Dear Adam,

Thank you for your message and your questions.

You ask ' What's the point of a design that isn't actualised?' 
Let me know if I'm wrong. I'm assuming you mean 'why distinguish between the
design and the real thing'? If so, there are some obvious practical reasons.
If you asked a car dealer to sell you  a particular design of car and they
took your money and gave you a set of engineering drawings (the design), you
might be upset (on the other hand you might be REALLY delighted (if you
wanted to make some serious money!)).

You ask, 
' [You state that the important aspect of design research as you see it is
the
capacity to predict user outcomes,] but how are you going to do that without
making something and gathering user data?

It is often necessary for designers to design things that haven't been made
before.
If a designer makes something and gathers user data about it, then the only
benefit will be in designing the same thing. What is needed IN ADDITION is a
theory model that predicts behaviours. Being able to combine relevant data
with theory that predict behaviour of designed outcomes  enables you to
design things that you haven't already made and done user testing for. 

Sometimes the use of behavioural design theory is simply beneficial for
reducing the time needed to create good designs that can be justified
(remember the legal action...). For example, a good behavioural theory about
website usability can provide the basis of website design guidelines that
justifiably predict  behavioural outcomes for any website that fit with the
applicability bounds of those guidelines, i.e. can support a designer to
design a good website. 

In other areas of design, it simply isn't possible to make something and
gather user data before creating a design proper. For example, the Chinese
government is currently designing strategies and policies for whole of China
rapid economic development whilst avoiding overheating some areas of the
economy and minimising social unrest below the point that it will
destabilise the government - a difficult design task as the only factors
that increase economic development also increase economic overheating and
social unrest. It is possible to use data from previous policy designs  (in
China and elsewhere), but to do that in a way that can be trusted, needs a
behaviourally-based suite of design theory that enables one to use
historical data to predict outcomes in current policy design. This kind of
design activity has already associated with it a raft of
behaviourally-based design theories (though not usually called such) from
areas including  economics, operations research, organisaitonal dynamics,
system dynamics, geo-politics, social dynamics, sociology, political economy
and historical analysis.

You say,
' Kind of sounds like a mathematics which is all workings and no answers to
me...'

Wondering what you mean. Please can you say more?

Best wishes,
Terry