Three Point shooting efficiency as it correlates to winning

Basketball-reference can help you with this using their NET criterias. In the Team Game Finder, just set them greater than 0, combined with NET Pts to determine the winner or loser.

I just ran this for FG%, 2FG% and 3FG% for 2017-2018. Turns out 3FG% is the worst indicator of the three, by far.

• Overall FG% matches the winner in ~ 80% of games (I was surprised by this)
• 2pt % matches the winner in ~ 70% of games
• 3pt % matches the winner in ~ 67% of games.
  
  that said, the Rockets lost the fewest game when shooting higher 3FG% than opponent -- just 3 games out of our 17 losses. So, there is likely a more advanced correlation between 3pt% at high volume and Wins.

Also ran FTAs. The team who shoots the most FTs is only the winner in about 57% of the games.

 

Overall FG% includes 2pt% and 3pt%,

Dun want to sound like a smart ass but the total package usually is defining.

One thing though ---- The Rockets have their own line of stats. They are outliers.

Rockets are still scoring 2pters in terms of layups, just fewer long 2s.

 

I also ran TOs and Off Rebounds. They are about equal to FTAs as far as Win indicators go.

wrt what surprised me: Yes, I was surprised that FG% was so high. Get this, it produced more wins than eFG%, too, (albeit, just barely).

As far as advanced indicators go (beyond the classic box score stats), did you know of the so-called 4-Factors. I'm not completely up on how they became the accepted 4-Factors, but I'm sure they are tied to some type of Win Predictors using various correlating factors.

The Offensive 4-Factors are: eFG%, TOV%, ORB%, FT / FGA (slightly different than free throw rate).

----------- Offensive Rebounding Effectiveness --------------------

This brings up another question I think is worthy of a thread: I'm curious if the new 14-second rule on ORB will affect the value of offensive rebounds, thus reduce the importance of ORB% as a 4-Factor. I can even imagine some coaches like Pops (and JVG in the past) who are big on the importance of transition defense, will even further decide to devalue Off Rebounds in favor of getting back on D more aggressively. That is to say, if the value of the Offensive Rebound goes down due to a shorter shot-clock (I don't see how it can't, just a matter of how much), then why go for Off Rebounds as much vs shoring up your transition defense.

 

"You can't win unless you outscore your opponents."

-anonymous

 

well, FG% doesn't count 3 points, so that only makes sense if each team is shooting the same volume of both. I'm not surprised its still the #1 box score indicator (I'm sure throughout history, too), but 80%??

So that's the whole point of my surprise that FG% would have remained that high (actually higher than eFG%, too)...because 3pt% will drag it down, as the league moves to higher 3pt Rate -- assuming not all the teams are increasing volume equally.

If every team adjust rates of 2s and 3s equally, then I can see the indicators remaining the same. But if only half the league embraces high volume 3s (still true as of last year), then you'd expect FG% to become less important as an indicator, versus what it's historically been.

bottomline: Be interested to see the trend over past 20yrs whether 3pt% has been making a gain on FG%, and whether the WIn% of FG% has been dropping.​

 

Shouldn’t matter, as possessions (and Pace) are essentially the same for each team at the game level. (I think the maximum possession delta per game is 2...not counting OT.) So per game, stretching or condensing totals to 100 possessions would effect both teams equally.

 

Right. That’s what Morey and MDA say, too. But, I think they equally talk about makes. Rockets have a tremendous Win% at something around 13 makes, regardless of %.

 

I’m going to guess Makes, % then Attempts. Basically, I’m choosing points as an indicator over efficiency...like you said in the OP—the old who scores most points indicator. That should be true for FGM over FG%, at least.

 

You are probably exactly right about this.

Was thinking more about the 4 Factors. The formula for ORtg (offensive efficiency) is a very long formula, that calculates possessions from boxscore data (thus indeed, the formula includes TOs and ORBs) and accounts for 3 pointers and FTs. You know how in math they teach you to isolate parts of formulas....I bet the 4 factors might indeed comprise the ORtg formula in its entirety. So, the 4 factors might be the actual four components of the formula....just not in per 100 terms, but rather percentages.

And ORtg = Points per 100. So, since possession are nearly equal per game, the better ORtg team will win the game 100% of the time, except for a few anomalies surrounding the fact that one team can actually have 2 (at most, not counting OTs) possessions (POS, as defined by the NBA stats) in one game more than their opponent.

(Think of an OT game. The team who gets the ball first in OT, can actually get it last, too.) But those are minor anomalies.

 

The formula is mainly just estimating possessions, so it then can provide Points per 100 possessions. It is actually very simple what it is trying to achieve. It’s just more complicated than one might think to estimate numbers of possessions from the Box Score.

The big goal is to be able to provide possessions from Box Score data going back as many years as possible. If PbP data was available for all years, the ORtg formula would be unnecessary.

 

The NBA stats definition of possession is just like football,,,defined by change of possession. It doesn’t matter how many shots you put up, it’s still one possession until the other team gets the ball.

So, possessions exactly see-saw, except for the fact (just like halftime in footbal), teams alternate start of quarter possession based upon opening tip, regardless of who had the ball last in the previous quarter. So, the possible Max diff in possession is 2....meaning one team always had the ball last, but could *only* also have the ball first twice in a 4 quarter game.

Like this by quarter, noting teams alternate possessions by team who wins tip gets ball to start the 4th as well, other team gets 2nd and 3rd.

1. Team A has ball first and last = one more POS than Team B
2. Team B has ball first and Team A last = same amt of POSs, a PUSH
3. Team B has ball first and Team A last = same amt of POSs, a PUSH
4. Team A has ball first and last = one more POS than Team B

Thus, Team A had ball two extra possessions.

​

 

This thread makes my brain hurt, and I like it.

 

At first thought it seems to me that if FG% is more of an indicator than 3pt% for winning, then teams that shoot less 3pt shots should have higher winning % because 3pt% is always lower than FG%. The more 3pt attempt one shoots, the lower the overall FG%. But that's not the case. Teams that have higher 3PA do not necessarily have lower FG%.

The Rockets were #1 in 3PA and #15 in FG%.
Toronto was #3 in 3PA and #6 in FG%.
Cleveland was #5 in 3PA and #4 in FG%.
So it is not surprising that Houston, Toronto, and Cleveland are #2, #3, and #5 in 2P% respectively. And these teams were #1, #2, and #6 in wins.

Some other teams are more "normal" in that 3PA correlates inversely with FG%. Brooklyn, for example, was #2 in 3PA and #29 in FG%.

I wonder for teams like Houston and Toronto if their high volume 3p shooting helped their 2p% (spread the floor?) or they simply had the personnel to excel in shooting.

BTW, I am mildly surprised that GS was #16 in 3PA. I thought they would be much higher.