As @timcliff noted in his reply, you are absolutely right. But at the same time I don't think it is completely wrong to say that it 'works' in a sense, since it did not only return to $1 but it did so largely by the intended mechanism of printing more and more of it until the price declines.
In the short term, yes speculators can move the price (absent additional mechanism to peg it more strongly), but the higher the price goes, the more capital inflow is required to hold it there against the steady "drip, drip, drip" of continued new supply, and that ultimately becomes unsustainable. While sometimes slow, even a steady drip can eventually produce the grand canyon.
Or another way to say that is the speculators are not entirely external to Steemit, because Steemit users can and do "cash out" their rewards. When they do so, it requires a speculator to inject additional money to buy those newly-rewarded SBDs at the inflated price, or the price must decline.
Thanks for the explanation @smooth; appreciate you taking the time. Steemit/Steem seem like interesting experiments, in their own right.
My interest in the SBD peg primarily comes from a commercial interest. It's functionally difficult to use a currency to trade goods and services if it fluctuates wildly.
Quoting from one of my other replies here:
As noted by @timcliff in the post, perhaps other changes can be made in the future to reduce the upward spikes. Even without changes I do think it is possible (though hardly certain) that the price may become more stable over time as the system matures.