new stat, defensive / offensive efficiency

Discussion in 'Pistons and NBA' started by Lee356, Nov 25, 2005.

  1. Lee356

    Lee356 All-Star

    Joined:
    Aug 31, 2005
    Messages:
    4,410
    Likes Received:
    527
    Location:
    Flint
    Two main factors. One, the percentage of times you get a shot after the other team does. Your percentage goes down here if the other team gets offensive rebounds. It goes down if you turn the ball over. Whether you attmept a shot, or whether you get to the line, I cound that as an attempt for this stat. This factor to be exact goes like this:

    (Your shot attempts + your free throw attempts divided by 2.3) all divided by (your opponents shot attempts + your opponents free throws divided by 2.3)

    The 2nd factor is what percentage of your shots you make. For this, if you make 2 free throws, I count that as a made shot attempt. I also divide the number of your made threes by 2 and give you still more credit for made shots. For the denominator, its your shot attempts plus your free throw attempts divided by 2.3. This 2nd factor goes like this:

    (Your makes + your 3's made divided by 2 + your free throws made divided by 2) all divided by (your shot attempts + your free throw attempts divided by 2.3)

    Multiplying the two factors together gives you this new stat of mine. Defensive / offensive efficiency.

    Irvine, 50.8%, opp 50.3%

    Carlisle first year, 52.9%, opp 52.8%

    Carlisle 2nd year, 50.2%, opp 52.8%

    Brown first year, 47.9%, opp 51.8%

    Brown 2nd year, 49.9%, opp 52.6%

    So far Flip's first year, 49.7%. opp 57.5%

    Irvine's differential was a negative 1/2 percent, and his team did terrible. Carlisle in his first year was dead even, barely squeaked by Toronto and got pounded by Boston. Carlisle's 2nd year saw a bit of a positive differential and we got further in the playoffs. Brown had good differentials both years, and outstanding results. Flip so far is 9-1, with the best differential by far among the 6 years looked at.

    Some verbage I posted on this earlier in case it might help to show where I am coming from on this:

    Defensive / offensive efficiency. As a defensive stat, you look at effective shooting percentage by the other team based on your own number of shot attempts. As an offensive stat, well, you just use the exact same thing, but from the perspective of your opponents. So its just one number for you, and one for your opponents.

    This stat takes into account game pace, because its based on your own shot attempts.

    It takes into account your offensive rebounds. If you get a lot of offensive rebounds, you get a lot more shot attempts. Since I am dividing the opponents number by our shot attempts, it makes them look bad the more offensive rebounds we get.

    Free throws are taken into account twice. Once when figuring our number of shot attempts. If we got fouled every time down, and only scored at the line, we would have zero shot attempts for the year, but this stat would still work, as I factor in free throw attempts as shot attempts. There is more to this, but I am out of time for the moment. For now, I will just mention that our current team is effectively scoring 58% of the time after a shot attempt by the other team. This compares to our opponents at 50%. I have figured this all back for five year. I will post more later on this.
  2. Lee356

    Lee356 All-Star

    Joined:
    Aug 31, 2005
    Messages:
    4,410
    Likes Received:
    527
    Location:
    Flint
    more stats

    Det 1998 51.5% vs. opp 53.0%

    Det 1999 50.9% vs. opp 53.6%

    Det 2000 54.1% vs. opp 55.2%

    So other than our current team, the 1999-2000 team had the best offense, and also had the worst defense of any of the teams from 1997-1998 until present.
  3. max

    max All-Star

    Joined:
    Aug 15, 2005
    Messages:
    7,361
    Likes Received:
    358
    Gender:
    Male
    Location:
    my bunker in the Carribbean
    Stats

    You could seperate out 2pta*2FG + 3/2*3pta*3FG + 1/2*FTA*FT%

    But aren't 2-pt attempts, 3-pt attempts and FT's already factored into the total points and total FG%? Even if a team gets to line often and has lower shot attempts the FT's still factor into the total points.



    Really the bottom line to me besides winning % is the total point differential.
  4. Lee356

    Lee356 All-Star

    Joined:
    Aug 31, 2005
    Messages:
    4,410
    Likes Received:
    527
    Location:
    Flint
    same thing

    Max, what you are suggesting is the same as whats in my formula. I am not saying my formula could not be simplified, but I don't see anything fundamentally wrong with it. But here is the problem. I want to keep things in terms of percents, to make the stat understandable. I could have come out with a stat in terms of points per opponents possesion, but that does not easily translate in ones mind. What people are use to is field goal percentages. I just added to that a factor for how often they get to shoot, adjusting everything to make free throw attempted and free throws made a part of the percentages. And added in credit for triples having that extra point.
  5. Lee356

    Lee356 All-Star

    Joined:
    Aug 31, 2005
    Messages:
    4,410
    Likes Received:
    527
    Location:
    Flint
    for the Washington game late November

    With this stat, we say Washington effectively scored 49% of the time after we put up a shot attempt or set of free throws.

    Conversely, we had a 47.5% success rate.
  6. lemonpen

    lemonpen First Round Draft Pick

    Joined:
    Sep 1, 2005
    Messages:
    591
    Likes Received:
    0
    Location:
    Southfield, MI
    Make sense test

    One of the things my chief engineer stresses is "the make sense test". Rather than sweat the numbers, what value is it to know the success rate me or my opponent has after an opposing shot atempt.

    Not judging, just trying to see the potential.
  7. max

    max All-Star

    Joined:
    Aug 15, 2005
    Messages:
    7,361
    Likes Received:
    358
    Gender:
    Male
    Location:
    my bunker in the Carribbean
    define

    And to define where constants are derived from. Like where is the 2.3 coming from? Why 2.3?
  8. Lee356

    Lee356 All-Star

    Joined:
    Aug 31, 2005
    Messages:
    4,410
    Likes Received:
    527
    Location:
    Flint
    answer

    My contention is that good defense and good offense are to an extent the same thing. Good offense flows into good defense, which in turn can create offense. Note, I did not say create good offense, just offense. If your defensive play comes up with the ball, you still then have to execute good offense to keep the cycle going.

    So you shoot the ball. What is the expectation. Its that either make the shot, or that the other team clears the defensive rebound and you are now playing defense either way.

    By focusing in on your own shot attempts, you are starting at the right place to study your defense. Because defense is certainly part of that shot selection.

    Do you have your offensive rebounders in place? If you do there is a chance you will get the offensive rebound if you miss the shot. And of course, if you do get the offensive rebound, the other team can't score.

    If you don't make the shot, and the other team gets the rebound, do you have people back on defense to prevent them from getting an easy fast break score. You better have thought of this before the shot goes up.

    You have heard that rebounding is part of defense. Indeed, it most cetainly is. If you get the ball, the other team can't score. And this goes for both ends of the floor.

    So back to that time you are about to take a shot. What are you going to judge the success of this shot attempt on. It should be two things. Your percentage of the time you will get your points from the shot. And next, the percentage of times the other team will get a score when they get the ball after your shot, whether it be a make or miss. Yes, even if you make the shot, you could be so badly out of position on defense that they get an easy fast break score.

    So when does your defense start? The moment you take a shot. If you make the shot, that puts the other team under their own basket to start their offensive posession. If you miss the shot, your positioning of your guys will have a large part in whether your opponent can fast break on you. First of all, if you get the offensive rebound, you just shut them out of a posession. Even if you don't get the offensive rebound, fighting hard for it forces the other team to keep more guys at their defensive end of the floor to clear that board, and less at the offensive end.

    Now, you shot the shot at the right time with your guys in position. The other team still got the rebound. But they had to go into a halfcourt set. You play tough D and steal the ball from them. But that steal would have never occurred if they had just dunked the ball on you from a fast break. You got to avoid fast break scores from them to even give your defense a chance. So that shot selection is really, really important.

    Now, this is not meant to be everything about shot selection. A lot more to it. Targeting particular players on the opposing team to get them into foul trouble, keeping your own guys out of foul trouble if they are needed too much on defense, and many other factors go into shot selection. But here is one thing that is very important pretty much always. Get to the line and make your free throws. It gets them into foul trouble, but about just as important, it puts you into the best defensive position. (Assuming you do make your free throws, offensive rebounding position is not needed.)

    My formula counts a couple tries at the line as a posession. But some free throw attempts are on traditional three point plays. And a few on techs. So arbitrarily, I chose to divide by 2.3 when doing the calculation to convert free throw attempts to posessions.

    If you turn the ball over, you don't get a shot attempt. So, your percentage goes down. Remember, the first factor is how many shot attempts you get after an opponents shot. The second factor is what percentage of time you score 2 points. Adusted for made 3's as it is an effective percentage to be a bit more useful.

Share This Page