(This article continues the thinking described in the recent “Is the Whole Truly Better than the Sum of the Parts?” article.)
This past season we spent considerable time and several Hoop Coach articles trying to arrive at a method for coaches to obtain an accurate number to ascribe to an individual player for his performance in a basketball game. In attempting to arrive as close as possible to a number that accurately portrays a player’s real game, we’ve been driven by the following conclusions:
- Using any system to evaluate YOUR team by an outside agency using third party taggers is only slightly telling, and therefore only slightly meaningful and valuable.
- Over the course of a season, the best way to arrive a player’s value is his net point value- (offensive contributions to points scored and defensive accountability for points given up).
- Since the object of a game is to outscore the opponent, all actions that don’t lead to a score by either team aren’t included. The premise is simple-mistakes and omissions will be assimilated in scoring results over the long haul. For instance, if Player A regularly doesn’t block out, he’ll get docked for points for that omission on scores. If he lucks out on occasion, the points he gives up with this issue will eventually nail him. If a coach really wants to micro- focus on this player’s problem, he should chart Blockout %.
- On the offensive side, both PER and EFF do not give value to players for screen assists or second-assists.
We’ve already discussed the defensive side of this issue (See previous Hoop Coach articles, Charting Defensive Accountability I and II). But, let’s quickly review some key points:
- On the defensive side, 3rd party taggers don’t have a clue as to how YOU really want to guard the ball, defend screen/roll or how want to help-let alone where you want to pick up.
- Third party taggers are also clueless as to your game-to-game defensive adjustments or your IN-GAME adjustments.
- “Points Given-Up” are rarely accountable to one defender; usually two or three defenders are accountable for an opponent’s score.
So, today we’ll share a methodology for ascribing credit for offensive contributions other than scoring. As an example, I took a D1 game from last season where I already analyzed the defense. Team A beat Team B 78-68 and I assigned accountability to individual players for all 68 points. (See attached chart.)
I then analyzed each of the 78 points Team A scored in that game and gave credit to individual players for assists, screen assists and sometimes second assists when I determined that they directly instrumental in a score.
These are the guidelines that I used. Obviously, you can invent your own system and methods:
- There were a number of occasions where the scorer received ALL the credit. There was a post isolation where all other 4 offensive players stood and watched the scorer. There was a pullup triple in transition but it was really one-on-one. There was a put-back. You get the gist.
- When a scorer didn’t receive ALL the credit, I gave him HALF of the credit for the points. The other half went to 1, 2 or 3 other players depending on whether there were these contributions: ASSIST, HOCKEY ASSIST, SCREEN (ON OR OFF BALL), OFFENSIVE REBOUND.
- On a 2pt FG or 3pt FG created only by an assist, the assist man got credited with half the points.
- There were possessions where a 2 or 3 was divided by 4 players. In the case of a 2, the scorer receiver credit for half the points and an assister and two screeners shared the other point-so each got credit for .33.
- There were several instances where Team A missed a FG attempt and an offensive rebound saved the possession for a subsequent score. That rebounder received credit for whatever portion of half the points scored.
- For non-shooting FT’s scored after the 1 and 1 went into effect, I took the total (5 for instance) and divided by the number of players who played in that half. So, if 7 players were in, they each received .7 pts of credit for that half. (If you want to be more precise, you could actually assign bonus FT’s credit by the number of minutes played in the half by individual players).
- Last, let’s examine how charting offensive contributions other than scoring might change one’s viewpoint on a player’s true value. If you refer to the chart, it is very graphically illustrated that:
- The #2 Man scored 32 points in this game and “gave up” 15 points for a value of +17 points but after you credit his teammates for all of their contributions to his scoring, he gets credit for only 23 of the 32 points for a net of +8 “true game value”.
- The #1 Man didn’t score a point but yet got credited for 3.5 pts because of his assists.
He gave up 9 points so his true game value went from a -9 to a minus 3.5.
- The #5 Man only scored 2 points but his screens and offensive rebounds increased his true game value to +6.25.
- The Guard Sub scored 11 actual points but like the #2 Man his true game value was reduced by all the work his teammates contributed to his scoring.
- The Frontline Sub did not score and gave up 4 points but his screens contributed 2.5 points to raise his true game value to a -1.5 from a -4.
The last calculation of dividing a player’s True Game Value by his minutes played gives you his True Game by Minute Value. (The chart also contains this calculation.)
In conclusion, if one wants to really measure a player’s value in a game, the work is a little tedious. However, this work can really pay off in assessing how players really perform at both ends of the floor and what their true game value is and consequently what their “True Season Value” is. This scrutiny will pay immediate dividends in motivating lesser scorers. Scorers will also eventually see the benefits of this practice as it will help them realize that they can accumulate value themselves with screens, assists, second assists and offensive rebounds as well as reduce their “points given up” at the defensive end. As mentioned previously, the bigger the sample size of games, the more accurate the modified plus/minus system becomes as well as the subsequent true value of a player in those games.
Here is a sample game with True Value based on Modified Plus-Minus