In my experience, most multi-band verticals that work on more than three
bands start to become seriously compromised on the lowest band. Sometimes
they are compromised on the highest band, as well.
The lowest-band compromise generally comes from having gadgets and gew-gaws
and all manner of what-not strung all over them to create resonances in the
ham bands. These generally soak up the juice, create high currents in many
exposed connections, and make the current distribution, uh, strange.
On the other end of their coverage, these verticals may be too physically
long to create a single, coherent current distribution that creates a
useful, low-angle radiation pattern. The trapped verticals, such as Hustler
6BTVs, do pretty well on their highest band because they become a single
near-quarter-wave radiating element there.
The trend towards adding coverage of the WARC bands to the traditional
80-40-20-15-10 (which are kind of harmonically related) complicates
multiband antenna design greatly, generally requiring fiddly bits, traps,
coils, strategically-placed capacitive hats, and miscellaneous hoo-hah.
The key thing to remember for multi-band antennas is that matching is not
necessarily radiating. To radiate, electrons must accelerate back and forth
in a straight line. Any folding, winding, or cross-coupling that reduces
the net amount of straight-line, back-and-forth acceleration of our little
negatively-charged buddies results in reduction of the amount of RF launched
into the aether. Like Steve said, we have written about this extensively in
our Antenna Test books.
73, Ward N0AX