San Francisco: By day, Robert Leitao manages a Catholic church in Southern California. By night, he indulges his other passion: predicting Apple Inc's results. 

Leitao is part of a cadre of amateur forecasters, bloggers and hobbyists who sift through reams of data every quarter to guess at Apple's quarterly results - often putting professional analysts to shame by coming up with more accurate predictions. 

Co-founder of the Apple Independent Analysts Group, Leitao is ranked seventh for the December quarter out of 50 analysts who cover Apple by Fortune magazine, which found that his estimates turned out to be much closer to the results than those from prestigious banks such as Goldman Sachs and Morgan Stanley. 

While Wall Street analysts' forecasts for Apple's revenue and earnings per share were off by an average of 21 percent in the latest quarter, amateur analysts missed by just 10 percent, according to the Fortune data.  

This raises questions about how good Wall Street is at forecasting Apple, the largest U.S. company by market value, famous for trouncing market forecasts quarter after quarter. 

Leitao does not believe he and the more than 100 members of the Apple Independent Analysts Group - which he started as a hobby - are smarter than their professional peers. He suspects Wall Street is more inclined to play it safe. 

It does not help that Apple itself tends to lowball its guidance. "In their work, there is a greater risk in coming out with aggressive estimates that are too high," Leitao said of analysts at top-tier brokerage firms. 

In the last couple of years, Apple's earnings have exceeded Wall Street expectations tracked by Thomson Reuters I/B/E/S by at least 13 percent, and often a lot more. The exception was the September quarter of 2011, when they fell short. 

Technology analysts say Apple is particularly tough to predict because of its secretive nature, and because it has a hand in everything from hardware (iPhones, iPads, Macs and iPods) to digital distribution (iTunes and its emerging iCloud remote storage service). 

They also point to a lack of historical data on the tablet market, and Apple's fast expansion into more than 100 countries as challenges. Yet another blind spot is sales from Apple's own retail stores and - data that only the company has. 

"If you're talking about Apple where you're trying to model Macs and phones and new markets like tablets, that's obviously more challenging than making assumptions around Intel where you're only modeling PCs," said JMP analyst Alex Gauna.