The moderate consumption red wine seems to be a plausible component of the low number of realized CHD deaths?
The "French paradox" isn't the only one indicating the statistics are horse shit on saturated fat: consider the Israeli paradox.
In at least one study (can dig it out if wanted) the result was that it looked as if moderate drinking was associated with better cardiovascular outcomes than no drinking at all (as well as drinking too much) because the people who didn't drink at all had stopped drinking on doctors' orders, essentially.
More recent studies have excluded participants with underlying heart conditions and the effect disappeared. Again, I can dig out the links if required (but it'll take me a bit of work).