The week before the Nobel Prizes are awarded is, among other things, an opportunity to remember that science is a deeply human endeavor.

Star Trek's Mr. Spock would surely deem the frenzy of speculation that precedes the actual announcements – which will begin on Oct. 6 with the prize in Physiology or Medicine – highly illogical.

For the rest of the world, though, the speculation is an integral part of the fun.

Thankfully, despite the instant fame that comes with the prize, the scientific community is pretty safe from naked selfies of its winners. Gossip, though, is fair and enjoyable game – who will win? Will some fairly obvious contenders be honored this year, and if so, will it be in physiology, or chemistry?

Thomson Reuters released its annual Nobel predictions last week, naming citation laureates "whose landmark discoveries and advances, not to mention their measurable esteem in the scientific community, place them within reach of science's highest prize."

Some critics contend that the citation laureate success rate is low. The majority of people on the list will not win in the year they are named.

But to analyst David Pendlebury, that is not the point.

"We have done that before" – for the medicine or physiology prize, most recently in 2009, when citation laureates Elizabeth Blackburn, Carol Greider and Jack Szostak won for their research on telomeres. "But I would say that's a bit of an accident" when the citation laureate wins the same year they are named.

The fact of the matter is that even though most scientists do not do Nobel-quality work, there is still more Nobel-quality work being done by more people than there are prizes to give out.

In some ways, he said, "it's a matter of taste. That's the problem the Nobel Committee has, and that we have, too."

Taste, of course, only comes into play within a very rarefied sphere of quality. An Oreo cookie – even a Double Stuf Oreo – is not a macaroon.

The strength of the type of analysis that is used to name the citation laureates, Pendlebury said, is to quickly identify the high-quality research within any given area of research.

"When you are new to a field, it is an introductory method to find what the key methods are, and who the key people are." The goal is not to predict whether it's the pistachio or the almond macaroon that will win a prize next year, but to say whether research is macaroon or Oreo-quality in the first place.

For the Thomson Reuters analysis, the process of naming citation laureates starts with looking at scientists whose works are highly cited. That narrows the field considerably: Of roughly 58 million papers published in the last 50 years, only about 15,000 have been cited more than a thousand times.

Pendlebury warned that using more citations as an index for higher quality only works well with papers that are extremely highly cited. At lower frequencies, he said, there is "a lot of false precision." Whether a paper is cited 75 times or 150 times can be due to many different factors besides the quality of the research – the field, the journal, the power of the author.

But "at these ultra-high frequencies, the signal to noise ratio is very good," he noted.

Even at ultra-high frequencies, though, citations are only a partial indicator of quality. Some of the most highly cited papers are methods papers that have given scientists new ways to look for answers without really changing the intellectual framework of which questions get asked in the first place.

Some methods developments papers, on the other hand, do lead to new ways of thinking about questions in the first place. Pendlebury noted that he would not be surprised to see functional magnetic resonance imaging (fMRI) be honored by the Nobel Committee. "It has had multidisciplinary impact," he said. "They use fMRI now to watch traders on the stock exchange making economic decisions."

Beyond citation frequency, there are more qualitative indicators of who might be picked. Previous winners matter – the committee is unlikely to pick the same subspecialty twice in a row. So do previous wins – some awards, like the Lasker Award or the Wolf Prize, are good statistical predictors of future winners.

And finally, there's the surprise factor. "Findings that overturn dogma seem to be particularly attractive to the Nobel Committee," Pendlebury said.