Thursday, May 10, 2018
Russia meddled to get Trump elected, right? Let's look at the evidence.
Russia meddled in the election to get Trump elected, right? OK, let's examine the evidence. Democrats today released the "3500" FB ads placed by "the Russians" (in reality, placed by a Russian company called the Internet Research Agency, with alleged but totally unproven connection to the Russian government). I downloaded some and painstakingly went through every ad placed in July and September 2016 (remember half or more of the ads were placed after the election, but I wanted to look at the ones which ostensibly might have influenced the election).
Before getting to content, I note that 30 of the 99 ads in July, and 38 of the 86 in September, had ZERO impressions. NO ONE saw them. Why they're even included, other than to pad the number "3500", is unclear. Many of those have the bizarre situation that the ad creation date is after the ad end date (so, no surprise no one saw it!). I can't believe that FB's ad creator even allows this, but apparently it did. On top of that, there are many duplicates in the ads. The same ad is placed two or three or even four days in a row, and each placement is counted as a separate ad. So again that number "3500" (which is already inflated by including ads after the election) is greatly inflated.
That being said, here's some content. In July, 47 of the 99 ads (almost half) were either "Black Lives Matter" ads and/or anti-police brutality (there's obvious overlap between the two subjects). In September, a whopping 75 of the 86 ads fit those categories. Now would anyone like to tell me how such ads were supposed to help elect the racist Donald Trump?
As for the rest, July's ads included 5 Pokemon-related ads and 17 which were in the "funny photos" category, i.e., completely non-political. Only literal handfuls of ads had anything to do with politics. There were 5 anti-Clinton ads in July (fewer distinct ads), but 2 pro-Clinton ads as well. In September, 4 pro-Trump ads (actually all the same ad counted 4 times), and ZERO mentioning Clinton either pro- or anti-. And none either month mentioning Jill Stein.
Still think there was an elaborate Russian plot to elect Trump? Tell me how.
Before getting to content, I note that 30 of the 99 ads in July, and 38 of the 86 in September, had ZERO impressions. NO ONE saw them. Why they're even included, other than to pad the number "3500", is unclear. Many of those have the bizarre situation that the ad creation date is after the ad end date (so, no surprise no one saw it!). I can't believe that FB's ad creator even allows this, but apparently it did. On top of that, there are many duplicates in the ads. The same ad is placed two or three or even four days in a row, and each placement is counted as a separate ad. So again that number "3500" (which is already inflated by including ads after the election) is greatly inflated.
That being said, here's some content. In July, 47 of the 99 ads (almost half) were either "Black Lives Matter" ads and/or anti-police brutality (there's obvious overlap between the two subjects). In September, a whopping 75 of the 86 ads fit those categories. Now would anyone like to tell me how such ads were supposed to help elect the racist Donald Trump?
As for the rest, July's ads included 5 Pokemon-related ads and 17 which were in the "funny photos" category, i.e., completely non-political. Only literal handfuls of ads had anything to do with politics. There were 5 anti-Clinton ads in July (fewer distinct ads), but 2 pro-Clinton ads as well. In September, 4 pro-Trump ads (actually all the same ad counted 4 times), and ZERO mentioning Clinton either pro- or anti-. And none either month mentioning Jill Stein.
Still think there was an elaborate Russian plot to elect Trump? Tell me how.
