To answer the question would require us to know what the predicted global surface temperature at this point in time?
There are several problems with that, the first being that the models did not model the surface temperatures,
but we do have some reference points.
An article in Nature in discussing the global warming hiatus, talks about what they expected.
Climate change: The case of the missing heat : Nature News & Comment
Since the article was written in 2014, and the reference data has changed, I will refer to the current data.
http://www.metoffice.gov.uk/hadobs/...time_series/HadCRUT.4.4.0.0.annual_ns_avg.txt
In 2014 they said the 2 decades preceding 1998 had warming of 0.21 °C per decade, that they expected to continue.
1998 .536
2015 .746
delta .21, .21/1.7 decades = .12 C per decade
Under the criteria used in the article, the current temperatures are .09 X 1.7=.153 C below the predictions.
2015 is questionable data however as 25% of the year is contaminated with strong El Nino temperatures.
(I know that they used another El Nino year (1998) in the original warming calculation, but it is constrained
within their model already.)
If we evaluated only to 2014 (pre El Nino) the numbers change quite a bit.
1998 .536
2014 .567
delta .031, .031/1.6 decades = .019 C per decade.
In this case the error from the predictions is much greater, .21-.019= .191 X 1.6= .305 C below prediction.
What all this really shows, is that the noise of the normal temperature cycles is is so great that
sliding one year can double the per decade rate.
Our climate system is very noisy for the surface temperatures, which is why the models,
model the surface troposphere system.