Polymarket Election Whales: Who Holds the Midterm Markets

The biggest wallets in election prediction markets are visible on-chain. We census them across every race we track. Here is what the top-20 boards show.

GhostLeg, 2026-08-11, 5 min

Holder data in this post is a snapshot from Thursday, August 7, 2026 (latest capture 8:40 AM ET). Top-20 boards refresh through the day; the live race pages carry the current state.

Election markets have whales, and on Polymarket you can see them

In sports prediction markets, we have been tracking Polymarket's biggest wallets for months: every position on the exchange settles on-chain, which makes the big money visible in a way no sportsbook or regulated exchange allows. The same transparency applies to the election markets, where the stakes of the 2026 midterms have drawn serious size.

GhostLeg now runs a whale census across the election board: for every tracked contract, we capture the top-20 holders every six hours, currently across 190 contracts spanning the 87 open races on our politics board. Each race page shows its own Big Money card. This post is the cross-race view: who shows up on the most top-20 boards, and what their footprints look like.

Two honesty rules before the names, because this data invites over-reading:

  1. The census sees top-20 boards only. A wallet holding positions too small to crack any contract's top 20 is invisible to it. This measures the big money, not all money.
  2. The names are Polymarket's own public pseudonyms. Every wallet on Polymarket gets a public handle; that is what we show, exactly as the venue does. We do not connect wallets to people, and neither should you.

The cross-race census

As of this snapshot, the wallets appearing on the most top-20 holder boards across our tracked races:

Wallet (Polymarket pseudonym) Races with a top-20 position Total shares held
Ample-Instance 65 ~107,800
Belated-Glue 62 ~7,300
Bossy-Wasting 58 ~13,000
Tiny-Maracas 54 ~8,400
Those-Skywalk 51 ~14,700
Little-Gold 46 ~42,700
Our-Dough 42 ~136,500

A note on units: shares, not dollars. One share pays out $1 if its outcome resolves yes, so a share count is a maximum-payout figure; what a wallet actually paid depends on the prices it got.

Breadth and size are different strategies

The table's most interesting feature is the contrast between its two extremes.

Belated-Glue sits on 62 different top-20 boards holding only about 7,300 shares in total: an average of roughly a hundred dollars of maximum payout per race. That is not conviction sizing; it reads like indexing, someone spreading small positions across the whole election board.

Our-Dough is the opposite: 42 races but roughly 136,500 shares, the largest total in the census: fewer markets, real size. Ample-Instance does both at once, appearing on more boards than anyone (65) while carrying six figures of shares.

We report the footprints, not the motives. A large position is not automatically a prediction: it can be a hedge against another position, a market-making inventory, or the second leg of a spread across venues. What the census gives you is the shape of the big money, live: when a new wallet suddenly appears across a dozen race boards, or a longtime holder exits one, that shows up in the six-hourly snapshots on the affected race pages.

Why this only exists for Polymarket

There is no Kalshi column in this census, and there cannot be one. Kalshi is a CFTC-regulated exchange with private brokerage-style accounts; no public record connects positions to holders. Polymarket settles on-chain, where every position is publicly auditable by design. This is the same structural split we documented for sports whale tracking: transparency is a Polymarket-specific property, not a prediction-market property.

That asymmetry is worth remembering when the venues disagree. Our politics board tracks Kalshi-Polymarket divergence on every race, and on Polymarket's side you can at least inspect who is holding the disagreeing side. The current gaps, including a 25-point one on House control, are documented here.

How to use it

FAQ

Can you really see who is betting on the midterms? On Polymarket, you can see every wallet's positions on-chain, under the venue's public pseudonyms. On Kalshi, accounts are private; no equivalent view exists.

Are the whales usually right? In elections, that is a graded question our record will answer as races settle; the census is new and the midterms have not resolved. On sports markets, where our graded wallet records run deeper, the answer is: some are, most are not, and the track record is the filter.

Does a big position mean the whale thinks that outcome wins? Not necessarily. Hedges, market-making, and cross-venue spreads all produce large positions without a directional view. Footprints, not forecasts.

How fresh is the holder data? Top-20 boards are captured every six hours per contract, across 190 contracts as of this snapshot.


GhostLeg reports what prediction markets price and what public on-chain data shows. Prices and positions are not predictions or recommendations. For entertainment purposes only. Past performance does not indicate future results.

All GhostLeg blog posts