Understanding Effective Field Goal Percentage and Utilizing It

Back before the onset of the 3pt FG at all levels of play, there was obviously only one FG%.  No matter where one took a shot, layup, 24 footer or anything in between, the reward was always 2 points.  Then, the various arcs came into the game at various times and at varying distances.  So, for shots beyond the arc, the reward was 3 points.  Naturally, this necessitated keeping track of both overall FG% and 3pt FG%.  If someone wanted to isolate 2pt FG%, that was simple enough to do by subtracting those attempts and makes from the overall numbers.

Well, of course, even the least math oriented coaches came to the realization that players didn’t have to shoot as high a percentage at the arc as they did outside the arc to be effective.  Within hours of the introduction of the arc at the college level (where I was at the time), I clearly remember thinking that a player who would shoot .333 from the arc would be equally effective as one who shot .500 from inside the arc- certainly just a rudimentary math observation.  Many coaches came to similar conclusions in varying ways simultaneously.

Certainly, we’ve come a long way since then as the metric trend has exploded over the world-wide basketball landscape.  Basically fostered initially by math aficionados, who weren’t even directly connected to the game in any way, now the trend is embraced and utilized by even the staunchest of early objecting coaches.

With post arc FG% analysis, one of the first reactions by coaches was emphasizing that and even insisting that their players shoot only two types of shots-outright layups and triple attempts.  The premise, (almost always proven mathematically sound) is that any player’s FG% on layups is going to be higher than other inside the arc shots like runners, jump-hooks and pull-up jumpers AND that a player’s EFFECTIVE % on 3 pt. attempts will very often be higher than all combined non-layups inside the arc.  A simple example follows: If Player A only shoots .300 (which is actually considered very poor at any level) from outside the arc, that translates to .450 from inside the arc because of the reward difference.  That .450% in many cases would be higher than the % players would shoot inside the arc on the more difficult shots than layups.  Obviously, in the real world, most teams take non-layups inside the arc because those are the shots they can get-especially at the end of a shot clock.

Around the same time, a parallel math equation was formulated by a number of the math analysts that confirmed coaches’ suspicions that the most accurate way to analyze FG% was to combine the statistics inside and outside the arc- hence Effective FG%.  The formula is simple:

eFG%= (FGM + (0.5 X 3PM))/FGA.

Let’s illustrate by examining 3 vastly different 2015/16 NCAA players:

Player #1 is Sam Beeler, a 6-10 interior player from UC Santa Barbara, who only attempted shots inside the arc (not a single triple attempt).   Beeler was 101 of 140 from the floor so the equation looks like this for him:

72.1% =101 + (0.5 X 0= 0)/140.  At 72.1%, Beeler led D1 in both “regular” FG% and “effective FG%.

Player #2 is Max Hooper, a 6-5 wing from Oakland University, who only attempted shots outside the arc (not a single 2pt. attempt).  Hooper was ranked #7 in the country in effective FG% and was 30th in 3pt FG% at .455.  Hooper was 117 of 257 from beyond the arc and his equation looks like this:

68.3% = 117 + (0.5 X 117=58.5)/257.

Player #3 is Buddy Hield, the All-American wing from Oklahoma, who had an almost equal number of shots inside and outside the arc.  Hield was ranked 39th in the country in Effective FG% and 25th in the country at .457 on 3pt %.   Overall, he was 301 of 601 and outside the arc he was 147 of 322.  His equation includes both facets, unlike the other two examples:

62.3%= 301 + (0.5 X 147=73.5)/601.

What conclusions, then, can be drawn from these examples and all other Effective FG% numbers?

  1. Buddy Hield for sure, and Sam Beeler possibly were going to get all the minutes from their respective coaches that they received for additional reasons other than Effective FG%’s.  In the case of Hield, his athleticism, floor game, rebounding for a guard and on-ball defense were all plus-plus assets.  In Beeler’s case, his rebounding and basket protection assets were other reasons to keep him on the floor.
  2. However, for the most part in Max Hooper’s case, his Effective FG% would have screamed for him to be on the floor on that alone, even though his off-ball defense was very good and camouflaged average athleticism and the absence of an overall floor game,
  3. Most players at all levels don’t fall into the Beeler and Hooper extremes.  In other words, most players will have accumulated some combination of two and three point attempts.   The more average players are in this category, the more a coach can use Effective FG% to delineate differences.
  4. Most players will fall somewhere in the middle of a traditional bell curve when it comes to ability.  Hield’s athletic and basketball ability is 99th percentile worthy.  Again, using Effective FG% will aid delineating differences among a number of similarly talented players.  The great ones don’t need to be analyzed with the same scrutiny as players in the middle of the pack.  Coaches make far fewer mistakes on the very good or the bad than they do on middle of the bell curve players.   
  5. The more tools a coach has to analyze his players against each other, the better.  PER, Usage Rate, Effective FG%, Points and Rebounds Per Minute (Next Article) are all easy to compute and don’t require video work.  In conjunction with a defensive rating system which most certainly requires video work, with some of the above offensive metrics, a coach can hone in statistically on his players’ performances and eliminate guesswork-as best as possible.

Like any other phase of the game, metrics or analytics are worthy of serious consideration.  The minute a coach says to himself, “I don’t need to know (X)”, he restricts his understanding of an ever evolving game and stunts his growth as a coach.  Rather, it seems that an evolving coach might rather say, “I know and understand X, but I’ve thought it through and choose not to utilize it in my system.”     

3 comments

  1. Curious if anyone has seen a metric that not only includes the additional point scored with a 3 point shot, but also recognizes players that get to the Free Throw Line and make a good percentage. For example, last night D Wade was only 5-13 (38%) from the field, but one of those 5 was a triple and he was 11-13 from the free throw line. So this should really be seen as an efficient night of shooting. His stat line would make it see that he attempted 19 shots (assuming one free throw was for an “and one” vs. having missed the front end of a 1and1), got fouled on 6 of them, and scored 20 points. So on a 2 point shot scale, he was really 53% (10-19). Another way to look at this would be to say he scored 1.05 points on every shot attempt (“shot efficiency”). Wondering how broadly this view is used?

  2. Interesting to see this post again. I’ve started tracking the metric mentioned in my last comment in 2016 for my local club players, and am calling it PPSD – Points Per Shot Decision. This is a great indicator of what the team yields in points every time a particular player makes the decision to shoot. Looking at shooting effectiveness this way really reinforces the core tenants of the “3 or Key” generation of basketball, and seems to help players in their shot selection and decision making on where/how to attack via shooting or driving. James Harden really exemplifies shooting efficiency and has a ridiculous PPSD as he makes a ton of 3s, gets to the rim a lot, gets to the free throw line a lot and makes his free throws. Does anyone know if there is a metric at the pro/college level that is the same as this?

    Note: If you use this metric you will need to make a decision on what to do about free throws that were not the result of a player shooting the ball (ex: bonus or double bonus situation, technical foul). It might be easier to leave those out, but there’s an argument to include the points associated with them while assigning a portion of a shot to each scenario at a “discount” to a real shot decision (ex: 1 FT due to technical foul is .25 shot instead of .5 shot, bonus situation (1 and 1) is .5 shot, double bonus or 2 technical free throws is .75 shot).

Leave a Reply