For two days last July in Woodstock, Vt., I looked at hundreds of portfolios -- at times they seemed like thousands -- showing houses and housing projects from all over the United States. Much of what I saw was discouraging; a few designs were inspiring.

The occasion was the seventh annual "Builder's Choice" design and planning awards program sponsored by Builder magazine, the official trade journal of the National Association of Home Builders. Builder has the biggest circulation, and is reputedly the most widely read, of all publications in the American building industry.

The magazine invited 12 jurors, a mixture of architects and developers mostly from the East and West coasts, who were assigned to one of four panels. Each three-person panel judged one category of submissions. While my panel reviewed detached "for sale" and "custom" houses, others looked at attached housing, large-scale site developments and commercial projects.

The competition sponsor subdivided the detached for-sale housing category into four subcategories: "small" houses (less than 1,500 square feet), "midsized" houses (1,501 to 2,200 square feet), "bigger" houses (2,201 to 3,000 square feet) and "large" houses (over 3,000 square feet). Likewise, custom houses were either "small" (under 3,000 square feet) or "large."

Despite project size differences, I and my fellow panelists -- a developer from Florida and an architect from New York City -- wasted little time during the first round rejecting about two-thirds of the submissions, and with little dissent.

A large percentage of them were from Sunbelt states, particularly Florida and California. They appeared to have been produced by some interstate conspiracy bent on selling stucco, roofing tiles and assorted exterior embellishments that showed up over and over again, no doubt from being advertised frequently in widely read trade journals.

Such embellishments included bay windows in a variety of geometries, window flower boxes, balconies, Palladian windows and pediments, dormers, cupolas, turrets, skylights, neoclassical columns, wrought-iron railings and window grills, and ornately embossed front doors.

There is nothing inherently wrong with any one of these elements. But assembling them in one composition proved to be a major aesthetic stumbling block for most of the houses submitted. Yielding to the temptation of wretched excess usually produced either architectural dissonance or a kind of Mr. Potatohead image, as if design was just a matter of selecting and sticking components onto an inert building lump.

Too often, builders and architects acted like cosmeticians, the exterior counterpart of interior decorators massaging the inside with sconces, potted plants and wallpaper. A fairly simple, unpretentious house, one with a perfectly workable layout of spaces, would be transformed into marketable architecture by a process of applique and, in some instances, needless deformation.

Consequently jurors saw disparate architectural elements contributing little to establishing or reinforcing some overall, unifying spatial or volumetric theme. Typically there was no theme and no decipherable language of design and construction expressing something about how the house was built or used.

Wanting to fragment the bulk and enrich the visual scale of otherwise simple, boxy volumes, designers couldn't resist making arbitrary shifts and jogs and rotations, a kind of architectural "bumping and grinding," in floor plans and site plans. Such moves can produce awkward spaces inside as and complicated volumes outside. Roofs can get especially awkward when sheared, segmented, tilted or rotated to accommodate bumps and grinds below.

Such geometric and decorative habits seemed particularly unforgivable in some of the smaller houses. Their forced architecture was unduly complex and tortured, and the encrustation of elements affixed to facades looked ready to fall off in the first storm or earthquake. Of course, architectural boogie-woogie, by distorting the apparent scale of a building and perhaps reminding us of historical antecedents, is intended to make a small house look superficially like a much bigger one.

Indeed, it was the overwrought superficiality and faddishness of most designs, and not any specific element or decorative motif or style, that caused the jury to reject them. Those few that we blessed and recommended for awards were diverse in size and in stylistic approach, and all had been orchestrated artfully to make coherent buildings, and rooms within, even when the first impression was one of collage and variety.

And most winners exhibited some degree of honesty, as far as can be determined from photographs, in their use and expression of materials -- a notable achievement given the number of opportunities and products available to make houses look like they are built out of something they are not.

Consider also the floor plans that we had to comb through as we looked for designs that seemed at least sensible, if not innovative or exceptional. Based on what we saw, consumers appear ready to accept almost anything, including many of the ghastly decorating suggestions found in model homes.

Some dwellings offered separate dining rooms big enough for a table, but not for chairs. Occasionally living rooms and bedrooms, even relatively large ones, were difficult to furnish because of conflicting circulation patterns and placement of windows, doors, fireplaces and closets. By contrast, most kitchens and bathrooms seemed adequately configured.

We particularly looked for healthy marriages between building and landscape, for projects that capitalized upon or took significant cues from their site. In many cases, yards or courts seemed residual, no more than left-over space where the house wasn't. You walked through a door or looked through a double-hung window, and you could find nature, but it was nature contained in a separate zone.

Better designs wove together the interior and the exterior. Buildings could extend themselves by capturing usable outdoor space with loggias and trellises, or by partially enclosing courtyards with wings and walls. Some of the nicest rooms in the nicest houses opened generously onto decks, patios and porches.

Perhaps it's not surprising that we found more award winners among the custom houses than among the built-for-sale houses. Most of the former were submitted by architects, the latter by developers. Yet construction cost was not the difference; there were a number of good, modestly budgeted custom houses.

Economics does not explain why production houses can't be as well designed as custom houses. But there is an economic reason why so many production houses tend to be overdesigned: marketing and consumer taste. Judging from what we saw, most builders and buyers still operate on the "curb appeal" theory. If in doubt, add "zoot" and it will sell. They may be right in the short run, but in the long run they create a shallow architectural legacy -- not to mention future maintenance headaches. Roger K. Lewis is a practicing architect and a professor of architecture at the University of Maryland.