By the same metric we could question January 1977 at around an 11 degree average, as the coldest month of all time by around 6 degrees. Now there are people who can corroborate that as many were alive then, and it is an actual temperature. How do we know September 1881 wasn't that warm? We've seen very warm Septembers, maybe that was on steriods. It can happen. Plus that September recorded a few 100 degree days, which has never happened since, oh except in 1884.
I also find it interesting that it seems we don't see highs that are as extreme as they used to be. Looking through the records it seems years ago we had yearly maxes that were vhigher than today. Taking each decade since 1940 Most are relatively close, with the 40s having the highest decadal mean high at 95, the 2000s are the lowest at 91. The last decade (2010s) was 93. 40s were 95, 50s were 94, 60s were 94, 70s were 92, 80s were 94, 90s were 94, 00s were 91 and 10s were 93. While they're all close, as averages it seems significant that our highs don't seem to be getting quite as high as they used to.