Data Crunchers Turn Their Algorithms on the Papal Conclaveby
Analytics savant Nate Silver and his team at the FiveThirtyEight blog have hit a wall with the upcoming papal election. They will not be sticking their necks out with a prediction for the conclave vote, citing a dearth of conventional polling data to plug into their models.
Sitting out the most-watched election on the planet speaks to the difficulties of building a statistical model to short-list the papabili, or men most likely to be appointed pope. The outcome of the papal conclave vote is far more difficult to forecast than, say, a political election. For starters, you have to account for a two-thirds clinching vote rather than a simple majority. Secondly, historical voting data from the College of Cardinals will get you only so far. For example, 67 of the 115 cardinal electors voting in this conclave (or, 58 percent) are first-timers.
Still, these obstacles aren’t stopping Vatican watchers and oddsmakers from predicting the next pontiff. They contend there is enough available data to crunch from previous conclaves to handicap the 2013 field. After running the numbers, here is what the papal forecasters have to say about the upcoming vote:
The frontrunner factor
As conclave veteran John L. Allen Jr. wrote in National Catholic Reporter last week, you can throw out the old bromide, “He who enters as pope exits as a cardinal.” Allen crunched the data on the past six conclaves dating back to 1939, concluding that there has only been one complete stunner—Cardinal Karol Wojtyla, elected Pope John Paul II in 1978. All other future popes entered the conclave riding a wave of positive buzz. This time round, many oddsmakers, including Oddschecker.com and Paddy Power, are tipping Italian cardinals Angelo Scola and Tarcisio Bertone and Ghana’s Cardinal Peter Turkson as the ones to watch, with Cardinal Sean O’Malley of Boston emerging as a dark-horse candidate.
The fab five
There may be 115 candidates for this conclave, but history shows that after the initial voting rounds, the serious contenders will be winnowed down to “a subset of five or fewer,” according to a 2004 study by J.T. Toman, an econometrics and business statistics professor at the University of Sydney. If this holds true, you can write off 110 cardinals who will be in Rome this month simply to cast a few votes and maybe dine on a dish of bucatini all’amatriciana before boarding a flight back home.
The first baby boomer pope?
The next pope will be between the age of 63 and 73, predicts Adam Brickley, a Washington (D.C.) blogger and Vatican watcher. The younger side of this age range would make the next pontiff the first boomer pope. Brickley, best known for launching in 2007 the blog Draft Sarah Palin for Vice President, developed a unique model for short-listing papabili in a 2008 undergraduate thesis while majoring in political science at the University of Colorado in Colorado Springs. After analyzing the results of the previous nine papal elections dating back to 1903, he concluded it is possible to make “several concrete predictions” about the next pope’s “age, region of origin, and ideology.” Using this methodology, Brickley has whittled down the field of likely popes to 28. Surprisingly, the list includes no Italians, Germans, or Americans.
Bet long on the conclave
According to Brickley, over the last century, the shorter conclaves have averaged between three and four voting rounds. For the relatively brief conclaves that would mean a successor is usually named on either day two or day three. That won’t be the case this year, he predicts. “The best analogy for this year’s papal vote is 1922, the one that elected [Pope] Pius XI. That went 14 ballots,” he says. “The cardinals then were looking for a steady hand as they were deeply divided.” That sounds a lot like the tense atmosphere inside the walls of Vatican City this week, where major divisions have emerged between the New World cardinals, represented primarily by the Americans, and the old guard Europeans. Reading the smoke signals then: Wager long.