First, remember that there is a lot of base load (!) gas power in the US. Second, a robust continent-sized HVDC grid (which you'll need to get all that "stranded wind" to the consumers) should smooth out the intermittency. And for those days when that isn't enough, the gas plants are still there and can be fed a little gas until the situation rights itself.
When it comes to coal you can replace it with nuclear, MW for MW and BTU for BTU.
And I don't worry terribly much about GHG emissions. Peak oil is my bogeyman. Peak oil is not an energy crisis. It is a liquid fuel crisis.
But that only answers half the question ... why replace the lesser of two evils in terms of fossil fuel electricity while leaving the greater of two evils alone.
It does not answer the second half of the question ... how to rig the policy so that it is, in fact, natural gas taken off line, when on in a pure commercial calculation, expanding wind power increases the appeal of natural gas turbines compared to thermal coal. I've been accused of being a Marxist, yet while Harpo's my favourite, it's Groucho I'm always quoting. Odd, that.
For the second half of your question, the policy is simply a blanket ban on building new coal plants or upgrading the current ones in such a way that they consume more coal. Peak oil is not an energy crisis. It is a liquid fuel crisis.
Nuclear is inflexible and will have a hard time replacing gas in the peak role. But it works in the baseload, and I did write that nuclear should replace coal, not gas.
And among the non-renewable energy sources, gas remains a more natural complement to wind than either nuclear or coal.
The second half of the question is how to rig the system to force out natural gas. Any policy that successfully promotes wind will differentially push out coal rather than natural gas, so channeling the displacement from coal to natural gas requires an additional policy intervention. I've been accused of being a Marxist, yet while Harpo's my favourite, it's Groucho I'm always quoting. Odd, that.