Thursday, January 20, 2011

Fouling up three, under seven seconds left

After Georgetown's 5-point win over Seton Hall on Tuesday, there was a bit of doom-and-gloom here and elsewhere about the Hoyas.  Ray Floriani, erstwhile contributor here and still blogging elsewhere, made it to the game, and sent along this missive.

NEWARK, NJ - A confession is in order. I had an officiating assignment of two high school games in Jersey City and got to the Rock at the half. The trip was only eight miles and at that time, there was little traffic to deal with. Our last high school game they turned the heat down, so media room coffee at the Rock hit the spot. At any rate…

Impressed with…Georgetown poise. You lose a seven point halftime lead and are now down seven with twelve to play as the home team is in the midst of a 21-7 run; many would have panicked. During the Seton Hall rally the Hoyas stayed together. No finger pointing or the like. They just maintained focus until they could get out of the slump. On the bench coach John Thompson III also maintained his cool. The only outburst of displeasure would be off a careless unforced turnover [ed. note: that was Nate Lubick's behind-the-back pass in the general direction of Julian Vaughn].

Also impressed with Austin Freeman. How could one not be? He had 28 points, and hit 10 of 13 from the floor including a few big ones at crunch time. Even Seton Hall coach Kevin Willard said “Freeman is fun to watch.” Just not too much fun to try and stop.

Correct me if I’m wrong…But I would like to see a few more touches for Julian Vaughn. He has size and a low post presence. Vaughn is attempting 16% of the Hoya shots in conference play. Even utilizing him as a passer wouldn’t hurt. Vaughn has 11 assists, third on the Hoyas in Big East play. More touches and shots could mean more trips to the line where he is a 72% shooter. As noted, just a thought.

Rebounding . . . Read through this site some links and essays on Georgetown rebounding. In raw numbers the Hoyas were 36-35 in favor. The OREB percentage saw them trail the Hall 42-40%. That was largely due to a better second half. The first 20 minutes they had a 41-27% OREB pct deficit, but still led by seven. A 67% eFG mark during that time was largely responsible.

Coach Thompson mentioned in one piece that rebounding was not just on the big men. The guards contributed as Chris Wright and Jason Clark both had 6 (tied with Vaughn for the team lead on the game) and Freeman had 5.

Ray also sent along a link to another post he had on the game, discussing Coach Thompson's late game strategy of fouling when up by three and time for probably one decent look from deep by Seton Hall (link).

The quotes from the coaches after the game are interesting:
Asked why he chose the strategy, Georgetown coach John Thompson III said partially in jest, “because Julian (Vaughn) promised to get the rebound.” In a more serious vein, Thompson said it is a strategy he would regularly employ in that situation.

Kevin Willard of Seton Hall agreed with Thompson’s move saying he would use a similar tactic in that scenario but had a plan for his club. “If we hit the first shot,” Willard said ,” we were going to miss the second and hope to put the rebound in (to tie). But that all went out when we missed.”

There have now been several studies performed to look at exactly this situation - whether it makes sense to foul when up 3 and just a few seconds left. John Ezekowitz looked at how well this strategy worked for the 2009-10 college basketball season (link), and concluded that it's neutral - both fouling and not-fouling worked equally as well.

Then Luke Winn followed that up with an article at SI (link) looking at the theoretical odds for the play, as calculated by Bill Fenlon, who coaches at DePauw University.
In Fenlon's circle of coaching friends he was the Evangelist of Fouling Up-Three Under Seven Seconds. He hoped Stevens was on the same wavelength, thinking the same thing: You HAVE to foul here.

Fenlon once commissioned the drawing of mathematical decision trees on the subject, and believes in the following odds:
  • A 19 percent chance of getting sent to overtime if you hunker down and play no-fouls defense.
  • A 4.9 percent chance of getting sent to overtime if you foul correctly.
  • A 0.67 percent chance of going to overtime by accidentally fouling a three-point shooter.
  • A 0.13 percent chance of losing in regulation by accidentally fouling the shooter.
  • A 0.16 percent chance of losing in regulation off a made first free throw, missed second, offensive rebound and a three.
Now I'm not going to sweat those odds too much - obviously they are strongly dependent upon how well the trailing team shoots threes, as well as how well each team rebounds - but the general upshot is that there's about a 1 in 5 chance of going to overtime if the trailing team gets a shot off behind the line, and about a 1 in 20 chance if the leading team foul correctly (i.e. before a shot attempt).

At the end of the Georgetown - Seton Hall game, the Pirates in-bounded the ball down three with nine seconds left, and Austin Freeman gave a foul with five seconds left.  This was exactly the situation described above.  But were there any additional circumstances that made the odds better or worse for the Hoyas to want to foul?

Seton Hall had the following players on the court:
Player      Height    3FG%     FT%     OReb%
Theodore     6'0"      31      82       2.5 
Lawrence     6'2"      28      72       2.6
Hazell       6'5"      47      77       1.4
Polynice     6'5"       8      59       4.0
Robinson     6'6"      34      78       6.2
Freeman fouled Lawrence, about the ideal candidate:  he was the worst FT shooter of those likely to handle the ball.  Also, he isn't a great 3FG shooter, so he's more likely looking to pass to Hazell or Theodore rather than shoot (and therefore there's less chance of fouling in the act of shooting).  And indeed, Lawrence was running a play to hand off the ball to Hazell when Freeman grabbed him.

But the most important bit of info is who wasn't on the floor:  Herb Pope, and his 12.4% OReb rate.  Meanwhile Julian Vaughn and Nate Lubick were on the court for the Hoyas, and they're much bigger than the Pirates' Robinson and Polynice.  So the most important step for Seton Hall - grabbing a missed FT - was going to be a difficult proposition.

I should note that JTIII subbed in Lubick for Hollis Thompson before the free throw:  Thompson is actually a marginally better rebounder than Lubick, but Lubick is the better FT shooter (which assumes that whoever grabs the rebound will get fouled and may need to make 1 or 2 free throws to ice the game).  I don't have a problem with that move.

Sure enough, Keon Lawrence missed the front end of his 1-and-1, Vaughn grabbed the board and quickly passed the ball to Austin Freeman (and then on to Chris Wright), and that was the game.

Coach Thompson made the correct strategy call, and the team - namely Austin Freeman and Julian Vaughn - executed it perfectly.

Stats pages will get updated tonight.


  1. While Hollis has slightly better rebound numbers than Lubick, I suspect Lubick might be the better option in that situation because it is basically a block out mission and he has more bulk.
    Hollis rebounds fairly well but he tips a lot of balls. We needed Seton Hall to have to go over the back to get to the ball.

  2. Yeah, if it were me I would have subbed in Lubick for Clark, not Thompson.