Oil is not half of America's imports. It could be viewed as a percentage of America's trade deficit, but then, so could other categories of imports like agricultural goods or construction materials or electronic equipment or textiles or cars or even virtual imports like tourism.
By volume I have no idea, although I suspect it is number 1.   Its very close to half (at $60/bbl) of the trade deficit dollar value each month, = 25 billion vs total trade deficits  in range of 50-64 billion as of late.
Yeah, but the point is that the trade deficit (or surplus) is exports minus imports. It makes no sense to think of one imported commodity as a fraction of the trade deficit. Oil is not half of US imports.
Maybe somebody else said that.  I didn't say it was half the US imports.  I said its "almost half the trade deficit dollar value".  If I want to know how the trade deficit value in USD  originates and what % of the trade deficit can be attributed to oil imports - oil exports and I know the oil volume imported and the volume exported (assumed = relative 0) and the average cost of the imported oil, it makes sense to me.  I know a certain amount of oil import value is offset by wheat exports, bond sales to foreign entities, and HP computers etc.,  but I don't care about that.  I only want to know the relative impact of oil imports-exports towards calculating the value of the deficit, so I think it makes sense to think about it that way.  At least I think it makes "engineering sense".
Even relative to total imports it takes the pretty substantial 20% (rough estimate - oil imports: 13mln.bpd x 365 x $65 = $308 bln.; total imports as of 2005 ~ $1500 bln.)

If we could eliminate oil imports we'd cut our trade deficit by half.