Intrade may have shut down, but the election of a pope still offers a good chance to study the predictive value of betting odds. Eight years ago, when Cardinal Joseph Ratzinger ascended to St. Peter’s throne, the oddsmakers looked prescient: he was the favorite, or a co-favorite, in several places. Thirty-five years ago, the bookies - and, by extension, the wisdom of crowds - did much less well: Cardinal Karol Jozef Wojtyła, soon to become Pope John Paul II, was nowhere among the favorites.
This time, the favorite is Angelo Scola, the archbishop of Milan, as he was when we last checked in with the betting markets on March 4. A 71-year-old Italian with anintellectual bent, Cardinal Scola has a chance of between 25 percent and 33 percent of being elected, according to various oddsmakers.
But after Cardinal Scola the oddsmakers’ choices have shifted a bit. Cardinal Odilo Pedro Scherer, of Brazil, who was in ninth place with just 6 percent odds a week ago, is now just behind Scola, with a 22 percent chance, based on the odds at Paddy Power on Tuesday night.
Then comes Cardinal Peter Turkson, of Ghana, with a 15 percent chance; Cardinal Marc Ouellet, of Canada, at 11 percent; Cardinal Sean O’Malley, of the United States, at 9 percent; Cardinal Tarcisio Bertone, of Italy, at 7 percent; and Cardinal Christoph von Schönborn, of Austria, at 7 percent. Oddschecker.com has a fuller list - updated frequently - with even longe! r shots, although it lists the odds as betting lines rather than as percentages. (The odds also add up to more than 100 percent, so it is probably reducing each of the above odds a bit.)
There are no polls of the 115 cardinals who will be making the decision - at least no public polls - and no are other variables, like economic growth, with a clear record of predicting the winner. So everyone following the conclave, from mildly interested observers to bettors to observant Catholics, is left to piece together fragments of information, much of it of dubious quality.
“What public info is there to aggregate†asks Justin Wolfers, an economist at the University of Michigan, who has studied prediction markets. Even the odds on Oscars winners seem likely to be more accurate, he added: “It’s both a larger set of voters -large enough that they aren’t voting strategically - and the info they are using is very public.While we’ve seen the best movies, few of us have read the writings, or attended the sermons, of the various cardinals.â€
But if the papal odds are less reliable than the odds on, say, sporting events - which have a very good overall record - they still may be the most useful publicly available predictions. As this blog has noted in the past, pundits and high-profile experts generally have an abysmal record with forecasts. Louis Menand, writing in The New Yorker in 2005, summarized some of t! he resear! ch, much of it by the psychologist Philip Tetlock. “People who make prediction their business â€" people who appear as experts on television, get quoted in newspaper articles, advise governments and businesses, and participate in punditry roundtables â€" are no better than the rest of us,†Mr. Menand wrote.
The last two papal elections are good examples. Shortly before the 2005 conclave, CNN.com wrote:
Betting firms and local media are backing Cardinal Joseph Ratzinger of Germany, who gave John Paul’s funeral oration, but history advises punters to beware of favored candidates.
At lunchtime Tuesday after three ballots to pick the next pontiff, Ratzinger remained favorite on two out of the three online betting boards, with Intertops giving him the shortest odds at 5/2.
Ratzinger was a 11/2 second favorite for Dublin, Ireland-based Paddy Power and 9/2 favorite at Britain’s William Hill. Nigera’s Francis Arinze led with Paddy Power on 7/2.
French cardinal Jean-Marie Lustiger was third for Paddy Power (6/1) and also for William Hill (11/2) who put Arize second at 5/1.
Making up the top six in all three lists were 78-year-old Italian Carlo Maria Martini; his successor as the archbishop of Milan, Dionigi Tettamanzi and Claudio Hummes of Brazil.
Experts, though, seemed less solidly behind Cardinal Ratzinger. On the one hand, Laurie Goodstein and Ian Fisher of The Times called him the “central figure†entering the conclave, with more initial support than anyone else. On the other hand, the writer Michael Brendan Dougherty said on Twitter on Tuesday morning, “Although he seemed obvious after the fact, very few Vatican watchers gave Ratzinger a shot.â€
In 1978, neither experts nor oddsmakers foresaw the outcome. From the Aug. 11, 1978, edition of The Catholic Herald:
Ladbrokes opened the book on the election of the new Pope naming Cardinal Pignedoli favourite at 5-2. Cardinals Baggio and Poletti were joint second favourites at 7-2 with Cardinal Benelli fourth at 4-1. Also strongly fancied were Cardinal Willebrands at 8-1 and Cardinal Koenig at 16-1. Cardinal Hume was quoted at 25-1. The longest odds quoted were 33-1 [for] Cardinal Suenens.
In fact, Cardinal Wojtyła, archbishop of Krakow, won on the eighth ballot.
This time, the experts and oddsmakers seem to agree that Cardinal Scola is the favorite, but they don’t agree on all of the second-tier favorites. A survey of Vatican watchers by YouTrend listed Cardinal Timothy Dolan of the United States as the second most likely ope, followed by Cardinals Ouellet, Scherer and O’Malley. Cardinal Luis Tagle, of the Philippines, was sixth. A couple of the oddsmakers’ favorites, including Cardinals Turkson and Bertone, do not appear on the YouTrend list.
To be sure, experts aren’t always less wise than crowds. A few Supreme Court experts did a better job forecasting last summer’s health care decision than prediction markets did, and their success did not appear to be merely a matter of chance. Those experts managed to read some of the signs around the court in the weeks leading up the decision. The odds on the prediction market, by contrast, did not react to the signs at all.
Similarly, during the 2012 election, poll analysts arguably did a better job t! han prediction markets or oddsmakers.
A conclave is precisely the sort of event that can play to experts’ strengths. As with the health care decision, experts have the potential to collect useful information that isn’t available to the public. Of course, having the potential to do something and actually doing it are two different things.
