The statutory language at issue in this case was almost certainly not an instance of sloppy drafting, but one of many provisions that were designed to get citizens of each state to put pressure on their state governments to participate in setting up their own exchanges, at great expense that would break the budgets of most of them. These "nudges", as regulatory czar Cass Sunstein calls them, can work over the course of decades, if they are small, but it is a miscalculation when to work they would have to arouse public pressure within days or months, enough to overcome expenses that the states cannot responsibly incur.

That is not to say that statutory language is not often sloppy, as is regulatory and judicial language, but as often as not it reflects a miscalculation of how the provisions adopted will play out in the real world, something that does not come out until attempts are made to apply it.

Human beings are generally not very good at anticipating all of the consequences of interventions in complex systems, especially not members of Congress or their staffers. Most such legislation is not drafted by them, but by outside lobbyists, who have agendas of their own that do not include anticipating every adverse consequence.

The ACA was largely written by lobbyists for the health care industry, especially the health insurance companies, who offered their support for it because it promised to divert large sums of public funds into their profits. But there is no way that could be done without imposing additional costs on health care, which are already increasing without limit as new, more expensive treatments become available. At some point there have to be "death panels", and this administration wanted the insurance companies to take the heat for operating them. Other countries use "single-payer" systems, which are not financially viable, but in the U.S. they are also not politically viable, in part because they put the heat on the government for denying care. Now the reasons for that nonviability are becoming clear.

For an interesting study see the 1970 paper  "The Counterintuitive Behavior of Social Systems", by Jay Forrester.