Does Terping Work?

Since returning to the ACC to coach his alma mater, Maryland's Gary
Williams has been quick to attribute losses to North Carolina-based
ACC opponents to biased officiating. He'd often imply there was a
conspiracy by the ACC to prevent Maryland from winning. While this
initially looked like an effort to inspire his team to play better,
one should not deny the second potential benefit of spreading the
seeds of doubt.

This year, it seems that
the seeds of doubt may have borne fruit. Watching Duke this year,
one notes how frequently fouls are called on the Blue Devils and how
similar plays by opponents may not be whistled. There has arisen
a general feeling among Duke fans that the officiating in the ACC
has been bending over backwards to appear to be unbiased, to the
extent that there is a bias, unintentional (or not?), against
Duke.

After certain
events, we became interested to see if the anecdotal evidence was
supported by the statistics. Over the course of the year, we noted
that it seemed to be open season on some Duke players, notably
JJ Redick. Only three times in 19 games was a foul called on JJ's
defender when he was shooting beyond the arc. This, despite the
"word on the street" that the way to defend against JJ's shooting
was to hit him when he shoots. Last year, that foul call was
routine.

What brought this to a head was the game against Georgia
Tech in Durham where JJ was effectively tackled while shooting,
and no call was made. (Correction, a technical foul was called on
Coach K by Karl Hess after he went ballistic over that no-call and
other calls in the game.)

The next game, Duke was pounding the
ball inside against North Carolina to the extent that 32 of Duke's
36 first half points were in the paint. Normally, you'd expect
to see this kind of aggressive attacking of the basket to draw
fouls -- Duke shot a grand total of four foul shots on the half.
Duke continued to attack the basket in the second half, but despite
four late fouls on North Carolina as they were fouling to get the
ball, more fouls were called on Duke for the game.

In the ACC
Tournament, Isma'il Muhammad commited a blatant intentional foul on
a Daniel Ewing layup. It was clear to the photographer who got the
picture of Daniel's shirt being pulled from behind, but apparently
Karl Hess's officiating crew were unable to see that. Karl did call
a technical on Coach K, though. This mis-call was displayed for all
to see on the front page of the sports section, and was prominently
shown to Karl Hess when he came out to officiate the ACC Championship
Game.

Once more, there were some interesting calls in this game.
It was notable that when Nik Caner-Medley lowered his shoulder to
drive to the basket against JJ Redick, this was a blocking foul.
99 times out of 100, when a player lowers his shoulder to push a
defender out of the way, it's called a charge. There were similar
times in the last few minutes when fouls should have been called on
Maryland defenders (notably when Chris Duhon was driving for a hoop)
but were missed by this crew. Mind you, we lost because when we did
get a chance to go to the line, we only hit 51% of the free throws.
One more free throw in regulation, and the game would have been
won.

After Duke's defeat of Xavier on Sunday, the standard rant of
"Duke gets all the calls" was started again by members of the media.
This, despite the fact that 23 fouls were called on Duke, as opposed
to 21 on Xavier. Xavier was in the bonus with over 13 minutes
remaining the second half, when they had only two team fouls, for
the middle 16 minutes of the second half, Xavier went to the line
14 times as compared to Duke's three trips...

All this merited
further investigation. We went through Charlie Board's data in
detail, examining the data for all ACC teams and collecting the number of fouls for and against per team,
and creating a new measure, the foul-ratio. We discarded any team's
data if fewer than 10 games of data was present. This resulted in a
collection of 161 sets of points, illustrated below.

Clearly, examining the graph,
there is a trend to having a higher number of fouls called on
losing teams. The next step was to attempt to quantify this trend.
We first attempted to fit a straight line, y=mx+b
using a least squares approach. The slope of the resultant
line was -0.37, showing a clear trend towards fewer fouls
called on good teams.

Also, just to see if
it made any difference, we tried a trinomial curve.

As expected,
there is a trend towards having a winning team have fewer fouls
called on them. So, we come to the 2004 Duke men's basketball team,
and we look at the numbers. First, we see that in 19 ACC games,
Duke had 393 fouls called on them, and only 401 fouls called on
their opponents. This is a foul ratio of .980. While not the
worst, it is well higher than the normal for a team that won 78.9%
of their games. Here's the historical data showing foul ratios for ACC teams against their winning percentage:

Duke 2004 in bold

Interesting, no?

Next, we took a look at the regular season league winners, to see where this fit. Here's the complete set of data for league leaders, where there were at least 10 box scores present. (League leaders with fewer than 10 box scores are not listed.)

Summing up the data, we find the normal foul ratio for an ACC regular season winner is .8689; for every seven fouls called on a league leader, 8 fouls are called on their opponent. (Of Duke's last six regular season championships, three were higher than this ratio, and three were lower.)

The first thing that stands out is the 393 foul calls. This is the most fouls called on any regular season winner in Charlie Board's dataset, but it also reflects an increase in the number of fouls called, generally. Still, as the graph below indicates, the foul ratio of .98 is unusual.

How have we reached this point? Gary Williams has been terping about fouls for over a decade, so what changed? We suspect the turning point was in the 2001 Final Four, when Billy Packer was calling the Duke vs Maryland semifinal. Early in the game, Maryland was whistled for a foul when Casey Sanders was putting up a follow shot. Packer's comment when it was called a shooting foul was "Duke got a break there." Never mind that the NCAA rules indicate that when the player starts his shooting motion, the shot is being attempted. Packer incorrectly assumed that for the foul to be a shooting foul, the ball would have had to leave the shooter's hands. Then, at 16:33 remaining in the first half, a second foul was whistled on Maryland, and Billy Packer said, "Wow, Duke's getting a lot of the calls so far." Of course, what Packer ignored was that two team fouls had already been called on Duke. So, in Packer's interpretation, "Duke getting all the calls" means that the number of fouls called were even. When Gary Williams complained to the officials, "How much do you want Duke to win?" Packer's commentary in the game only reinforced the belief that there was an officiating conspiracy favoring the Blue Devils.

Packer compounded the mistake in the championship game. One of his favorite lines about Duke was "Duke makes more foul shots than their opponent attempts." Going into the game, it was true: Duke made 684 foul shots in their first 38 games, and allowed 682 opponent foul shots. By the end of the game, it was no longer true for Duke, but it was still true for Arizona. At no point did Packer mention the larger margin for Arizona:

There's nothing like incomplete reporting to reinforce a stereotype. Since this Final Four, it has been generally accepted among non-Duke fans that Duke "gets all the calls," and with the epidemic of envious Duke hating, the standard lie is reinforced.

(It should be noted that all four teams in this year's Final Four have not hit more foul shots than their opponents attempted.)

We do not believe there is a general conspiracy among officials to call Duke games harsher on Duke than previous years,
anymore than we believe there is a conspiracy amongst the ACC, the national
media, the NCAA, and the officials to benefit Duke, but we would be surprised if the general
rumor-mongering by the sports media hasn't at least subconsciously changed the way an official might call the game. We have seen some subtle
changes - officials standing away from the Duke coaches at timeouts, without a similar distance to opposing coaches, and the like.

As Duke fans,
like anyone else, all we want is the game to be called evenly; if something is called a foul on Duke, then when an opponent does similar, it should be a foul on Duke's opponent. If you stand away from the Duke coaches, stand away from the opponent coaches, and vice versa. Let the game be decided by the young men of the teams, not the
myths and prejudices of the media.