Your Graph is Bad

By now you've noticed the Facebook phenomenon of posting seemingly convincing graphs that underscore the remarkable victories -- or crimes against humanity -- of a certain politician, political party, or socioeconomic class. Invariably, the graph confirms the ideology of the poster, heated discussion ensues, and in about 4 minutes, Godwin's Law is enacted.

To all those earnest graph posters and re-posters out there: your graph is bad.

Most of these graphs floating around on Facebook are pure abominations of critical reasoning. Although they use real data, there are literally dozens of tricks and slight-of-hand deceptions -- or, if we want to be polite, graphical nuances -- that distort the data to support a predetermined viewpoint. 

This is what political scientists, in their fun-loving and whacky way, call propaganda.  While it might be true that a picture is worth 1,000 words, chances are that most of those words are spin.

To illustrate the point, let's take a look at one popular graph making the rounds after gasoline hit approximately $38/gallon in San Francisco and Republican presidential candidates stumbled all over each other promising to “end our dependence on foreign oil.”

In an effort to stave off potential criticism of President Obama’s energy policy, which has included some politically dicey decisions such as off-shore drilling moratoriums (since lifted) and blocking the construction of the Keystone XL pipeline, the following graph has been circulating (I don't mean to pick on the President -- Republicans pull this kind of crap too):

The obvious message here is that things “got worse” under President George W. Bush and have “gotten better” during President Obama.

Claptrap, I say.

Let’s take a look at some of the problems with this graph:

1)     Distorting the Time Frame:  The graph incorrectly identifies when President Obama came into office.  If you look closely, you will see that it identifies 2008 as the start of the Obama presidency, instead of the actual 2009.  But because it uses that fancy little arrow, your eye will immediately follow it to over the 2009 data point, and you probably won't notice the distortion. 

Just changing the graph to accurately reflect when President Obama came into office and adding a trend line from the dependency high paints a different picture.  It now becomes quite clear that dependency on foreign oil was already on its way down well before the time President Obama assumed office:

As it turns out, our reliance on foreign oil as a percentage of net imports has been declining since 2005.  It’s hard to credit the Obama administration for something that started right around the time President Obama was finishing up cleaning out his office in the Illinois State Senate and learning where the bathrooms are in the Dirksen Senate Office Building.

2)     Ignoring the Real Causes

This graph tries to get away with the logical fallacy of post hoc ergo propter hoc, which is what Latin-speaking guys wearing togas would say when you tried to take credit for something you had nothing to do with.  In today’s more genteel times, we call this “correlation without causation.” 

The graph implies that something about the Obama presidency has resulted in lower dependence on foreign oil.  Since it doesn’t say what policies, exactly, the graph not only appears convincing, but also is hard to refute.  After all, there are no pesky data points that contradict the prevailing message of the graph.

So let’s pick it apart a bit.

Even though the height of dependence was in 2005, the significant, obvious downward trend begins in about 2007-2008. There are literally dozens of factors influencing our net imports for petroleum products, but nothing matters more than consumption.  To use a graph helpfully provided by the US Energy Information Administration (EIA):

One can see that while there is some correlation between Production and Net Imports, the lines for Consumption and Net Imports are almost identical.  As our petroleum consumption has increased, so have our net imports. 

Petroleum consumption itself is a reflection of broader economic conditions – in a bad economy we consume less, and in a good economy we consume more.  Because of the economics of the global energy trade, the more we use, the more we import. 

As the EIA concludes in its report, “"This decline partly reflects the downturn in the underlying economy after the financial crisis of 2008. Not surprisingly, demand has bounced back somewhat from a low of 18.8 million barrels per day in 2009, when the U.S. economy bottomed out. But the downward trend in consumption started two years before the 2008 crisis and reflects factors such as changes in efficiency and consumer behavior as well as patterns of economic growth."

The corollary to this analysis is that when the economy recovers, net imports of foreign petroleum will increase.  Unless we all go out and buy a Prius.

Conclusion

And there you have it.  In two easy, but deceitful, steps, supporters of President Obama have made what looked like a convincing graph to demonstrate the Administration’s triumph of lowering our dependence on foreign oil.  Further investigation and application of that ol' gray matter between the ears demonstrates how silly a claim it really is.

Have a graph that you think is junk?  Email it to TheBadGraph@gmail.com.