North Carolina continued their inconsistent ways yesterday with an overtime victory over Davidson. UNC picked up the pressure in the final few possessions of regulation. Davidson struggled taking care of the ball, but managed to put themselves in a position to win the game with the shot clock off. Based on the final minutes of the game, it’s probably no surprise that Roy Williams decided to trap Davidson instead of simply letting the clock run down. The final possession is below:
Under regular conditions, UNC’s trapping strategy is probably a risk. The Tarheels are sacrificing shot-defense for the chance at a steal. Teams tend to generally be contempt with letting the clock run down and playing for overtime. As previously mentioned, Davdison had been struggling against the pressure. There’s no doubt personnel is a big factor here. However, let’s take a look at the decision to gamble defensively in a vacuum:
Outcome 1: Successful trap
Obviously this is the preferable outcome for the defensive team. A turnover would have given UNC a chance to actually win the game in regulation. A live-ball turnover has an added bonus of a potential easy basket on the other end. However, a live-ball turnover could still leave time on the clock for another possession. I’d estimate that creating a turnover in this situation probably increases win probability to around 70% assuming two equal teams.
Outcome 2: Offense stalled
This is what happened in the UNC game. The trap didn’t create a turnover, but it did force Davidson into an extremely difficult shot. However, teams do tend to go to hero ball in these situations anyways (regardless of the defense). My educated guess would be that shooting percentage goes slightly down compared to when a defense simply plays straight up. Barring a fluke foul, the only way for the defense to win here is in overtime. Win probability is somewhere around 35% again assuming two equal teams.
Outcome 3: Failed trap
Here’s where the risk comes into play. The trapping team is one poor rotation away from allowing a wide open shot. Shooting percentage will go way up if this is the case. Still, the trap forces the offensive team to go earlier than desired. UNC would have a chance to tie (or possibly win) the game with a reasonable amount of time left on the clock. I’m not saying that a completely failed trap would put UNC in a good position, but it might be slightly better than you would expect at first thought. This is where estimating win probability gets pretty tricky, but somewhere around 20%-25% seems about right.
Being forced to defend with the shot clock off in a tied game isn’t a desirable situation. No coaching move is ultimately going to change that. However, I think there is some merit to not allowing the offense to take the clock all the way down for the final shot. This strategy becomes even more favorable with a great trapping defense and/or a great isolation player on the other team. Think about how the offense acts in these situations. They generally won’t even look at the hoop until around eight seconds left. Coaches value getting the last shot over getting the best shot. Yet these same coaches don’t seem to be willing to take a risk when the roles are reversed. I’ll try to monitor these situations going forward.