Actually, BIJ's numbers actually are good as their methodology is based on finding multiple sources for drone strike incidents. (
Covert US strikes in Pakistan,Yemen and Somalia – our methodology: The Bureau of Investigative Journalism)
While they do use Pakistani, Yemeni, and Somali sources, their assertions are also backed up with evidence from "the New America Foundation, Critical Threats, Long War Journal, Jamestown Foundation, Jihadology, Empty Wheel, Wired, WikiLeaks, the UN and Amnesty International" and "CNN, MSNBC, ABC News, Fox News, Reuters, the BBC, Associated Press, the Guardian, the Telegraph, the Independent, TIME, the Wall Street Journal, the Washington Post, the New York Times, the Los Angeles Times, the Nation, the Atlantic, Salon, Xinhua, Army Times, Navy Times, Bloomberg, AFP, NPR, Al Jazeera, and Al Arabiya." Thus, the sources are quite diverse and thus the possibility of bias decreases.
You say that both the Long War Journal and the New America Foundation "have estimates for militant lows and highs and civilian lows and highs, they break down as 1,600-2,800 and 150-190 respectively." However, I have to ask, in what month and year did you post this? The data changes frequently. Secondly, you rely mainly on LWJ and the NAF. Both of which are quite good sources, no doubt, but one cannot possibly get a full picture of the civilian casualties that drones have when focusing on only two sources.
In 2012, Columbia Law School's Counterterrorism and Human Rights Project, in conjunction with the Center for Civilians in Conflict conducted a study in which they concluded that (
http://web.law.columbia.edu/sites/d.../files/Drones summary and recommendations.pdf)
The study (
http://web.law.columbia.edu/sites/d...itute/files/The Civilian Impact of Drones.pdf) itself states that US drone strikes actually
hurt US interests in Pakistan.
It also notes that drone strikes can potentially create terrorists: