It's surely not confusing for us - it may be for the general public though, and of course misleading too. Very unlikely for there to have been any short warmer period - unless it is from within the period since 2015! To answer your original question, yes, I think roughly it is fairly believable - It seems to match fairly well the 'officially accepted' global temp series from various orgs. But that's not really the point is it?
See for example
http://www.cru.uea.ac.uk/ The HADCRUT4 temp series (the British one) which is always displayed on the CRU homepage. The only discrepancy is that 2016 was the hottest year on the whole series whereas in the twitter graph 2015 looks rather warmer.
And if you want to check some of the other global series, without wishing to suggest I read the
Quarterly Journal (I cannot understand most of the article titles!) there is a relevant article by Simmons et al 2017 (QJ v 143, issue 702, 101-119 which can be found at
https://rmets.onlinelibrary.wiley.com/doi/10.1002/qj.2949 and is open access to all. The figure reproduced on that page says it all.......
so, the point is....we must look at the last 150 years before coming to any conclusions about the last three!! The last three years still represents an astonishing peak in global temperatures and the 'cool' 2018 was warmer than anything experienced up until 2014....see
https://crudata.uea.ac.uk/~timo/diag/tempts_decadesmooth_global.png
We still do not know whose twitter account this is from. Do you know, Len?
Julian Mayes Molesey Surrey