The temperature and the spin
Jan 17, 2015
Bishop Hill in Climate: Statistics, Climate: Surface

Many scientists on the whole seem to have been suitably cautious about alleged record-breaking temperatures, taking care to place the new data in the context of the error bars. It's also fair to say that others have been a bit wild.

The Science Media Centre has a couple of moderately level-headed responses, from Tim Palmer and Rowan Sutton, but as always with the SMC it's seen as important to get some input on climate change from a paleopiezometrist, from whom we learn that:

The new global temperature record announced today completely exposes the myth that global warming has stopped.

And if that isn't a lot of hoary old tosh I don't know what is.

However, giving credit where credit is due, even Bob's contribution has been put in the shade by Seth Borenstein of the Associated Press:

So how likely are these temperatures to be random? The Associated Press consulted with statisticians to calculate the odds of this hot streak happening at random. Here are some statistics and the odds they calculated, with the caveat that high temperatures tend to persist so that can skew odds a bit:

The three hottest years on record — 2014, 2010 and 2005 — have occurred in the last 10 years. The odds of that happening randomly are 3,341 to 1, calculated John Grego of the University of South Carolina. Kai Zhu of Stanford University, Robert Lund of Clemson University and David Peterson, a retired Duke statistician, agreed.

Nine of the 10 hottest years on record have occurred in the 21st century. The odds of that being random are 650 million to 1, the statisticians said.

So persistence can skew the odds "a bit" can it? Others begged to differ:


@borenbears @ClimateSystem Golly. That's astonishingly bad statistics.

— Bishop Hill (@aDissentient) January 16, 2015

Sorry to have occasion to say this but @aDissentient is correct @borenbears @ClimateSystem Numbers are based on an invalid simplification

— mtobis (@mtobis) January 16, 2015


 Interestingly, it was not just Borenstein who had been spinning this particular yarn. The great sage of "modern" statistics had tweeted something similar:


"There is less than a 1-in-27 million chance that Earth's [2014] record hot streak is natural" @AFreedma @Mashable:

— Michael E. Mann (@MichaelEMann) January 16, 2015


...despite several scientists having pointed out the error of his ways:


Connolley, Annan, Grumbine & I iirc tried without success to dissuade him @aDissentient @AndyMeanie @borenbears @ClimateSystem

— mtobis (@mtobis) January 16, 2015


Peter Gleick had been doing likewise, with a corrective issued by CRU's Tim Osborn:

@PeterGleick 1-in-27 million looks way wrong. Probably poor assumptions about persistence, natural variability

— Tim Osborn (@TimOsbornClim) January 16, 2015

It's amazing to see how this kind of disinformation (misinformation?) gets generated and disseminated. If I, a humble blogger, can detect the error at a glance, why can't people whose job it is to uncover and communicate the truth about climate get it right? Why was Borenstein's contribution retweeted by the  "ARC Centre of Excellence for Climate System Science" at the University of New South Wales? Why was Mann's regurgitated by a climatologist from Copenhagen?

Update on Jan 17, 2015 by Registered CommenterBishop Hill

Here's another:


I was also amused to see this being retweeted by someone who is interested in improving science students' mathematical understanding.


Article originally appeared on (
See website for complete article licensing information.