← Back to the article

Provenance · The Debate

The debate behind The Election Odds Are Becoming Campaign Material

The questionAre Prediction Markets Measuring Politics or Manufacturing It?

How this debate works

Before writing, The Arbiter stress-tests each story by framing the two strongest opposing positions and arguing both sides of a structured three-round debate: opening arguments, rebuttals, then steel-manning the opponent and answering one question — what specific, verifiable evidence would change my mind?

OpenAI GPT-5.5 argued both sides under a debate constitution that requires empirical evidence, specific citations, and engaging the strongest version of the opposing argument. The published article was written separately: the debate supplies the questions, and the author verifies key claims with its own research before taking a position.

Sources in this transcript are evidence as each advocate presented it during the debate — research leads, not independently verified endorsements.

Why we covered this

Strong publish. It connects election odds, influencer promotion, market structure and legitimacy narratives in a way mainstream campaign coverage often misses. Clear research path and broad relevance for politics, media, finance and regulation.

The positions

Advocate A argued

Election prediction markets can remain useful, reliable information tools if they are treated as regulated financial-information venues rather than neutral public-opinion gauges. Under this view, platforms like Kalshi and Polymarket do not become invalid simply because influencers, insiders or partisans try to move prices; every market faces manipulation risk. The key question is whether rules on paid promotion, disclosure, wash trading, concentrated positions, employment-based insider access and enforcement are strong enough to make prices more informative than polls, punditry or social-media narratives. Advocates for this side would argue that CFTC-supervised markets, transparent order books, stricter influencer-disclosure policies, liquidity monitoring, whistleblower systems and limits on conflicted trading can expose rather than conceal politically motivated narratives. Even thin or volatile odds may still provide valuable real-time signals when users understand liquidity, trader concentration and regulatory context.

Advocate B argued

Election prediction markets are increasingly at risk of manufacturing political reality rather than measuring it, especially when their odds are amplified by influencers, campaigns and media outlets as evidence of momentum or legitimacy. Under this view, thin liquidity, concentrated whale positions, paid promotion and possible insider access can allow a small number of actors to move prices and then cite those prices as proof that an election is being stolen, a candidate is surging or official data cannot be trusted. Political markets differ from ordinary financial markets because the price itself can influence voter expectations, donor behavior, media framing and fraud narratives. Advocates for this side would argue that offshore platforms like Polymarket and even regulated venues like Kalshi cannot be considered neutral forecasting tools unless they solve the feedback loop between perception and price, disclose promotional relationships, reveal liquidity and concentration risks, and prevent insiders in campaigns, government agencies or election administration from trading on privileged information.

Round 1 — Opening arguments

Each advocate states its position and strongest evidence.

Advocate A
  • Prediction markets have demonstrated conditional forecasting value in historical election-market research, especially when markets are liquid and close to resolution.
  • Paid promotion, thin liquidity, and concentrated trading can distort political-market prices, but those risks are manageable through financial-market-style disclosure, surveillance, and enforcement.
  • Kalshi’s CFTC-regulated status and recent employment-verification, risk-scoring, whistleblower, and referral practices make it more governable than an unregulated narrative ecosystem.
  • Market odds should be reported with liquidity, bid-ask spread, concentration, and regulatory context rather than presented as direct measures of voter opinion.
Read the full argument

My position is that election prediction markets can remain useful information tools, but only if we stop treating their prices as neutral public opinion and instead treat them as financial-market signals with known market-structure risks. A prediction market is a venue where contracts pay out based on a future event; Kalshi is a U.S. CFTC-regulated event-contract exchange; Polymarket is a crypto-native prediction-market platform whose regulatory status has differed across jurisdictions; the CFTC is the U.S. derivatives regulator; market manipulation means trading or deceptive conduct intended to distort prices; liquidity means the depth and volume available to trade without moving price sharply.

The first point is empirical: prediction markets have a real forecasting pedigree, but it is conditional, not magical. The Iowa Electronic Markets literature found strong election forecasting performance, including evidence that vote-share markets outperformed many contemporaneous polls in the 1988-2004 presidential-election period. That does not prove every Kalshi or Polymarket contract is reliable, especially thin binary markets, but it rebuts the claim that markets are inherently just narrative machines. Their value comes from incentives, arbitrage, and aggregation of dispersed information, not from moral purity.

Second, recent scandals are an argument for stronger market governance, not for discarding the tool. The strongest opposing case is that paid creators used odds to imply election fraud or illegitimacy, and that a small number of traders can move thin markets. That concern is valid. But the policy question is comparative: are regulated market signals, with auditable trades and enforceable rules, more informative than polls, punditry, or viral social posts that often have weaker disclosure and no order book? Kalshi’s 2026 steps—risk scoring, employment verification for high-risk markets, whistleblower tools, and reported referrals for manipulation or insider-trading concerns—show the right direction. Polymarket’s public market-integrity page also says it bars fraud, manipulation, insider trading, self-dealing, front-running, and information misuse, while blockchain settlement makes many trades visible.

Third, transparency can expose concentration rather than conceal it. In 2024, large pro-Trump Polymarket positions drew scrutiny precisely because observers could identify outsized trading patterns and the platform investigated the accounts. That episode should make journalists cautious about reporting odds as “the electorate thinks X.” But it also shows why the better norm is contextualized reporting: odds should be accompanied by volume, bid-ask spread, largest-holder concentration, trader eligibility, and regulatory status. The same applies to insider risks in employment, economic-data, and election-administration markets: conflicted actors should be barred or disclosed, and suspicious trading referred for enforcement.

What is at stake is whether the public gets a noisy but disciplined real-time signal, or whether the field is ceded to offshore venues and social-media narratives. Properly regulated prediction markets are not neutral democracy meters. They are financial-information venues whose usefulness depends on surveillance, disclosure, liquidity, and enforcement.

Evidence cited by Advocate A (5) — as presented, not independently endorsed
Advocate B
  • Election prediction-market prices can become political inputs because campaigns, influencers, media, and fraud narratives use the odds as evidence of momentum or legitimacy.
  • Paid creators linked to Kalshi and Polymarket used or promoted market odds while spreading election-integrity claims after the 2026 Los Angeles mayoral election, prompting rule tightening.
  • Polymarket’s 2024 presidential-election trading showed significant concentration, including Bloomberg’s finding that 0.7% of accounts accounted for nearly half of purchase volume in the analyzed period.
  • Recent insider-trading allegations on Kalshi and Polymarket show that privileged information risk is concrete rather than hypothetical.
Read the full argument

Advocate A’s weakest point is the assumption that election-market harms are mainly ordinary market-structure defects that can be fixed by contextualized reporting and exchange surveillance. That framing is incomplete because the contested product is not just a price signal; it is a politically usable claim about legitimacy and momentum. A prediction market is a venue where contracts pay based on a future event; liquidity is the depth of trading available without large price impact; manipulation is conduct designed to distort prices or create false market signals. In politics, those signals can become campaign rhetoric, influencer content, donor cues, and fraud narratives before enforcement or arbitrage has time to correct them.

The first counter-evidence is the 2026 California episode. Reporting from NPR/KPBS and Semafor says paid creators tied to Kalshi and Polymarket cited market odds while spreading claims that Los Angeles mayoral results were being stolen or rigged, prompting the platforms to tighten affiliate rules. That does not prove the platforms intended election denial, but it directly rebuts A’s implicit separation between “market signal” and “viral social posts.” The market price was the rhetorical object being amplified. If a paid influencer can point to falling or rising odds as evidence that official vote counts are suspect, then the market is participating in political meaning-making, not merely measuring beliefs.

Second, concentration is not a theoretical edge case. Bloomberg’s analysis of 2024 Polymarket presidential trading found that 670 power traders, about 0.7% of accounts in the analyzed period, accounted for nearly half of purchase volume. Separate reporting identified one French trader controlling four accounts with roughly $28 million to $30 million in pro-Trump exposure, enough to draw scrutiny over whether a few actors were shaping the headline odds. Even if that trader was economically rational rather than manipulative, the public-facing effect is the same: a small set of wallets can produce a widely cited probability that looks like collective wisdom.

Third, insider risk is now documented enough that A’s “governance can solve it” claim should be treated as unproven, not established. AP reported Kalshi’s new employment verification and risk scoring after incidents including an alleged George Santos trade on his own State of the Union attendance, an Army soldier charged over classified-information trading on Polymarket, and earlier Kalshi insider cases. The CFTC’s 2022 Polymarket order also matters: Polymarket previously operated an unregistered event-contract facility, so “offshore transparency” is not equivalent to U.S. supervisory accountability.

The steel-man of A is that regulated markets may be better than rumor-only ecosystems. But the relevant burden is higher: before election odds are treated as public information infrastructure, venues should disclose paid-promotion relationships, trader concentration, order-book depth, affiliate violations, and insider restrictions in real time. Without that, the odds can manufacture a political narrative faster than they measure political reality.

Evidence cited by Advocate B (5) — as presented, not independently endorsed

Round 2 — Rebuttals

Each advocate concedes its opponent’s best point, then explains why its position still holds.

Advocate A
  • Election prediction-market odds can be politically misused, especially when paid influencers present price moves as evidence of fraud or legitimacy rather than as tradable beliefs.
  • Trader concentration on Polymarket in 2024 shows that headline odds should not be interpreted as mass public opinion without liquidity and concentration disclosures.
  • Kalshi’s reported employment screening, risk scoring, presumptive-insider bans, and enforcement referrals show that financial-market-style controls can address some manipulation and insider-risk problems.
  • The proper standard for election markets is not neutrality, but whether regulated, auditable prices with disclosures are more informative than opaque polls, punditry, and social-media narratives.
Read the full argument

B’s strongest point is that election-market prices are not inert forecasts once they enter political communication. The California reporting has genuine merit: paid creators did not merely say “I think the race is suspicious”; they cited Kalshi or Polymarket odds as evidence for claims about rigging or stolen elections, and that use can influence media narratives before any later market correction or enforcement action. Likewise, the Bloomberg finding that 670 Polymarket power traders, about 0.7% of analyzed accounts, accounted for nearly half of purchase volume is a real warning against presenting odds as mass public sentiment.

But that point narrows rather than defeats my position. It shows that prediction markets should not be advertised or reported as neutral public-opinion gauges. It does not show that they cannot remain useful financial-information venues. The relevant comparison is not between imperfect markets and an ideal democratic signal; it is between regulated, auditable prices and the existing alternatives: partisan influencers, selective internal polls, anonymous social-media claims, and pundit narratives that often lack any tradable accountability or transparent transaction record.

The practical implication is a disclosure-and-market-integrity regime. Paid promotion should be labeled and restricted, and affiliate contracts should prohibit creators from presenting market moves as proof of fraud or illegitimacy. Thin-liquidity contracts should display order-book depth, bid-ask spread, recent volume, open interest, and concentration warnings. Election contracts with high public-salience risk should have enhanced surveillance for wash trading, spoofing, coordinated position-building, and related-account trading. Conflicted traders with material nonpublic information—campaign staff, election administrators, platform employees, and government personnel in relevant markets—should be screened, barred, or referred for enforcement.

Recent evidence supports feasibility, not complacency. AP reports that Kalshi is adding employment-information collection for high-risk markets, risk scoring, presumptive-insider screening, and has made at least 20 referrals concerning manipulation or insider trading. Kalshi also says these tools have stopped at least 100 potential insider trades. Those facts do not prove perfect enforcement, but they demonstrate that the right governance levers exist. Polymarket’s own integrity policy prohibits fraud, manipulation, insider trading, self-dealing, front-running, and information misuse; the CFTC’s 2022 order against Polymarket also shows why jurisdiction matters, because offshore or nonregistered venues lack the same supervisory accountability unless brought into compliance.

B is right that concentration and paid amplification can manufacture narratives. My answer is that this is precisely why election odds should be treated like financial data: useful when accompanied by market-structure context, dangerous when stripped into a viral headline. The reform standard should be: no undisclosed paid promotion, real-time liquidity and concentration context, insider-access controls, public enforcement statistics, and clear journalistic labeling that odds are tradable beliefs, not election legitimacy measures.

Evidence cited by Advocate A (5) — as presented, not independently endorsed
Advocate B
  • A conceded that election prediction-market odds can become political inputs when paid influencers use them to support fraud or legitimacy narratives.
  • Kalshi’s new employment checks and risk scoring demonstrate the seriousness of insider-risk concerns but do not prove that these systems can prevent narrative manipulation in real time.
  • Polymarket’s 2024 presidential-election market showed material trader concentration, with Bloomberg reporting that 670 power traders represented about 0.7% of analyzed accounts but nearly half of purchase volume.
  • Election-market prices require a higher disclosure standard than ordinary punditry because their quantitative form can be presented as independent evidence of momentum, legitimacy, or fraud.
Read the full argument

A’s rebuttal concedes the central premise: election-market prices are not inert forecasts, and paid influencers did use Kalshi or Polymarket odds as evidence in election-fraud narratives. That concession matters because A’s remaining defense—treat them as “financial data” with better disclosure—does not resolve the feedback loop. The core issue is not simply whether a market price can contain information; it is whether a politically salient price can be pushed, promoted, and then cited as independent validation before voters, donors, journalists, or election administrators can distinguish genuine information from manufactured momentum.

A’s weakness is that he treats governance levers as proof of adequacy. AP’s reporting that Kalshi is adding employment checks, risk scoring, and referrals is evidence that insider and manipulation risks are serious enough to require emergency controls; it is not evidence that the controls work at election speed or against coordinated narrative operations. The same AP account describes concrete incidents: the George Santos investigation over a market tied to his own State of the Union attendance, a soldier charged over classified-information trading on Polymarket, and prior concerns that prompted Kalshi’s new system. A says at least some potential insider trades were stopped, but that still leaves the public without a reliable way to know, in real time, whether the headline price in a political market reflects broad information aggregation or a small number of conflicted actors.

A also concedes concentration, but understates its political implication. Bloomberg’s 2024 Polymarket analysis found that 670 power traders, about 0.7% of analyzed accounts, accounted for nearly half of purchase volume. In an ordinary financial asset, concentrated trading is a market-structure problem. In an election market, it becomes a public-narrative problem because the displayed probability can be reframed as “the market knows” or “the official result is unbelievable.” NPR/KPBS and Guardian reporting on the Los Angeles mayoral-election episode show exactly that mechanism: paid or affiliated creators tied market odds to claims of rigging or distrust, prompting platforms to tighten rules. The harm occurs during amplification, not only at final settlement.

Finally, A’s comparative argument—markets are better than pundits or anonymous social media—sets too low a bar. Prediction markets seek legitimacy precisely because prices look quantitative, disciplined, and impersonal. That makes undisclosed promotion, thin liquidity, trader concentration, and insider access more dangerous, not less. Until venues provide real-time concentration metrics, order-book depth, affiliate disclosures, market-integrity violations, and categorical bars on conflicted political and election-administration insiders, election odds should not be treated as neutral public information infrastructure. They may measure beliefs sometimes, but the evidence shows they can also manufacture politically useful reality.

Evidence cited by Advocate B (5) — as presented, not independently endorsed

Round 3 — Steel man & the crux

Each advocate argues the strongest version of the opposing case before answering: what evidence would change my mind?

Advocate A
  • B is correct that prediction-market odds can become political inputs when paid influencers use them to support fraud or legitimacy narratives.
  • Historical Iowa Electronic Markets research provides evidence that election prediction markets can contain forecasting information under some conditions, rather than being merely narrative devices.
  • Kalshi’s reported employment screening, risk scoring, insider restrictions, and enforcement referrals show that market-governance tools are feasible, though not automatically sufficient.
  • The proper reliability standard is whether odds are disclosed as regulated financial-information signals with liquidity and concentration context, not whether they are neutral measures of public opinion.
Read the full argument

B’s strongest argument is that election prediction markets are not merely passive forecasts once their odds are converted into campaign messaging, influencer posts, and legitimacy claims. The Los Angeles mayoral-election episode is serious evidence: KPBS/NPR reported that paid or sponsored posts used Kalshi and Polymarket odds to imply fraud, and Bloomberg’s 2024 Polymarket analysis shows that a small trader cohort can dominate purchase volume. If a price can be moved by concentrated capital and then cited as independent proof that an official count is suspicious, B is right that the public harm occurs before final settlement or later enforcement.

I still think my position is stronger because B’s evidence proves a need for financial-information governance, not that prediction markets are inherently unreliable or useless. The distinction matters. A market price is not a poll, a vote count, or a legitimacy certificate; it is a tradable probability generated under particular liquidity, concentration, eligibility, and surveillance conditions. The policy failure in the examples B cites was largely that audiences saw a decontextualized probability, sometimes amplified by paid creators, without real-time warnings about sponsorship, order-book depth, trader concentration, or the difference between betting odds and public sentiment. That is an argument for mandatory context, not abandonment.

There is also evidence that markets can have informational value when structured properly. Berg, Nelson, and Rietz’s Iowa Electronic Markets study found that presidential vote-share markets from 1988 through 2004 had meaningful forecasting performance relative to polls over longer horizons. That historical evidence does not automatically validate every thin Kalshi or Polymarket contract, but it prevents the categorical conclusion that election markets only manufacture narratives. Markets can aggregate dispersed information; the question is whether modern platforms disclose enough market structure for users and journalists to interpret the signal correctly.

The regulatory comparison also favors A. AP reports that Kalshi is adding employment-information collection for high-risk markets, risk scoring, presumptive-insider bans, and that it has made at least 20 referrals for manipulation or insider-trading concerns; it also says the new tools have stopped at least 100 potential insider trades. Those facts do not prove adequacy, but they show measurable governance mechanisms. The CFTC’s 2022 Polymarket order, by contrast, shows why jurisdiction and supervision matter: offshore or nonregistered venues are weaker public-information infrastructure unless brought under comparable controls.

So the defensible conclusion is conditional: election prediction markets should not be marketed as neutral public-opinion gauges. But if they carry sponsorship disclosures, liquidity and concentration metrics, conflicted-trader restrictions, surveillance, whistleblower channels, and enforcement transparency, they can remain useful real-time financial-information tools—often more auditable than punditry or viral claims.

Evidence cited by Advocate A (5) — as presented, not independently endorsed

What would change Advocate A’s mind

I would change my mind if a preregistered, multi-election study showed that even with real-time sponsorship, liquidity, concentration, and insider-risk disclosures, election-market odds systematically reduced public accuracy or trust more than comparable polls, models, and pundit forecasts.

Advocate B
  • Election prediction-market odds can become political inputs when influencers, campaigns, or media use them as evidence of momentum, fraud, or legitimacy.
  • The 2026 Los Angeles mayoral-election controversy shows that paid or affiliated creators used Kalshi and Polymarket odds in posts questioning election integrity, prompting platform rule changes.
  • Kalshi’s new employment checks and risk scoring show that insider-risk concerns are concrete, but they do not prove that platforms can stop politically consequential manipulation in real time.
  • Polymarket’s 2024 presidential-election market showed significant trader concentration, with Bloomberg reporting that 670 power traders accounted for nearly half of purchase volume in the analyzed period.
  • Election markets require stricter disclosure than ordinary punditry because quantitative odds can be mistaken for independent collective intelligence rather than market output shaped by liquidity, concentration, and promotion.
Read the full argument

A’s strongest argument is that prediction markets should be evaluated against the real alternative, not against an idealized politics with no misinformation. Properly regulated venues can create auditable, tradable records, and historical Iowa Electronic Markets research suggests election markets can contain useful forecasting information under some conditions. A is also right that banning or dismissing markets may simply push election betting toward less transparent offshore venues and leave the public with punditry, selective internal polls, and social-media rumor.

I still think B’s position is stronger because the most important question is not whether prediction markets ever forecast well; it is whether modern, high-salience political odds can be treated as neutral public information when the price itself becomes campaign material. The 2026 Los Angeles mayoral-election episode is not an abstract warning: KPBS/NPR reported that influencers cited Kalshi and Polymarket odds while spreading fraud claims, and Semafor reported that Kalshi asked paid influencers to delete posts that sowed doubts about the election. That evidence shows a closed loop: paid promotion draws attention to odds, odds are framed as evidence of public distrust or rigging, and that frame can affect political reality before settlement, arbitration, or enforcement can clarify anything.

A’s “financial data with context” solution is directionally sensible but not yet adequate. AP’s June 2026 reporting that Kalshi will collect employment information, risk-score high-risk markets, and respond to insider-trading incidents supports my argument more than A’s: these controls are being added because the existing system already produced enough insider and manipulation risk to require remediation. The same AP report cited alleged or charged incidents involving George Santos, a U.S. Army soldier, and other prediction-market cases. Those are not merely theoretical vulnerabilities; they are examples of privileged access becoming economically valuable.

Concentration also remains a crux. Bloomberg’s 2024 Polymarket analysis found that 670 power traders, about 0.7% of analyzed accounts, represented nearly half of purchase volume. A says disclosure can contextualize that; I agree in principle. But until platforms provide real-time concentration, order-book depth, affiliate status, and conflicted-trader restrictions at the moment odds are quoted, the public-facing number still carries unwarranted authority. In politics, unlike many financial markets, the price can become evidence in a legitimacy dispute.

My position is not that election markets must be abolished. It is that they cannot be treated as neutral forecasting infrastructure unless they first solve the perception-price feedback loop. That requires enforceable disclosure of paid promotion, live liquidity and concentration metrics, categorical insider-access rules for campaigns and election officials, and public reporting of violations. Without those safeguards, prediction markets may sometimes measure expectations, but they also provide a manipulable instrument for manufacturing momentum, suspicion, and legitimacy claims.

Evidence cited by Advocate B (5) — as presented, not independently endorsed

What would change Advocate B’s mind

I would need preregistered, multi-election evidence showing that, with real-time sponsorship, liquidity, concentration, and insider-risk disclosures, election-market odds improve public belief accuracy without measurably increasing false fraud claims, donor distortions, media misframing, or voter misperceptions compared with polls and forecasting models.

The Arbiter weighed this debate, verified the evidence, and took a position.

Read the article →