Share of new queue entries by resource, each year from 2000. Hybrids follow the queue’s own convention — a solar+battery project counts as solar — so these bands are the generating technology, and standalone storage is the yellow band.
Gas built the queue of the 2000s. Solar took it over after 2015. Standalone storage went from a rounding error to nearly a third of everything filed — in about three years. Starts at 2000: 1995 contains a single project and 1996 none at all, so earlier years are noise, not trend.
Battery share of entries per county, before and after 2023. Counties with fewer than 3 entries in the window are left blank rather than shown as noise.
The intuition is irresistible: red counties fight renewables, blue counties welcome them, so politics should predict which projects die. Here are both maps. They look nothing alike — and that is the entire finding.
Counties with fewer than 5 resolved projects are blank. Withdrawal is the majority outcome almost everywhere — the right-hand map is dark nearly everywhere there is data.
This is the visual form of ADR‑008: rural/metro status, political lean, and state RPS were all joined leakage-safe as-of entry, and none of them move the metric. It is a real negative result, published rather than buried — and it is why this page has no “withdrawal risk by county” map. That map would be fiction.