2026 US Midterms · Follow the money

The Best Politics Money Can Buy

Who is paying for the 2026 congressional elections, what they want in return, and how the money lines up — or fails to line up — against what voters say they actually care about.

Voters care about six things. The money is chasing three other ones.

Start with the finding that organises everything else in this dashboard. The six issues below are what registered voters name, unprompted, as their priorities for 2026. Set against them are the three industries writing the largest cheques of the cycle: crypto, artificial intelligence, and foreign-policy advocacy. None of the three appears anywhere near the top of the public's list.

Two ways to guess the result

Before any of the detail, the blunt question. If you projected November purely from the polls, you get one Congress. If you projected it purely from who has raised more money, you get a different one. Both models are built below from the same eight decisive Senate races and the same eighteen House toss-ups, so the gap between them is the gap between what voters are saying and what donors have bought.

Senate — 100 seats, 51 for control

House — 435 seats, 218 for control

Notice that the money model is not simply the pro-Republican one. Republican-aligned megadonors have out-given Democratic ones roughly three to one this cycle ($1.05bn to $360m). Yet in the races that decide the Senate, Democratic candidates have out-raised Republican ones almost everywhere — Ossoff by 13:1, Talarico by 7:1, Brown by 2.7:1. Hard money and outside money are flowing in opposite directions. Any claim that "money buys elections" has to account for a cycle in which the party with less outside money has more candidate money in every race that matters.

How each model calls the eight decisive Senate races

SeatHeld byMoney gapMoney model Poll marginPolls modelAgree?

Method. Baseline is the current chamber: Senate 53R–47D, House 220R–215D at full strength. Seats not listed as competitive are held by the incumbent party in both models. The money model awards each competitive seat to the candidate whose committee has reported more total receipts through 30 June 2026 (FEC); where a party has an unresolved primary the leading fundraiser is used. The polls model uses Decision Desk HQ polling averages and the most recent public surveys for the Senate, and published forecaster central estimates for the House. The two models are not built to the same tolerance: the money model only reallocates seats already rated competitive, while forecast-based House projections also move some Lean-Republican seats. The money model is therefore the more conservative of the two by construction, and its narrower House majority partly reflects that.

The eighteen House toss-ups, by money gap

Difference between the Democratic and Republican candidates' total receipts in each seat the Cook Political Report currently rates a toss-up. Bars right of centre are seats the money model gives to the Democrat, left of centre to the Republican. Ten of these seats are Democratic-held and eight Republican-held; the money model produces four flips (AZ-01, AZ-06, NE-02, PA-10) and no losses, for a net Democratic gain of four and a 219–216 chamber. Maine's second district is the closest call in the country: Jared Golden leads Paul LePage by about $18,000. Where a primary is unresolved, the leading fundraiser in each party is used. FEC receipts through 30 June 2026.

Voter salience vs. dedicated outside money

Both sides are shares, so they are directly comparable. Left: each issue's share of all voter attention, from the share of registered voters naming it as the one thing they most want their congressional candidates to discuss (Pew Research Center, open-ended question, 6–12 July 2026, n=3,554 adults / 2,779 registered voters). Right: each issue's share of all money raised by dedicated single-issue outside spending vehicles in the 2026 cycle (FEC, through 30 June 2026). Cyan issues are the ones voters do not raise unprompted. Healthcare, energy and abortion show no right-hand bar because no single-issue super PAC of comparable scale exists for them — those industries give heavily, but diffusely, through hundreds of small corporate and trade PACs rather than one named vehicle.

Read this carefully. The two scales measure different things and are deliberately shown side by side rather than combined into a single index. A low-salience issue attracting large money is not proof of corruption — niche industries facing existential regulation have an obvious rational reason to spend, and voters can be uninterested in a question that still matters. What the chart establishes is a structural asymmetry in who is buying attention, not a finding about anyone's motives.

The six issues, and who is funding each side

IssueVoter saliencePrincipal money on this issue 2026 receiptsDirection

Salience figures combine Pew's open-ended "what should candidates talk about" measure with Gallup's most-important-problem series and KFF's issue-importance tracker; where sources differ the range is shown. Money figures are FEC-reported 2026-cycle receipts for the committees named, through 30 June 2026.

Which party voters trust on each issue

Percentage-point advantage for the Democratic (blue) or Republican (red) party, "which party do you agree with on…". Pew Research Center, 6–12 July 2026. Note the bottom row: on AI policy, 51% of adults say they agree with neither party — the single largest "neither" figure Pew recorded, on the very issue where industry money is growing fastest.

The contributors

Fifteen committees account for the overwhelming majority of independent expenditure in the 2026 cycle so far. Each is shaded by the policy interest it exists to advance. Bar length is money actually spent on independent expenditures — advertising for or against named candidates — not money merely banked.

Source: FEC PAC totals endpoint, cycle 2026, sorted by independent expenditures, coverage through 30 June 2026. Hover for receipts and cash position.

CommitteeInterestInd. expenditure Total raisedCash on handFEC ID

The single largest concentration: three crypto firms, one network, three cheque-books

Coinbase
$48m+ disclosed to Fairshake · Oakland, CA
Ripple Labs
$48m in two $23–25m tranches · San Francisco, CA
Andreessen Horowitz
$23.8m via AH Capital Management · Menlo Park, CA
Fairshake
$137.0m raised · $127.0m still unspent · $51m transferred onward to affiliates
Defend American Jobs
Republican-facing arm · $29.4m spent attacking and defending
Protect Progress
Democratic-facing arm · $22.7m spent

All figures from FEC Schedule A itemised receipts and PAC totals for committees C00835959, C00836221 and C00848440. The three-entity structure lets the same industry fund candidates in both party primaries without any single committee appearing partisan. Reported network war chest across the three vehicles: roughly $193m. Fairshake's advertising, notably, rarely mentions cryptocurrency at all.

Why this is the cleanest example in the dataset. Most industry money is diffuse — spread across hundreds of corporate PACs giving $5,000 at a time, where attributing intent is guesswork. The crypto network is different: a named, self-declared single-issue vehicle, funded by three identifiable companies with a specific bill agenda, spending in both parties' primaries. It is the rare case where donor, interest, amount and target are all documented in the same filings.

Individual megadonors, first half of the 2026 cycle

Large disclosed donations to federal super PACs and party vehicles, January–June 2026, as compiled by the Washington Post's megadonor tracker. The top 50 donors combined gave about $1.6bn. Note that Andreessen and Horowitz appear here as individuals in addition to their firm's corporate giving to Fairshake shown above.

The candidates

Every Senate candidate below has raised more than $6m this cycle. Bars show total receipts through 30 June 2026 into the candidate's own committee — this is hard money and excludes the outside spending covered in the previous tab, which in several of these races exceeds the candidates' own budgets.

FEC candidate financial totals for the 2026 cycle, refreshed automatically. Includes every major-party Senate candidate above $6m in the states this dashboard tracks — some are no longer running, having lost a primary or withdrawn; the money was still raised and spent.

Marquee races: money on both sides

RaceDemocratRaisedRepublican RaisedD:R ratioPolling average

Where each contender stands on the six issues

A directional summary of publicly stated positions and voting records — not a precision instrument. Read the caveat below it before drawing conclusions.

CandidateRaceCost of livingHealthcare ImmigrationGovt ethicsEnergy/climate Abortion

Scale: ++ strongly progressive position · + leans progressive · ~ mixed, moderate or unclear · leans conservative · −− strongly conservative. Hover any cell for the specific position it encodes.

This table is the weakest data in the dashboard and should be treated as such. Campaign finance figures come from mandatory sworn filings with an enforcement agency. Positions come from campaign websites, floor votes, endorsements and press coverage, compressed by hand onto a five-point scale. Two honest analysts would code some of these cells differently. It is included because the money story is unreadable without some sense of what the money is buying — but any specific claim about a specific candidate should be checked against the primary record before it is repeated.

Money against polls

The direct question: in the eight races that will decide Senate control, does the fundraising gap predict the polling gap? Each dot is a race. The horizontal axis is how much more the Democrat raised than the Republican; the vertical axis is the Democrat's margin in the polling average.

Fundraising: FEC, through 30 June 2026. Polling: Decision Desk HQ averages and the most recent public surveys as of early July 2026 (Times/Siena, Fox News, AARP/Fabrizio-Impact, Emerson, Alaska Survey Research). Michigan is plotted with the leading Democratic primary candidate pending the 4 August primary.

The correlation here is real but weak, and the causal arrow points both ways. Ossoff sits far to the upper right — most money, biggest lead — which looks like money buying a lead. But a senator polling 13 points ahead in a swing state is exactly the kind of candidate donors rush to fund, so the lead plausibly caused the money rather than the reverse. Meanwhile North Carolina and Ohio have very different money gaps and nearly identical dynamics, and Iowa is tied despite the Republican outraising the Democrat by 63%. Fundraising and polling are both downstream of a third thing — perceived viability — which contaminates any simple reading.

Race by race

Blue and red bars: candidate receipts in millions. The line marker shows the current polling margin, positive toward the Democrat.

The clearest test the 2026 cycle has produced so far

Texas: $35.4m lost to $9.2m

Texans for a Conservative Majority spent $35.4m in independent expenditure — the largest single-committee outlay of the cycle — defending Senator John Cornyn. Cornyn's own committee raised a further $14.3m. Their combined $49.7m lost the 26 May Republican runoff to Ken Paxton, whose campaign had raised $9.2m. A five-to-one money advantage did not survive contact with a primary electorate that had already decided what it thought.

Georgia: $98m against a fragmented field

Jon Ossoff raised $98.0m — more than the entire Republican primary field combined, several times over. Mike Collins won the June runoff on $7.4m. Ossoff leads 56–43 in the only recent public poll. Here money and position move together, which is precisely why the race proves less than it appears: an incumbent with a national donor base and a 13-point lead is both cause and effect of his own fundraising.

Where outside money is aimed inside a single party

Michigan's 4 August Democratic primary is the cycle's sharpest illustration of outside money shaping a choice before voters get a general election. United Democracy Project, the super PAC affiliated with AIPAC, has raised $104.0m this cycle and spent $27.9m on independent expenditure — including millions supporting Representative Haley Stevens against Abdul El-Sayed, who is openly critical of AIPAC's influence. El-Sayed has out-raised Stevens $14.5m to $11.9m in hard money. General-election polling has both Democrats running roughly one point ahead of Mike Rogers, meaning the outside money is not resolving an electability question so much as choosing between two similarly-positioned candidates on a foreign policy question that 4% of voters name as their top issue.

Does money actually win?

Usually — but the headline number overstates the case badly, and understanding why is the most useful thing on this page.

Share of races won by the candidate who spent the most, by chamber and cycle. Source: OpenSecrets post-election analyses, 2020 and 2022 cycles.

Why 94% overstates it

Most congressional districts are not competitive. In a safe seat the incumbent raises more money and wins, but the money is not doing the work — the district is. Because safe seats dominate the denominator, the aggregate win rate mostly measures how good donors are at identifying likely winners, which is a much less interesting skill than buying elections. The Senate figure is consistently lower than the House figure for exactly this reason: statewide races are harder to call in advance, and in 2020 the top Senate spender won only 71.9% of the time.

What the money does buy

The defensible claims are narrower and mostly about the stages before election day. Money buys the ability to run at all; it buys name recognition against an unknown opponent; it buys the deterrence of strong challengers, who look at a $42.6m war chest and choose a different race; and it buys advantage in low-turnout, low-information primaries where few voters have prior opinions. The 2026 cycle's own evidence — Paxton beating a five-to-one spending advantage in a high-salience primary — suggests money matters least exactly where voters already have views.

Other mechanisms visible in the 2026 filings

MechanismWhat the data showsFigure

Where the money comes from, by candidate

Total receipts against cash still unspent, for both major-party candidates in each decisive Senate seat. A campaign sitting on a large unspent balance in July is buying the final eight weeks, when advertising is most expensive and least answerable. Figures update automatically from FEC filings.

Method, limits and sources

What is solid

Every dollar figure in this dashboard comes from a mandatory filing with the Federal Election Commission, retrieved automatically from the OpenFEC API by a scheduled job on this server; the freshness strip at the top of the page shows when the last successful pull ran and which reporting period it covers. Committee identifiers are given so any figure can be traced to source. Polling figures come from named public pollsters and published aggregator averages. Voter salience figures come from Pew, Gallup and KFF with field dates attached.

What is soft

Sources