Guessing who will win a Nobel Prize is a bit like forecasting the stock market: Experts don't seem to do it any better than laymen.
So if you hear professors and pundits predicting the "God particle" will be the theme of the physics prize next week, or that an American writer - finally - is due for the literature award, check their track record.
"My top candidate has never won, and it's the fourth year I do it now," admitted Norwegian peace researcher Kristian Harpviken, one of the most prominent voices in the annual guessing game for the Nobel Peace Prize.
A week ahead of that announcement, the Irish online betting agency Paddy Power gave the lowest odds Friday to retired American scholar Gene Sharp, Afghan women's rights activist Sima Samar and Tunisian blogger Lina Ben Mhenni. All have been among Harpviken's top picks in recent years.
Harpviken, who heads the PRIO peace institute in Oslo, admits his speculation is just that - speculation - based on current events, previous prizes and personal preference.
"I think guessing is important in that it brings attention to what the Nobel Peace Prize should be about," he said. "I would be very hesitant to speculate on a certain candidate who is absolutely undeserving."
The secretive prize committees rarely drop any hints and Harpviken doesn't have any inside information. Virtually none of the Nobel guess-makers do - but that doesn't stop them from trying.
The peace and literature prizes generate the strongest buzz, and are typically less difficult to predict than the awards for chemistry, physics, medicine and economics.
The six award committees will announce one prize a day, starting with medicine on Monday and ending with the economics award on Oct. 15. The Nobel Foundation this year lowered the prize money 20 percent to 8 million Swedish kronor ($1.2 million), citing turmoil on financial markets. All prizes will be handed out on Dec. 10, the anniversary of prize founder Alfred Nobel's death in 1896.
It would have been easier to guess the winners if the Nobel committees had stuck to the will of the Swedish industrialist, who wanted the annual awards to reflect the greatest achievements "during the preceding year." Instead, the Nobel statutes were changed so that committees can reward discoveries made decades ago, to make sure they have stood the test of time.
"I think Alfred would have been OK with that," said Per Carlson, a former chairman of the physics prize committee.
Handing out a prize too soon increases the risk of jurors failing to identify the right scientists behind a discovery, Carlson said.
That happened in 1974, when the Nobel Prize in physics went to British radio astronomers Sir Martin Ryle and Antony Hewish. The latter was cited for the discovery of pulsars - rotating neutron stars - though it later became clear that one of his graduate students deserved the credit.
The Nobel time lag could hurt the chances of the most talked-about scientific breakthrough this year: the identification of the Higgs boson, a subatomic particle also referred to as the "God particle."
Though British scientist Peter Higgs predicted the existence of the particle in the 1960s, it was only in July this year that scientists at an atom-smasher outside Geneva claimed to have identified it.
Whether Higgs was even nominated is unclear - the deadline was in February and nominations are not revealed for 50 years.
Secrecy is paramount to the Nobel committees. Literature jurors have been known to use code words when discussing Nobel candidates and using fake book covers when reading their work.
After suspected leaks ahead of recent announcements, the literature panel has taken additional measures this year. A press release declaring the winner will no longer be delivered by courier to the offices of major news organizations in Stockholm, including The Associated Press. And the panel's permanent secretary, Peter Englund, has stopped giving his customary interviews in the weeks leading up to the prize.
"We just think it's better this way," Englund told the AP in an email.
Betting on the literature prize is in full swing, with Ladbrokes offering 3/1 odds for Japanese writer Haruki Murakami and American singer-songwriter Bob Dylan - an outsider whose name appears in the Nobel buzz every year though the committee has never given the prize to a musician - placing fourth at 10/1. The last American to win the Nobel Prize in literature was Toni Morrison, in 1993.
In perhaps the most elaborate example of Nobel guesswork, the scientific division of news and financial information provider Thomson Reuters analyzes thousands of citations and papers in high-impact academic journals to identify possible winners of the science and economics awards.
Even though they allow themselves three guesses for each of those four prizes, the Thomson Reuters analysts led by David Pendlebury haven't predicted a winner since the 2009 medicine award. Of the 158 scientists they have identified as potential Nobel laureates since 2002, 26 have won, but only nine in the year they had predicted.
"If we do get any of these right it's all the more remarkable, because we have no field expertise," said Pendlebury, whose academic background is in ancient history. The main difficulty in making forecasts, he said, is that "there are more people of Nobel class than there are Nobel Prizes to go around."
The scholars and literature buffs speculating on Nobel Prizes may be different from the crowd predicting scores in sports bars, but the psychology is similar.
When accurate guesses are rare, a person making an inaccurate guess can easily dismiss it by saying that most other people got it wrong too, said Jiao Zhang, a professor at the University of Miami who has studied the dynamics of guess-making.
"Whereas in the rare case when his guess is accurate, he can say to himself, `I'm among the few who got it right,'" Zhang said in an email. "Vindication and validation are sweet."
Economics - which wasn't in Nobel's will but was tacked on to the other prizes by Sweden's central bank in 1968 - is arguably the most difficult discipline to predict. Unlike Nobel-winning scientists who discovered things like X-rays or the AIDS virus, economics laureates are typically rewarded for abstract theories about market behavior.
Hubert Fromlet, a professor of international economics at Sweden's Linnaeus University, each year offers a "narrow" list with the 10 economists he believes are most likely to win. He also provides a "relatively narrow" list with 20 names and a "broader" one with 40.
But with 40 names, does it really count as I-told-you-so if one of them wins?
"Of course I feel happy when I have someone in the Top 40," Fromlet said. "But if you have someone in the top 10, it's a little bit more fun. It's not a competition. It's just something I like to do."
Associated Press writer Louise Nordstrom in Stockholm contributed to this report.