What idiot puts up a horizontal array? It costs essentially nothing to angle them to the proper angle for any given latitude.
Theoretically there is an advantage to tilting the array, but from a practical standpoint, there is little to none. There are several factors that prevent achieving much of that theoretical improvement.
First, since 98% of solar installations are on sloped roofs, you'd first need one slope of your roof facing due south, and not many homes have that. The additional losses incurred from a slope facing even a few degrees off from due south is great enough that the flat plate example given above turns out to be a good approximation of real-world conditions.
Second, for cost and structural reasons, most installations are affixed directly to the roof. At the average U.S. latitude of 37º, the ideal roof pitch would have to be 9" per foot (that's how it's measured in the construction trades). However the most common roof pitch is only 4-5" per foot.
Third, due to the ~23º inclination of the earth to its orbital plane, the "ideal" tilt angle for the solar array would have to vary 46º per year, as the seasons changed.
Fourth, only about 50% of solar radiation incident upon the earth's surface is from direct sunlight. The remainder is indirect or diffuse radiation. When you tilt the array towards the direct rays to increase performance, at the same time you are decreasing performance for solar radiation arriving from diffuse sources.
Net result? The numbers given in the table in post #7 turn out to offer a very real approximation to what one can expect from a typical installation. If you doubt it, I'd be glad to post a published study by Shell Solar describing an installation they did in sunny southern California, and the resulting energy production performance.