The 58 percent solution
Since March , Rupert Murdoch's Wall Street Journal has been devoting a full page to sports coverage. They don't cover sports the way your local newspaper does, in print or on the Web, with game stories and results. Instead, their writers produce analytical stories that you often don't see elsewhere. For instance, they recently explained why a valid world record cannot be set at a Boston Marathon: because the course runs downhill.
Instead of sports results, the Journal publishes a daily dose of "game predictions" generated by a Los Angeles company called AccuScore. "Based on an average of 10,000 game simulations and rounded to one decimal point," the Journal both predicts "precise" scores and win-loss probabilities, mostly for baseball games.
While my Ouija board was at the repair shop, I clipped a few days' worth of Journals, and compared their predictions with the games' outcomes. Last week AccuScore, which must be AccuWeather's statistically challenged younger sibling, predicted that the Detroit Tigers had a 64 percent chance of beating Daisuke Matsuzaka. They didn't. Likewise, the data drones said that the St. Louis Cardinals would beat the Cincinnati Reds 4.6 to 4.1 on Wednesday. The Cardinals lost, 9-3.
Journal bigs aren't rushing to defend this silliness. When I first reached Dow Jones vice president of communications Robert Christie, he said: "What's AccuScore?" Later he offered a tepid endorsement of his newspaper's glorified coin toss: "AccuScore predictions fall in line with what we expected and provide our readers with a high rate of accurate, forward-looking information." Christie also laid some hate on the Red Sox, which I am too gentlemanly to reproduce here.
"Accurate, forward-looking information," eh? Matt Drudge was pilloried years ago for claiming that his website was 80 percent accurate, but it turns out that 58 percent - AccuScore's overall claim - is good enough for the Journal. Remember that the next time one of its columnists offers you a stock tip.
Yes, yes, I know this is a betting tool that dares not speak its name, because sports betting is illegal in most states, including this one. Why else would AccuScore claim that its "advanced simulations generate huge profits"? "We don't recommend it as a betting tool," says a straight-faced Zach Rosenfield, of AccuScore. "We aren't in the gambling business. We offer pregame opinions like Dan Shaughnessy or Jackie MacMullan do."
What about the hit-or-miss accuracy record? "People have an expectation of 100 percent accuracy, or 90, 80 percent," Rosenfield says. "Nobody hits at that clip."
"You got me," admitted a too-pliable McEnany when I reached him at his home in Franconia, N.H. I had assumed that McEnany - who got a nice review in the Wall Street Journal, by the way - was himself a logger. "No," he said. "I'm a pencil monkey, just like you."
Fair-minded person that I am, I distributed the cookies to my discerning carpool of 11th graders - and they loved them! "People see the box and they think the cookies aren't going to taste good," says clinical psychologist Jill Robbins, who started the business in the basement of her home. "But they do."
HomeFree now has a 9,000-square-foot production facility that outsiders are allowed to visit only if they agree to wear a clean, zip-up jumpsuit and washable shoes that Robbins will provide. "We cover them up just to make sure," she says. "You never know what they might be tracking in from their car seats." At $5 a box, they are a tad pricey, and now available at Shaw's and Whole Foods supermarkets.
Alex Beam is a Globe columnist. His e-dress is beam@globe.com. 