E.g. see http://www.washingtonmonthly.com/magazine/january_february_2...
"in investigatory stops, a black man age twenty-five or younger has a 28 percent chance of being stopped for an investigatory reason over the course of a year; a similar young white man has a 12.5 percent chance, and a similar young white woman has only a 7 percent chance. And this is after taking into account other possible influences on being stopped, like how you drive. " [in other words, this sample has been corrected for any difference in base-rate of justification for being stopped].
The interesting thing is, profiling like this can be a self-fulfilling prophecy; if the base-rate of drug possession is equal between blacks and whites, but blacks are stopped more often, then there will be a higher number of arrests of blacks per-capita, and it will look like blacks are more likely of committing a crime. This higher crime rate looks like a justification for profiling, when in fact it's just an artifact of the profiling that was done.
[Note I'm not making any claims about the actual base rate of drug possession, just illustrating an effect with an example]
FTFY
So if the police get complaints from neighbors they respond to that. If your rich neighbors tolerate your coke addiction, they don't come knocking. If you have a noisy neighbor who complains they do come knocking. Police respond very much to community complaints, from my experience with them growing up.
Whenever the police came to "bust" activities, it was mostly due to neighbors calling in "suspicious activity" I.e. Underage drinking and weed.