Each quarter, Fortune’s Philip Elmer-DeWitt publishes a table of unaffiliated and Wall Street analysts' predictions for Apple’s (AAPL) earnings release. Invariably the Wall Street crowd underestimates and the unaffiliated overestimate but end up much closer to the actual EPS. This trend is consistent, quarter after quarter. I calculate my own bottom/up estimate, but with so many other analysts out there (Dewitt surveyed 44 for Q1 2011), I didn’t think it would add much value to throw mine in.
Instead, I looked to see if there was a “collective intelligence” or pattern in all these estimates. I took all of DeWitt’s quarterly postings, plotted histograms of the EPS estimates, and saw a pattern emerge. With the exception of Q2 2010 (which was a total blow-out), actual EPS fell about 75% of the way between the largest groupings of Wall Street and unaffiliated analysts’ estimates.
Q2 was a blow-out for two reasons: 1) This was the first quarter where Apple dramatically lowered its tax rate, and 2) iPhone sales blew out the highest estimate by 1M units.
Using the emerging pattern, which becomes clearer as DeWitt expanded his survey (more data = better-looking graphs), the “composite” EPS estimate for Q1 2011 is $6.05. This point is about ¾ of the way between the Wall Street and unaffiliated analysts’ peaks. I don’t know if this pattern will continue to hold, or if it is just a fluke, but if nothing else I thought it was interesting enough to share.