn't use the NBA api, though, maybe I should have. I just scraped the play-by-play Yeah I understand that, and agree. I guess I was just trying to state the obvious, that each individual possession is going to be "bad" compared to our normal, because they are rushed. But overall the two bad ones are still much much better than one good one.
Of course it should be calculated as points per situation, not possession, because the whole thing is that we are trying to gain one more possession by rushing the first possession. So the investigation should be how many points do we get by the two rushed possessions against one normal possession. I like that you include opponent's PPP in these situations. The rushed possession might have some effect on the defense when the opponent gets the ball back. When you hurry up to bring the ball up to shoot, would that make your defense in a disorganized state and might let the opponent get an easy score? Or is it the other way around, when you hurry up, you leave a few players in the back court and that may prevent easy fast break?
hahahah THIS If you are pro stats then give me a stats system that measure momentum and psychology! enough of this 2>1 or 3 > 2 nonsense!!
I’ll take two James Harden stepbacks as opposed to whatever good offense your team is running, thank you.
Don’t we have amazing percentages on shots that take place within 7 seconds or so? I always though that was the case with D’Antoni teams.
I don't know if that claim has any statistical evidence. Even if it is true, you still need to know the context of those stats. Most fast breaks takes less than 7 seconds, and are high percentage possessions. But we are not talking about fast breaks. We are talking bout hurry-up offense.
I would like to see the percentages on this “hurry up” offense. Question for you. Do you really think a Daryl Morey ran team has not run the math on these numbers? Clearly, everything we do is predicated on data, and I have to believe that the data leads to 2 quick shots over 1 “good offense” shot. (Whatever that means...) The fact that we might be the GOAT offensive team even this year reflects that.
I agree. I am not questioning the validity of the 2 for 1. Just wanted to point out what kind of stats would validate it. Pretty sure Morey has more sophisticated data for this kind of things. Just like football, a hurry up offense's goal is to score quickly in a limited time. If that offense was a generally good offense, they would have used it the whole game, not just for the final minutes.
