QUOTE(La Herring Rouge @ Sep 27 2006, 08:01 PM)

You seem to misunderstand either the graph or my reason for posting it. The graph (showing gas prices as an inverse) shows how closely Bush's ratings match the price of gas at ANY given time interval with near perfection. If Bush is popular, you can bet that gas prices are low and if they are high, well guess what?
This is not voodoo, it is not conspiracy theory and it is not an accident. The correlation is not proof of some price fixing conspiracy. It is, however, clear proof that gas prices seem to loom large in the minds of Americans when they judge their leader.
What that graph demonstrates is the reason why there is a saying:
"There are three kinds of lies: lies, damn lies, and statistics."
If you look at OPEC and its membership, I think anyone can see a problem with the theory that Bush is manipulating prices. Why would OPEC members Iran, Venezuela, Nigeria or Russia help prop up the poll numbers of GWB?
You claim its the oil companies (all of them, apparently, moving in unison). But oil company profits amount to only a few cents on the dollar. They could maybe shave off 20 cents a gallon if they made no profit. But we know that they have made profits, despite (apparently) the desire to give Bush an edge in the approval polls.
You apparently discount the possibility that everyday Joes measure the economy by the prices at the pump. If the prices drop, the economy must be doing better, thus they are doing better, thus Bush is doing better. I have always found it curious that while the economy is very robust (just short of booming), people think Bush's handling of the economy is poor.
The easiest way to manipulate statistics is to change the scale, which seems to be what that graph you illustrated does.
Besides, the correlation is not scary. At the beginning of the graph (2001), we see gas prices rising while Bush's poll numbers also rise followed by gas prices falling with Bush's numbers also falling. Completely inverse to your "scary" correlation.
From April 2002 to it looks like February 2003, prices remained virtually constant (according to the graph) and yet Bush's poll numbers dropped from 130-something to around 100 (whatever that means). Again, that does not follow any correlation.
Then we have the summer of 2003 (looks like August thru November) where (again) gas prices fell and Bush's poll numbers....also fell.
No, what we have here is a broad trend indicating that Bush's poll numbers have dropped since 2001 and gas prices have risen. Nothing resembling a correlation, let alone any indication of causation.
Of course, this particular blogger wouldn't have any motivation to fix the data, now would he?
Maybe this
map from his website gives us an indication of where his loyalties lie.
The guy provides a more detailed graph for the last year:
chart which even further your theory (Note: the blogger himself only claims that gas prices are useful because it is the only measure of the economy that ordinary folks see. He does not claim a Bush-Big Oil collusion).
November 2005 thru February 2006: gas prices rise, Bush's numbers also rise.
The trend from February thru May seems to fit, but then in May, with prices flat (although high) Bush's number rise.
Recently, with a big drop in prices, Bush's numbers only rise moderately.
There is nothing to this graph.