One likely explanation for the incredibly optimistic 2014 earnings estimates, then, is that each company's analysts are merely providing justifications for their own buy ratings.
"As an analyst, you want to print a certain number, and you'll do whatever it takes to get to that number to get a buy recommendation," said ConvergEx Group chief market strategist Nicholas Colas, drawing on his own experience as an equity analyst covering the auto sector. "Maybe it takes a prediction of 10 to 12 times earnings growth to get to that number. Well, then that's what you're going to do."
The issue is that analysts feel a great deal of pressure to recommend stocks.
"It isn't just that analysts want to be optimistic. They have to have things to market. And maybe a good deal of your client base has to be long stocks," Colas said. "Well, at the end of the day, you have to justify those recommendations with some numbers."
Yet while long-range estimates can soak up optimism, analysts will keep nearer-term estimates closer to earth, because "analysts don't want to be proven too wrong on the current quarter," Colas said.
(Read more: Winter weather freezing corporate profit growth)
Michael Pachter, a widely followed analyst of video game, digital media and electronics stocks with Wedbush Securities, says the blame really lies with corporations, not research analysts.
"Companies low-ball at the beginning of the year, and back-end-load their earnings profile," Pachter said. "Companies are increasingly telling analysts that earnings are a hockey stick, and analysts do what the companies tell them."
Regardless of who is to blame for the odd earnings projections that have been cropping up, investors have probably gotten used to them by now.
Still, all the anticipated growth—which has been registered in expansion in price-to-earnings ratios more than in actual earnings—can't be pushed off forever.
"It happens every year in exactly the same way, and capital market professionals understand that, so you can have a market that goes up as estimates go down," Colas said. "But at some point, that game has to end. Because you can't have P/E infinity."