clock menu more-arrow no yes mobile

Filed under:

Hoiberg and the Bulls have Failed in Crunch Time

Jeff Swinger-USA TODAY Sports

The past few games have revealed a glaring truth: The Bulls are not a very good team in closing out games.

The evidence from ESPN Stats:

NBA Stats also tracks how teams perform in the clutch, which they define as the last 5 minutes of a game where the score is within 5 points. In these clutch situations, the Bulls rank 22nd in the league in Net Rating. They're getting outscored by 10.4 points per 100 possessions, which is a huge margin (The Lakers have the worst Net Rating in the league at -10.9 for comparison's sake).

These problems were there last year, but not nearly to this degree. Despite all of their struggles under Thibodeau, the Bulls were still the 7th-best team in the clutch and outscoring opponents by a large margin.

How much of the blame can be placed on Fred Hoiberg? The Bulls have not been able to execute his after time-out (ATO) plays this year in some critical situations.

In Dallas with the game on the line, the Bulls couldn't even inbound the ball off Hoiberg's ATO play. The Bulls unbelievably made the same error again last week in Utah. In contrast, the highlight of the season was off a broken ATO where Gasol and Butler ditched Hoiberg's play in lieu of a beautifully improvised lob to win the game.

Forget the high profile blunders for a second though. How has Hoiberg done overall on his out of timeout plays?

Thanks to NBA Reddit-er @HermmanM, we have access to this data. Hoiberg ranks 21st out of 33 coaches (including 3 interim coaches) in points per timeout:

image courtesy of @HermmanM

Hoiberg has been pretty mediocre at ATOs, which is also a good word to sum up his tenure with the Bulls so far.

It's important to note though that these ATO plays are far overstated by fans and media. There's a prevailing sentiment that coaches can win games with brilliant playcalling, but in practice this simply isn't true.

Kevin Broom has done a fantastic job researching this phenomenon for Vice Sports. He notes that ATOs are generally much less efficient than running normal offense:

Numbers support the idea that NBA coaches aren't dramatically affecting games with their ingenious timeout play-calling. An analysis of the play-by-play data compiled at reveals that after-timeout plays (ATO) over the past six seasons average 0.836 points per play—notably less efficient than other half-court plays, which average about 0.885 points per play. This may be due to the fact that by definition ATO plays start from a dead ball, and dead-ball plays are the league's least efficient.

Hoiberg also shouldn't be judged too harshly because there is a ton of variability in these ATO efficiency numbers and most coaches are not great at them.

Given these caveats, it's still fair to note that Hoiberg has made many mistakes in his late game execution and playcalling. Take the Bulls' loss at Denver on Friday as an example.

With Denver holding a one point lead and 51 seconds left to go, the Bulls had the ball on offense and called their final timeout. They subbed McDermott in for Snell and set up this play:

That's a poor use of a final timeout. The play design is extraordinarily simple. All other Bulls players remain stationary while Rose and Gibson run a high pick and roll.

It's not a problem to run a simple play in that situation, but it is a problem to use a timeout to set it up. That timeout could be very valuable later in the game to advance the ball past halfcourt if the Bulls had another chance with not much time left.  As Broom notes, it also gave the Nuggets a chance to set their defense.

In the Bulls' most recent loss to the Timberwolves, Hoiberg called timeout to set up all of the Bulls' last 3 offensive plays (Play 1Play 2Play 3).  The Bulls couldn't execute any of them. They went 0/3 and 2 of the plays resulted in Pau Gasol taking above the break 3's. Gasol entered this season shooting 22% on those shots for his career.

There were other small errors in execution besides settling for Gasol 3's. On the second play, Derrick Rose missed an open McDermott in the corner. On the third, Gasol's poor pick sunk McDermott's chances of getting open in the corner again.

Hoiberg's play designs weren't bad in this game - he had multiple options and movement all over the court. But, as with the offense in general, the Bulls couldn't carry out what their coach wanted. They were outscored 12-0 down the stretch and blew a 5 point lead late in the game.

The lesson here is that Hoiberg, like most coaches in the NBA, has a tendency to overcoach late-game. The Bulls are bad enough already down the stretch and they're not helped by Hoiberg's micromanaging of the offense. The players can't even follow what Hoiberg is diagramming out anyway. Let the guys play, and save the timeouts for clock management purposes.