Chevy Volt Production To Halt For 5 Weeks Due...
What's Wrong With Chevy Volt Ads & How To Fix...
Chevrolet To Offer Volt Modifications Against...
Analyzing electric-car sales numbers and prognosticating on them is somewhere between a parlor game and a black art.
Now one blogger has applied some Nate Silver-like statistical analysis to try to divine predictive patterns in sales of the Chevrolet Volt, the range-extended electric car that's currently the best-selling plug-in vehicle in the U.S.
Ryan Turner, of Raleigh, North Carolina, who writes the My Chevy Volt blog (he's also @VoltDriver on Twitter), suggested yesterday in a post that looking at Volt inventory in California gives some clues to future Volt sales.
He notes that he had not predicted the fall in Volt sales that took place last November, though he did expect reduced numbers last month as buyers who wanted to take advantage of the $7,500 Federal income-tax credit would rush to buy before December 31 of last year.
But now he thinks he's found the way to predict the monthly numbers: Reported Volt sales track inventory in the car's key market, California, where more than 50 percent of Volts are reportedly sold.
And he has a very nice graph to map the two trend lines, which we've taken the liberty of reproducing here (but, you should go read his whole analysis).
Specifically, he looks at the number of Volts available for sale within a 30-mile radius of the 90210 ZIP code. That's Beverly Hills, in case you never watched the famous Nineties TV show on Fox.
He's tracked those inventory numbers since last May, so he's got most of a year's worth of data built up now.
And the graph is pretty revealing.
![Chevrolet Volt sales vs. Volt inventory around ZIP 90210, May 2012-Jan 2013 [VoltOwner.Blogspot.com] Chevrolet Volt sales vs. Volt inventory around ZIP 90210, May 2012-Jan 2013 [VoltOwner.Blogspot.com]](http://images.thecarconnection.com/sml/chevrolet-volt-sales-vs-volt-inventory-around-zip-90210-may-2012-jan-2013-voltowner-blogspot-com_100417825_s.jpg)
Chevrolet Volt sales vs. Volt inventory around ZIP 90210, May 2012-Jan 2013 [VoltOwner.Blogspot.com]
Enlarge PhotoWe'll be testing out his premise in months to come, to see if it holds up.
Meanwhile, we're curious: What other metrics would you use to predict the sales of the Chevy Volt--or indeed any plug-in car?
Leave us your thoughts in the Comments below.
+++++++++++
Follow GreenCarReports on Facebook and Twitter.
Have an opinion?
That is, to predict sales nationally in X month, you have to have California sales for X month, so why is that better than using National sales for the month earlier or just getting the National sales at the same time as the California sales?
Further, since as you point out, California sales are generally 50% of the national total anyway, then California sales numbers are MOST of what is contained in the national total. Or, the data/approach is using "X to predict X."
It isn't California sales. It is California inventory levels. California inventory levels are know real time.
And the point of the blog wasn't really to make a model of sales. It 'can' do that in a reversed engineering base, but the intentions of the blog are to show that the lack of sufficient California inventory directly affects sales.
I do say, in the blog, that an increase in California inventory will likely mean an increase in sales. But there will come a point where inventory is matched to demand, and this will cease to provide any real use.
The point I am making is that the Volt market is artificially restrained.
But don't confuse inventory with sales :) Two different things.
I just don't understand why GM didn't figure it out. If they give enough inventory to LA and SF area, they can easily increase their California sales by at least 20-30%...
The Chevy ad about going for weeks/months without a fillup was right on
I think it is very apparent from my data that lack of inventory in California pulled down sales significantly. You don't even need to believe me. GM has said it repeatedly, but they've never really backed it up with real data. It left them vulnerable from critics that said that reason was incorrect.
Do you think that the Volt suddenly got unpopular in November, then people decide to like it again in December, then hate it again in January?
There is always going to be variability for a multitude of reasons. But the supply disruptions for almost 2 full months in the second hand of the year was absolutely THE primary reason for poor sales.
Read by blog, not this post.
If they had shipped more AT-PZEV Volts for California, they would have sold more each month...
Nationally, there is little correlation.
Cumulative Volts built less Volts sold, Volts sold that month and PCT of avlble Volts sold that month:
Mon Unsold Sold Pct of Avail
12Jan 7129 603 8%
12Feb 8453 1,023 12%
12Mar 7350 2,289 31%
12Apr 7450 1,462 20%
12May 9831 1680 17%
12Jun 9212 1760 19%
12Jul 10757 1849 17%
12Aug 11614 2831 24%
12Sep 9977 2851 29%
12Oct 8726 2961 34%
12Nov 9512 1519 16%
12Dec 8388 2633 31%
13Jan 10197 1140 11%
13Feb 11869 1626 14%
Volts sold well Aug-Sep during declining inventory, selling up to 24% of available and have tanked despite increasing inventory, especially in the most recent two months.
Green Cars have a dirty little secret. Informative, but has nothing to do with the Volt's inventory, GM's channel stuffing.
Have an opinion?Join the conversation!