[Editor's note. The Tyee today revives its acclaimed Election Central superblog. Many time a day you'll find fresh scoops and analysis on the federal election which, let's face it, is just now revving up in earnest. To visit Election Central, click here.] On Election Day 2004, Victoria's Times Colonist ran a story on A1 predicting the campaign had come down to a two-way race between the Liberals and Conservatives in the capital city. The Liberals were running David Anderson. The Conservative candidate was Logan Wenham. Michael Prince, a political scientist at the University of Victoria, told the reporter who wrote the story that the election would hinge on how much support David Turner, the NDP candidate, drew. If Turner gets 25 percent of the vote "Anderson is dead in the water because it will mean Logan Wenham will win quite solidly," Prince said in the article. "In a sense, it's not a three-way race in Victoria, it's a two-way race. Two of them are neck-and-neck and who wins depends on how well Turner does." Prince, as it turned out, was completely wrong. Anderson squeaked back to Ottawa, beating Turner by less than four percent of the vote. The Conservatives' Wenham was way back in third. The fact is, much as journalists love to quote them, pundits aren't much better at predictions than the average person. In fact, according to a new book by Berkeley psychologist Phillip Tetlock, sometimes they're actually worse. In Expert Political Judgment: How good is it? How do we know? (reviewed by The New Yorker's Louis Menand here) Tetlock tracked 284 professional pundits and opinion makers for 20 years, asking them to predict the probability that a variety of events would, or would not, happen. For most of their more than 80,000 predictions, the experts were asked to pick a probability of one of three things: A continuation of the status quo; more of something (economic growth, parliamentary seats for a party, etcetera); or less of something. The results, in Menand's words, were not promising: The experts performed worse than they would have if they had simply assigned an equal probability to all three outcomes-if they had given each possible future a thirty-three-percent chance of occurring. Human beings who spend their lives studying the state of the world, in other words, are poorer forecasters than dart-throwing monkeys, who would have distributed their picks evenly over the three choices. What's more, the experts proved no better at predicting what was going to happen in their fields of expertise than they did in areas where they had, ostensibly, no more knowledge than the rest of us. In fact, in most cases, the more specialized their knowledge, the worse their predictive abilities. Why? Well, to oversimplify, when you have a lot of special knowledge, you want to use it. And when it comes to probabilities, it's often the simple answer that's right. With the federal election less than three weeks away and BC already labeled a battleground; it's a lesson worth remembering. It may sound over-simple, but when it comes down to it, whether or not a seat changes parties depends on how many people change their votes. And probabilities tell us that the fewer voters that have to switch, the more likely the seat will change. So that's it. The Tyee's expert formula for determining seats and regions to watch in BC comes down to this: They were close last time. From Vancouver, to the suburbs, the Island and the Interior, the races we'll be watching and reporting on were all painfully close in 2004.That's not to say upsets and big swings can't and won't happen. And we'll report and blog on stories from across the province (and occasionally the country), swing seats or not. But when it comes down to the seats that will most likely change hands, these are the ones we'll be watching: The Interior Last election, the Interior was solidly Conservative. The only chinks in the blue armor were Skeena-Bulkley Valley, where New Democrat Nathan Cullen beat Conservative Andy Burton by less than 1,000 votes, and the Southern Interior, where Conservative Jim Gouk edged his NDP opponent by a similar margin. Both seats should be contested again. Gouk retired, leaving his former assistant Derek Zeisman to defend the seat. Zeisman made headlines early in the campaign when a reporter with the Trail Daily Times uncovered an old essay wherein the candidate called for the elimination of the Charter of Rights and Freedoms and a North American monetary union. However, all was forgotten when Zeisman suffered a serious car crash just before Christmas. Zeisman plans to continue campaigning from his hospital bed. The Suburbs Surrey: Vancouver's suburbs are the largest battleground in terms of tight races. In Surrey, three of the four seats were decided by less than 4,000 votes in 2004. In the fourth, Surrey North, the now-deceased Chuck Cadman won in a landslide. But former Provincial NDP cabinet Minister Penny Priddy is competing in the riding this time and was a good friend of the popular Cadman, which could make this tighter than expected. The tightest race in Surrey was between Conservative Gurmant Grewal, Liberal Sukh Dhaliwal and New Democrat Nancy Clegg in Newton - North Delta. Grewal bowed out, but Clegg and Dhaliwal are both running again, this time against fisheries activist Phil Eidsvik. Grewal's wife Nina is still running in Fleetwood - Port Kells and will try to slay her second former provincial Liberal in as many contests. Burnaby, too, produced tight races in 2004. New Democrat Bill Siksay beat Liberal star candidate Bill Cunningham by less than a thousand votes in Burnaby - Douglas and the Conservatives weren't far behind. Burnaby - New Westminster was an even tighter NDP win. In Richmond, Immigration Minister Raymond Chan won handily in 2004. Watch Election Central in the coming weeks for a story on why this riding might swing anyway. Both seats on the North Shore were tight in the last election. West Vancouver-Sunshine Coast, a bizarre conglomeration of fishing towns, resorts and some of Canada's wealthiest voters, elected the now retired John Reynolds by less than 2000 votes. Liberal Don Bell won by a similar margin in North Vancouver. The last seat to watch in the burbs is New Westminster Coquitlam. Conservative Paul Forseth won here in 2004 with less than a third of the vote, only 113 votes ahead of the NDP. The Liberals are running former provincial Liberal Cabinet Minister Joyce Murray in the riding, while the NDP have former New West - Burnaby MP Dawn Black. The Island There were only three relatively comfortable wins on the island in 2004. Among the others: The Liberals hold the two (mostly) urban ridings of Victoria and Esquimalt - Juan de Fuca. The NDP pushed incumbents David Anderson and Keith Martin hard in both ridings 19 months ago. And with Anderson retired, things don't promise to be easier this time. And Vancouver Island North was one of the closest races in the country, with Conservative John Duncan winning by just 483 votes. Vancouver And last but not least, the big city itself, Vancouver. Hedy Fry has hung on to Vancouver Centre for more than a decade now, but not since she dispatched Kim Campbell in 1993 has she faced someone with the name recognition of Svend Robinson. Meanwhile, in Vancouver Kingsway, NDP veteran Ian Waddell is back for another go at Industry Minister David Emerson. Emerson simmered briefly in what some have already dubbed the silliest controversy of the campaign after Jaime Elmhirst, the President of the Federal Liberals in BC, reported on his blog that the incumbent said Jack Layton had a "boiled dogs head smile," which may or may not be an insult in Cantonese. So that's it. Those are the tightest ridings, at least according to The Tyee's patented statistical analytical process known as what happened 19 months ago. Which ones will swing and which ones will stay will come down to how many people changed their minds. Hopefully, Election Central will help you figure out why. Richard Warnica edits The Tyee's Election Central superblog. To visit, click here.