PDA

View Full Version : Why not foul everybody?

Putnam
10-03-2006, 12:28 PM
With training camp starting today, we're going to have real things to talk about rather than the idle speculations and surmises of summer. But thunderbird1245's thread about Harrison (and especially the comments about his propensity to foul) brought back to mind a thought I've often had.

Why not foul often?

Everyone knows the Hack-a-Shaq strategy, which makes sense because the guy is so much less likely to hit from the line than from the rim. Shaq is an extreme case, but the principle applies for most NBA players. If you look at each player's FT% and FG%, the probable outcome of a field goal attempt is higher than the probably outcome of a single free throw for most NBA players (using last season's data). Consider Jermaine O'Neal:

Jermaine's FG% last season was .472. His FT% was .709. So, the probable value of a JO field goal attempt was .472*2=.944 and the value of a free throw attempt was .709*1=.709.

Seen another way, if you foul Jermaine and send him to the line 10 times in a row, he's probably going to score 7 points. If you let him work, he's probably going to score 4 or 5 buckets, worth 8 or 10 points. The other team's defense ought to recognize that he's going to score more if they play him than if they foul.

Now, lemme say I would hate this. I hate the way fouls disrupt the flow of play in the NBA. But the question remains: Aren't fouls before the shot a good way of minimizing the other team's scoring probability? Each team has 12 players * 6 fouls = 72 fouls to give. You could give 67 fouls and still have five guys on the floor at the end. Why not send John Edwards and the other end-of-the-benchers into every game with instructions to foul out in six straight possessions?

Will Galen
10-03-2006, 05:19 PM
Now, lemme say I would hate this. I hate the way fouls disrupt the flow of play in the NBA.

Watching shootouts in soccer gave me an idea. Why not have the players shoot their foul shots after each quarter.

All accumulated foul shots in the fourth quarter would be shot with two minutes to go. After the two minute mark all foul shots would be shot as they are now. Same for overtime.

If a player can't shoot his foul shots because of injury or being thrown out of the game etc., the opposing coach picks someone.

An option would be having teams shoot free throws at the same time thus shortning the length of the game.

Probably not a good idea because other things would also change. For instant, no rebounds after a missed free throw in the first three quarters. Hmmm . . . I could live with that.

How else would this change the game?

mildlysane
10-03-2006, 05:34 PM
With training camp starting today, we're going to have real things to talk about rather than the idle speculations and surmises of summer. But thunderbird1245's thread about Harrison (and especially the comments about his propensity to foul) brought back to mind a thought I've often had.

Why not foul often?

Everyone knows the Hack-a-Shaq strategy, which makes sense because the guy is so much less likely to hit from the line than from the rim. Shaq is an extreme case, but the principle applies for most NBA players. If you look at each player's FT% and FG%, the probable outcome of a field goal attempt is higher than the probably outcome of a single free throw for most NBA players (using last season's data). Consider Jermaine O'Neal:

Jermaine's FG% last season was .472. His FT% was .709. So, the probable value of a JO field goal attempt was .472*2=.944 and the value of a free throw attempt was .709*1=.709.

Seen another way, if you foul Jermaine and send him to the line 10 times in a row, he's probably going to score 7 points. If you let him work, he's probably going to score 4 or 5 buckets, worth 8 or 10 points. The other team's defense ought to recognize that he's going to score more if they play him than if they foul.

Now, lemme say I would hate this. I hate the way fouls disrupt the flow of play in the NBA. But the question remains: Aren't fouls before the shot a good way of minimizing the other team's scoring probability? Each team has 12 players * 6 fouls = 72 fouls to give. You could give 67 fouls and still have five guys on the floor at the end. Why not send John Edwards and the other end-of-the-benchers into every game with instructions to foul out in six straight possessions?

Would make for a VERY long game, especially if both teams deploy this scheme.

Eindar
10-04-2006, 12:07 AM
With training camp starting today, we're going to have real things to talk about rather than the idle speculations and surmises of summer. But thunderbird1245's thread about Harrison (and especially the comments about his propensity to foul) brought back to mind a thought I've often had.

Why not foul often?

Everyone knows the Hack-a-Shaq strategy, which makes sense because the guy is so much less likely to hit from the line than from the rim. Shaq is an extreme case, but the principle applies for most NBA players. If you look at each player's FT% and FG%, the probable outcome of a field goal attempt is higher than the probably outcome of a single free throw for most NBA players (using last season's data). Consider Jermaine O'Neal:

Jermaine's FG% last season was .472. His FT% was .709. So, the probable value of a JO field goal attempt was .472*2=.944 and the value of a free throw attempt was .709*1=.709.

Seen another way, if you foul Jermaine and send him to the line 10 times in a row, he's probably going to score 7 points. If you let him work, he's probably going to score 4 or 5 buckets, worth 8 or 10 points. The other team's defense ought to recognize that he's going to score more if they play him than if they foul.

Now, lemme say I would hate this. I hate the way fouls disrupt the flow of play in the NBA. But the question remains: Aren't fouls before the shot a good way of minimizing the other team's scoring probability? Each team has 12 players * 6 fouls = 72 fouls to give. You could give 67 fouls and still have five guys on the floor at the end. Why not send John Edwards and the other end-of-the-benchers into every game with instructions to foul out in six straight possessions?

I think your math is off.

5 possessions = 10 free throws = roughly 7 points

Essentially, the only reason you would hack-a-Shaq a player is if his free throw percentage is lower than his FG%, or possibly if they were similar and you were running a Grinnell type offense, wanting to use the rebound off a 2nd miss to start a fast break.

skyfire
10-04-2006, 12:59 AM
Yea your maths is wrong. If his ft% is .709, the odds of him converting both free throws in any given trip to the line would be .502 (= 0.709* 0.709).

Ofcourse this doesn't take into account plays where JO makes the basket with a foul and gets to go for a 3pt play.

Fouling all the time would ensure the other team is almost always in the bonus, which could likely be exploited by the opposing teams better ft shooters.

PostArtestEra
10-04-2006, 01:07 AM
This makes no sense. Lets say that the Pacers get the ball 100 times in a game(It would probably be even higher if we were fouled every play, whatever. If the opponent chooses to put O'neal on the line every play that gives him two hundred free throws. Statistics would say that he will make over 140 of those free throws, and he's not a very good free throw shooter. The "foul everybody" method would not work.

Naptown_Seth
10-04-2006, 01:41 AM
Yea your maths is wrong. If his ft% is .709, the odds of him converting both free throws in any given trip to the line would be .502 (= 0.709* 0.709).
And in terms of "points per possession" you also have to figure in the probablity of making 1 of 2 which is where the points jump to the FT% number (70% instead of just .502).

Eindar has put it in the simplest terms, 5 possessions are 5 FGs at .472 or 10 FTs at .702, or 47 points out of 100 possible points vs 70 points out of 100 possible points for FTs.

It's not just a shooting PCT, it can also be thought of as a scoring PCT in this case. PCT of possible points you will score since the number of possible points per possession is equal for either option.

I think you got confused because of the "Adjusted FG%" that takes into account the bonus point you get for a 3pt make vs a 2pt FG.

50 possesions for a 3pt shooter gives you a total of 150 possible points. From those you would get the players 3P% of them. A 50% 3pt shooter would score 75 points on 50 possessions. So if he shoots a FT% less than 75% you would "win" by fouling him before the shot or only on 2pt shots that you can guarantee a miss due to the foul.

How many great 3pt shooters shoot 75% or less from the FT line? My guess is none. Even though a player gets a bonus point on 3pt makes it still requires at least an average FT% and that's for a great 3pt shooter. Dropping it down to 40% requires the player to be as bad as 60% or worse from the FT line.

So the strategy rarely works.

Israfan
10-04-2006, 04:58 AM
Jermaine's FG% last season was .472. His FT% was .709. So, the probable value of a JO field goal attempt was .472*2=.944 and the value of a free throw attempt was .709*1=.709.

You are forgetting that player has TWO free throws on each foul. So the value of a free throw is .709*2=1.418.

One positive of fouling that nobody mentioned yet is stopping game flow and not letting opponents to get "hot".

Shaq is fouled because he FT% is lower than FG%.

Putnam
10-04-2006, 09:20 AM
You are all correct, of course. I was considering only the foul situations of (5) and (6), and not very clearly at that. I was thinking about whether Harrison's tendency to foul hurts the team more than his other contributions benefit it, and then the extreme argument of fouling all the time came into my head.

Just one comment to Israfan. The probability of making both free throws is not .709*2=1.418, but .709*.709=.502 as skyfire, eindar and Naptown_Seth have shown.

Anyway, fouling as a defensive strategy is a horrible thing for the game.

PENALTIES: The offender is charged with a personal foul. The offended team is charged with a team foul if the illegal contact was caused by the defender. There is no team foul if there are personal fouls on one member of each team or the per- sonal foul is against an offensive player. The offended team is awarded:
(1) the ball out-of-bounds on the sideline at the nearest spot where play was interrupted but no nearer to the baseline than the free throw line extend-ed if an offensive foul is assessed.
(2) the ball out-of-bounds on the sideline where play was interrupted but no nearer to the baseline than the free throw line extended if the personal foul is on the defender and if the penalty situation is not in effect.
(3) one free throw attempt if the personal foul is on the defender and there is a successful field goal or free throw on the play.
(4) two/three free throw attempts if the personal foul is on the defender and the offensive player is in the act of shooting an unsuccessful field goal.
(5) one free throw attempt plus a penalty free throw attempt if the personal foul is on the defender and the offensive player is not in the act of attempting a field goal if the penalty situation is in effect.
(6) one free throw attempt and possession of the ball on the sideline nearest the spot where play was interrupted if an offensive player, or a teammate, is fouled while having a clear-path-to-the-basket. The ball and an offen-sive player must be positioned between the tip-of-circle extended in the backcourt and the basket in the frontcourt, with no defender between the ball and the basket when the personal foul occurs. There must be team control and the new play must originate in the backcourt, including throw-ins, and the offended team must be deprived of an opportunity to score an uncontested basket.
(7) two free throw attempts if the personal foul is for illegal contact with an elbow. The elbow foul may be assessed whether the ball is dead or alive. Free throw attempts are awarded whether the ball is dead, alive, loose or away-from-the-play in the last two minutes of regulation or overtime(s). Contact must occur for an elbow foul to be assessed. It is an unsports-manlike act whether or not there is contact. (See Rule 12A--Section VII-- d(6) for non-contact.) If the deliberate elbow contact is above shoulder level, the player will be ejected. If the elbow contact is shoulder level or below, the player may be ejected at the discretion of the official. In all of these situations, the official has the discretion of assessing a fla-grant foul (1) or (2).
(8) two free throw attempts if a personal foul is committed by a defender prior to the ball being released on a throw-in.
EXCEPTION: Rule 12B--Section X.
(9) two free throw attempts if a personal foul is committed against an offen-sive player without the ball when his team has at least a one-man advan-tage on a fast break and the defensive player takes a foul to stop play.

Putnam
10-04-2006, 09:26 AM
Incidentally, just to salvage a bit of value from an admittedly dumb idea, below are listed the NBA players who, like Shaq, do have a worse free throw pct that FG%.

D. Harrison
A. Mourning
J. Hodge
A. Varejao
J. Przybilla
D. Lee
D. Mbenga
L. Woods
D. Smith
T. Chandler
S. Hunter
B. Wallace
H. Seung-Jin
B. Skinner
J. Maxiell
S. O'Neal
A. Davis
A. Bynum
R. Dupree
R. Gaines
A. Biedrins
C. Taft
J. Sampson
A. Acker
S. Medvedenko

Israfan
10-04-2006, 09:59 AM
Just one comment to Israfan. The probability of making both free throws is not .709*2=1.418, but .709*.709=.502 as skyfire, eindar and Naptown_Seth have shown.

Sorry to be a nagger but...

What is important is the mathematical expectation of two free throws attempts which is an average value of two free throws attempts is 709*2=1.418.
While .709*.709=.5027 is the probability to make both free throws, you are neglecting all the points you get from one three throw. If you add this points resulting from one out of two three throw (which is 0.709*(1-.709)*2 first in second out) the total value will be exactly the same:
.5027*2+.709*(1-.709)*2=1.418

Sorry to be a nagger...