Certainly an argument could be made that that AP announcement the night before suppressed voter turnout in CA and in other states such as NM, SD and CA. Clinton might have gotten the needed number of delegates even if she had lost those three states given that she had a big lead in NJ in the pre election polls there but it would've been embarrassing for her going up against Trump. Two million votes have yet to be counted in California and in Puerto Rico as is discussed in the clip below..
Complaints of media bias against Sanders were many long before this last round of primaries. The media watchdog group FAIR (Fairness and Accuracy in Reporting) published many studies showing how little air time was given to Sanders relative to Trump. The challenge now for Sanders' supporters is not only to advocate for his ideas. It is to level the playing field for candidates like him. The struggle is going on now to curtail super delegates which like open primaries were created to favor more moderate candidates like Clinton as Mother Jones reports.
Another FAIR story on bias against Sanders was directed at internet stats guru Nate Silver. It argues that he has consistently made negative predictions against Sanders. Many of which have been wrong such as the one in the photo at the right. Silver has also been consistently wrong on Donald Trump predicting that he would fade like the rest of the GOP establishment assumed he would.
In the last two Presidential elections which more or less followed the conventional wisdom, Nate Silver was amazingly correct at predicting the outcome of elections. This election year has been one of the least conventional that people can remember and Silver's prediction rate has taken a hit. He is now predicting a close general election based on current polling.
Predictions are based on certain assumptions which may or may not be valid. An astrologer's predictions rest on the erroneous assumption that the stars in the sky influence people's behavior. Conventional wisdom can cloud the inferences of even the most expert statisticians. It is important to make as few assumptions as possible. In the end it is the people who determine the outcome of elections not the supposed experts.