Across Congress, the estimated taxpayer exposure to special-election costs from potential seat vacancies totals $128M. That figure is not a prediction that seats will vacate. It is a risk estimate, the kind an actuary runs before anyone gets sick, retires, or resigns. It exists because vacant congressional seats are not free to fill. Every special election requires ballots, poll workers, election infrastructure, and administrative overhead, all paid for by the public whether or not a given seat ever actually opens up.
Think of $128M as a standing liability sitting quietly on the books of the federal system, distributed unevenly across 535 seats. Most of that liability will never be paid out, because most members will finish their terms. But the number does not go away just because it is unlikely to be triggered. It sits there, a permanent line item in the cost of representative government, and this site's job is to quantify it seat by seat rather than treat it as an abstraction.
Five members of Congress currently carry a completion probability under 50 percent, meaning the site's model gives them less than even odds of finishing out their current term. That is the group worth watching, since a low completion probability is what turns a theoretical special-election cost into a more plausible one.
Separately, the site tracks which members would trigger the highest special-election cost if their seat vacated today. The top five, by cost, are:
Maxine Waters, California, House: $100M estimated cost, 66 percent completion probability.
Nancy Pelosi, California, House: $100M estimated cost, 70 percent completion probability.
John Garamendi, California, House: $100M estimated cost, 83 percent completion probability.
Doris O. Matsui, California, House: $100M estimated cost, 83 percent completion probability.
Zoe Lofgren, California, House: $100M estimated cost, 88 percent completion probability.
Notice that all five carry the identical $100M cost figure. That is not a coincidence of politics, it is a function of geography and election administration cost structure, which the methodology section below explains. Notice also that completion probability varies meaningfully within this group, from 66 percent up to 88 percent. Cost and risk are two separate variables, and reading them together, not in isolation, is the entire point of this dataset.
It is worth pausing on why five California House seats dominate the top of this list. Special-election costs scale with the size and complexity of the jurisdiction required to run a standalone election. California districts, with large populations and extensive election infrastructure requirements, generate higher baseline costs than smaller or more sparsely populated districts elsewhere in the country. The members themselves are almost incidental to the number, the geography is doing most of the work.
A useful way to think about this data: cost tells you what a vacancy would cost, completion probability tells you how likely that vacancy is. A high cost paired with a high completion probability is a low-concern data point, since it reflects a seat that is expensive to refill but statistically unlikely to need refilling soon. A high cost paired with a low completion probability is the combination worth attention, because it stacks real financial exposure on top of real uncertainty.
This is why Too Damn Old separates the two metrics rather than blending them into a single score. Readers using this site to understand vacancy risk should look at completion probability first to identify which seats are genuinely uncertain, then check the cost figure to understand what is financially at stake if that uncertainty resolves into an actual vacancy. Neither number alone tells the full story. Together, they give a clear, non-speculative picture of where taxpayer money is most exposed to the mechanics of congressional turnover.
This matters practically for anyone trying to understand where public money is at risk in the ordinary functioning of Congress, separate from any political judgment about individual members. A reader comparing districts, states, or chambers can use these two figures side by side to see where the system carries the most combined financial and probabilistic exposure, without needing to speculate about why a given member's completion probability sits where it does.
Special-election cost estimates are built from the administrative and operational expense of running a standalone election in a given district or state, reflecting the fact that election costs vary by jurisdiction size, infrastructure, and existing election-administration budgets. Completion probability is a separate calculation reflecting the statistical likelihood that a sitting member finishes their current term, based on factors the site tracks independently of any cost projection. Neither figure is a forecast of intent. Both are drawn directly from Too Damn Old's own dataset, computed on a per-member basis, not estimated or adjusted after the fact for this piece.
The two calculations are deliberately kept separate rather than merged into a single composite score. Cost is a structural fact about a jurisdiction, largely unrelated to the individual holding the seat. Completion probability is a per-member statistic. Keeping them apart lets readers reason about each independently instead of relying on a blended number that would obscure which variable is actually driving concern for a given seat.
For the complete breakdown of how special-election costs and completion probabilities are calculated, including the underlying data sources and assumptions, visit toodamnold.com/data.