This game defied my extra possessions stats theory. The Celtics gave up more ORebs (11 to 6) and more turnovers (11-5) which resulted in 11 more possessions for the Heat. They ended up getting 14 more shots (94 to 80 FGA). We had a few more FTA that accounted for the other 3 additional FGA by MIA. But we hit 9 more shots, including 6 more 3s, and 9 more FTs. I guess when you shoot that well, you can overcome giving up 11 extra possessions.
We average about 43 3PA per game (most in the league) and took 40 in this game. I like the trend. More than 40 seems like too many. We are now 15-9 on the road, OKC is best at 14-8. We have been playing much better on the road the last month or so. Like that trend too.
Is there any correlation to # of 3's attempted and either losses or close finishes (or giving up double-digit leads)? It may not play out statistically especially since lots of 3's go up in garbage time in blowouts - and when 3's are falling more get attempted. Maybe there is a sweet spot +/- # of threes where it becomes statistically detrimental.
The Celtics are at about 38% from 3 on the season (6th best). There is going to be inherent variability game to game. I don't know the statistics of this but my guess that a normal range would be say 30% to 45% on any given night would be an expected statistical normal deviation from the mean.
Qualitatively, what I see is that when they are taking less than 40 3PA, they are taking better shots on average so the percentage made is better. Nothing wrong with taking a 3 when it is a good look. The team average 3PA ranges from a low of 30.5 (Lakers) to a high of 42.7 (Celtics). So even <40 is still pretty high. We play with 3 guards and a big who can shot 3s so you expect this team to take more 3s than most teams. But I do not believe that more 3PA is always better.
It would be interesting to see winning percentage and 3P% as a function of 3PA, but I don't know how to mine that data. My conclusion of 40 3PA being some kind of a magic number or sweet spot is purely anecdotal based on looking at individual box scores over the course of the season. At this point, I can't back that up with any hard aggregate numbers.