However, as it turns out, 95% or more of the world believe that materialism is wrong and that some sort of theism is right. They believe that there is another dimension to the world, although there is of course wide disagreement about the details. About 2/3 of the world is monotheistic, and about the remaining third is pantheistic or panentheistic. Of the residual 5%, we can include such beliefs as outright paganism, polytheism, and others, all of which stand as a contradiction to atheism. In short, atheism, taken globally and historically, is profoundly weird, amounting to less than 5% of the entire world's population at the moment (let alone all of history). So if anything, we should be skeptical of atheism as our default position.
I don't think it's particularily valid to lump all forms of theism together as if they actually make up some block of common belief. In reality, many forms of theism are just as much in opposition to other forms of theism as they are to atheism.
What you're doing here is like stating that political party 'X' is likely incorrect because all the other parties all agree that party 'X' is wrong... even if that group of parties runs the gamut from Marxist-Leninists to Libertarians.
Notice, this is NOT an argument of X is popular and therefore true. Nor is it an argument that you should believe X because it's popular. It's an argument about where default skepticism should lie. If "everybody knows" something, the contrarians hold a preponderance of burden to show that everybody is wrong. In our case, everybody knows that atheism is wrong. So the atheist bears a preponderance of the burden to show that the reverse is the case.
The more statistically independent tests you have (e.g. rolls of a die, or participants in a survey), the more likely that the average of your sample will approach the actual average... so, the more people who say "I'm voting for party 'A'" when you ask them, the more likely it is that party 'A' will win the election.
However, this principle only works when the tests are statistically independent. When it comes to matters of faith, there are strong mechanisms that will encourage a person to have the same faith as his or her parents and community, regardless of the truth of those beliefs. If a person is a believing member of their church because they were raised in it and taught it from an early age, then this implies that they are not a believer for other reasons, such as the confirmed truth of their beliefs.
Religion has a tendency to spread regardless of whether or not it's true. Turning your atheism/theism dichotomy around for a minute, you can see this by looking at the beliefs of the world: whatever your pick as the "true" religion, most religious adherents believe in something else. The world is awash in false belief. Given this fact, why should anyone assume that one particular religion that spreads like the others and has no more apparent support than the ones that we recognize are mostly false, why should we assume that one to be true?
Basically, your argument against atheism can be applied to any belief system that's made up of a minority of the world's population... i.e. all of them.
Perhaps a bit more substantially, these facts might indicate that the atheist is blind to what is obvious to everyone else. Namely, that the world is comprised of more than just physical stuff (indeed, some -- a larger proportion of the world than that claimed by atheists -- argue that the universe of matter is actually illusory!). There's something deeply spiritual about it. It's either spirit itself or is incompletely described by appeal to physics and chemistry and their derivatives, as useful as those descriptions have proved to be.
I acknowledge what I consider to be "deep spirituality" in the universe, but I do not consider it to be any sort of god. I don't think spirituality requires belief in God at all.