Yeah, and contrary to what some think, CRS took the blame for the WW2OL fiasco on themselves; they did not shift it to strategy first.
A distributor doesn't say all of a sudden "Hey, I got an idea, you are scheduled to release in fall 2004, but we're going to release in fall 2003."
The problem is usually inexperience and optimism. My experience is that debugging time increases something on the order of the complexity of the software to a power of two. In other words, a game with twice the features of another will take four times longer to debug. Add to this the naive enthusiasm of coders, who when the bulk of the coding is done think the product is going to work out by the third compile at the latest, and you get products that always seem to finish late.
Now you've got a publisher, who has reserved some time with the duplicators, worked up release date to fit in with other products, has made the contacts with the retailers, and set up the advertisements. A delay in release is going to be costly. Usually, the contract will specify that the developer who incurs the overruns incurs the costs (=pays a penalty). CRS couldn't afford those costs, and odds are Mr. Smart couldn't either.
But when it comes time to assign the blame or credit for a product, that goes always and ultimately to the developer. The developer selects the distributor and agrees to the terms. If the developer doesn't deliver a functional product on the deadline, don't put the blame elsewhere.
And,y eah, Mr. Smart better have "escape" clauses in everything else he releases. That's the only way he can get people to trust him.