Short manifolds enhance power production at high rpm.
With a longer manifold, peak volumetric efficiency (VE) occurs further down the rpm range - better for all round driveability on the street.

It's not just about the manifold... you must take into account OVERALL length of the entire induction system: from top of velocity stack to valve seat.
Running a short manifold, very tall velocity stack combo has almost the same effect as long manifold with short stack - if OVERALL combined length is the same.

There's a very narrow rpm at which optimum wave tuning boosts volumetric efficiency (and therefore torque and power).
Overall length of the intake tract is used as a tuning tool to maximise power output at a specific point in the rpm range.

High rpm engine .. use shorter overall length
Low rpm, street use .. go longer

A set of short manifolds doesn't restrict tuning the system to high rpm... fit taller velocity stacks to increase overall induction length.
Testing on the dyno shows peak torque drop down the rpm range as length of intake system increases.

Couple of examples, with optimal length tuned between peak torque and peak power rpm (measurement is from valve seat to top of stack)

2110cc street engine, 150bhp @ 5900rpm ... 16"
2332cc n/a race engine, 268bhp @ 8300rpm... 11"

You're robbing Peter to pay Paul but if the area you're robbing from is seldom used, it makes sense to tune things to your advantage so performance is improved in the area you spend most time.

Some modern day engines use variable induction length to increase torque across a much wider rpm range than is possible with a single fixed length.

If you're building a large capacity, low rpm, torquey street motor (i.e. you know you need a fairly long intake length), better to start with a taller manifold from the off, as opposed to short manifolds and very long stacks...
the further away the butterfly is from the valve, the better chance you have of feeding well atomised fuel into the combustion chamber.

Most engine bays limit the type/length of induction system.
We're also restricted (in most cases) by what's commercially available.
And some will ignore all the physics and go for what looks "cool" 

Interesting, and the opposite of what I thought, can you explain this?
I talked with Johannes at JPM about this some weeks ago and he said to tune for 3rd pulse up to a certain duration, and 2nd above this... have you found this to be true also?

i reread this topic today. trying to figure stuff out. i flipped through both testaments last night looking at engines. and the many short manifolds that were ran back in the day. albeit the early 70ies, late 70ies cars seemed to run tall manifolds.
but the only conclusion i can come to is to "compensate" for the cam. i looked at the specs as well, and most of the short manifold engines ran a w-110 engle. whereas, the guys running a w-130 or w-140 ran tall manifolds. so, i take it the tall manifolds gave some bottom end power to the higher strung engines? and vice versa on the relatively mild choice of cam on the 1700 engines...

could this in any way be correct? oh and by the way, naturally they all ran the same velocity stack, so no fooling around there

My head's starting to spin

A lot of good info here and great article John!
I always knew longer manifolds and ported heads increased air speed (port velocity) which makes torque at lower RPMs,
and now I have the math to analyze it. I'm just not sure I want to go thru the brain damage to calculate it out.
Sarge said it.... My heads starting to spin