I agree with the well articulated points above (about tillering), but I think we're straying a bit from the original question which is about the ratio of bow length to draw length with respect to arrow speed.
It appears to be complicated. But we can break it down a bit, right?
For example, the longer the draw -- the longer the power stroke. E.g. the longer the draw, the more time an arrow is in contact with the string and accelerating, which allows the bow to impart more stored energy into the arrow. So longer draw (in a vacuum) is a plus. BUT, if that longer draw is gained by stressing the wood, and increasing set, well than you may have negated its benefits by reducing stored energy and increasing hysteresis.
So, one way to talk about design is to say...the longest draw you can get without increasing set or stressing/breaking down the wood is a tillering goal when making wooden bows that you want to shoot fast. Usually though draw length is about the person shooting the bow, not necessarily how fast you can make an arrow go...
Just a few thoughts. I'm open to changing them if somebody can convince me otherwise.
Gabe