The iPad/Netbook Graphs are Wrong: Buy Morgan Stanley a Calendar and Everyone Else a Calculator

image Let me preface this post by saying that I don’t obsessively follow the punditry of Philip Elmer-Dewitt. He could be, in general, a solid pundit and a great asset for CNN/Fortune’s tech division.

After reading his “hard hitting” analysis of some Morgan-Stanley Numbers regarding netbook sales, I have more questions than answers, and chief amongst the questions is “why are so many tech pundits so bad at math?”

Philip says, in his post:

What caught my eye, however, was what her proprietary research shows about the impact of the iPad and other tablets on the broader gadget market, starting with netbooks. As her chart (above) shows, sales growth of these low-cost, low-powered computing devices peaked last summer at an astonishing 641% year-over-year growth rate. It fell off a cliff in January and shrank again in April — collateral damage, according to Huberty, from the January introduction and April launch of the iPad.

There are three problems I spotted with his write-up: first of all, the descriptive language he uses seems to indicate that sales are shrinking.  The important but subtle distinction to make here is that marketshare and sales for the netbook class devices continues to grow to this day, just at a lesser rate than it did in previous months.

Beyond Philip’s writeup, there is critical fault with Morgan Stanley’s research numbers.  It’s difficult to tell if this drop off is partially due to seasonal effects since the data shown only goes back ten months. What is clear is that the netbook is still a new class of device, and the rate at which it grew in popularity simply isn’t sustainable for much longer than what it ran for – you can’t have 641% year-over-year growth rate forever when you own 20% of the market, as netbooks do.

If Philip was a bit better at reading graphs or basic math, he might have caught this point, and called attention to it.

Of course, the biggest problem with this write-up: despite the fact that Philip acknowledges Morgan Stanley’s assumption is flawed because they can’t remember when the iPad was actually released, he still publishes the post as authoritative and correct:

“Her timing seems a little off. The NPD data she cites is from January, but Steve Jobs didn’t unveil the iPad until the end of January.”

image What was even more disturbing that he let through without comment was the graph showing that “44% of U.S. consumers who were planning to buy an iPad said that they were buying it instead of a netbook or notebook computer.”

Really? What was the sample size? What was the polling method? What cities were used? The type of devices used by mainstream varies widely from region to region, in my experience.  None of this data is questioned or explained, and passed off as fact.

This is the same thing I went over the last time the blogosphere fawned over the death of netbooks which was, if my calendar is correct, oh yeah, last week!

Even though the post was the most popular article on the site for several days running, not a single post following on to the coverage today can be bothered to remember it:

The other problem with this analysis is in CNet’s actual reporting – there is no “dramatic slowdown” here.  If anything, it’s better described as slowed growth.

Electronista had the numbers everyone’s referring to:

Interest in netbooks has slowed dramatically in the first few months of 2010, according to new data from IDC. The number of the mini netbooks shipping grew less than 20 percent in the first quarter of the year and was accompanied by a steep drop in interest for the Atom processors that power them. Intel’s shipments of Atoms dropped from just over 24 percent of all notebook processors to 20 percent; the revenue from the chips dropped 19 percent at the same time.

If we look back to data discussed by Paul Thurott on WindowsIT Pro back in September of 2009, mini netbook sales:

According to a report by DisplaySearch, sales growth of netbooks outpaced sales growth of traditional notebook computers by almost 200 percent. PC makers shipped about 38 million portable computers in Q3 2009, according to the study, with overall sales growing 22 percent. But sales of netbooks grew a whopping 40 percent.

The rate of growth has indeed, if the numbers are correct and compatible, slowed by half, but then so have sales of the notebooks as a whole. Indeed, in 2009, netbooks were 22% of all notebook sales. In 2010, they’re 20%.

The iPad isn’t killing netbooks – or at least not yet.

The numbers haven’t changed since last week – we’re still talking about the same basic data. The growth numbers continue to look like they’re on a volatile decline, but the marketshare numbers remain relatively constant – between 19-24%, depending on which source you look at.

Even more mind-boggling, as of the time of this writing, the wall of shame continues to grow as other tech pundits take a look at this post and re-write everything as fact rather than doing a basic Google search or whipping out a calculator to find out if the data is in any way realistic. Computerworld, Yankee Group Blog, VentureBeat, MacRumors, AppleInsider,ReadWriteWeb, 9 to 5 Mac, TheAppleBlog, ZDNet, Erictric, The Next Web, Mashable!, GottaBeMobile,Silicon Alley Insider, Liliputing, Electronista, and MacDailyNews all posted rewrites with no new information and no critical analysis.

Take from that what you will, but chiefly, I ask that you take from it that the iPad, with it’s one million sales, did not dent the estimated 3-5 million netbooks sold over the same time period.

Update: This comment came in over Google Buzz from reader Nicholas Chase: “Add Gizmodo to the wall of shame – I guess they aren’t serious tech blogs like the others, but their ludicrous coverage got me steamed. The title says it all:” Done, Nicholas.

Update II: As is noted in the comments, Ian’s post at Technovia shouldn’t have been included in the “wall of shame” list, as his was a tertiarily related post to the meme. It’s inclusion in the list was an error on my part. The list has grown some since the initial post, and now includes The Register, Yahoo! News, and PadGadget.