← Back to scans

Will the 2028 presidential election occur? — In 2028

KXPRESELECTIONOCCUR-28 · Elections · 2026-04-16
95%
Agent
93%
Market Price
+2.1%
Edge
82%
Confidence
Volume: 49,348
Spread: 2.9c
Days to resolution: 959
Markets in event: 1
Final Rationale
The 59/59 historical base rate of U.S. presidential elections occurring on schedule is the strongest anchor, supporting a probability well above 90%. However, I ACCEPT the Devil's Advocate's critique that my ensemble showed zero spread (indicating insufficient stress-testing), that the combination method was asserted rather than derived, and that the 2.5-year horizon with compounding tail risks (authoritarian action, nuclear conflict, catastrophic events, cyber attacks) across ~959 days warrants more than 3% residual risk. I PARTIALLY ACCEPT Challenge 3: while no president has openly discussed cancelling elections while in office, the absence of any concrete legal/legislative mechanism and the strength of institutional checks (courts, Congress, states running elections, 22nd Amendment) still make cancellation extremely unlikely — but this is a genuinely novel condition deserving more than a 1.5% adjustment. I REJECT the implication that the market at 92.9% should pull me all the way down: the wide spread ($0.998) and low daily volume suggest this is an anchored market rather than one actively incorporating new information, and some discount likely reflects structural hedging behavior on prediction markets for distant events. My 95% estimate sits between the pure base-rate signal (~98%) and the market price (92.9%), reflecting genuine but small tail risks across multiple domains over 2.5 years.
Price vs. prediction
How the market moved after the agent's call. Solid blue line is the actual market price; dashed line is the agent's static forecast at $t{=}0$.
$n = 58$ follow-ups
1. Planner Opus · Fermi decomposition + tool plan
Tools chosen: kalshi_data kalshi_orderbook wikipedia_lookup article_search web_search code_execution
Sub-questions (Fermi decomposition)
  1. Will the U.S. constitutional and democratic order remain intact through 2028, preserving the electoral process? (w=0.40)
  2. Will there be no catastrophic event (war, pandemic, natural disaster, constitutional crisis) severe enough to cancel or indefinitely postpone a presidential election before November 2028? (w=0.35)
  3. Has any U.S. presidential election ever been cancelled or postponed in American history? (w=0.15)
  4. Is there currently credible political or legal activity that could result in cancellation or suspension of the 2028 election? (w=0.10)
Planner reasoning
This question asks about the occurrence of a scheduled constitutional event — the 2028 U.S. presidential election. The key insight is that NO U.S. presidential election has ever been cancelled or indefinitely postponed in American history, including during the Civil War, both World Wars, the Great Depression, and COVID-19. The reference class of ~59 presidential elections yields a 0% cancellation rate historically. The question resolves YES if 'the 2028 presidential election occurs' — this is a very low bar, as even a flawed, delayed, or disputed election would likely count as 'occurring.' The primary causal risks to assess are: (1) authoritarian consolidation sufficient to cancel elections outright — which would require not just executive desire but congressional, judicial, and military acquiescence; (2) catastrophic infrastructure collapse or extreme national emergency; (3) constitutional amendment or coup preventing elections. Temporal analysis: With 959 days until resolution (closing December 2028), the sequence of events needed for NO would be: within the next ~2.5 years, a sufficient concentration of political power to overcome constitutional constraints, the 22nd Amendment, and institutional resistance. The 2026 midterms (November 2026) would be an important indicator — if those occur normally, it substantially raises confidence that 2028 will also proceed. Any realistic path to cancellation would likely first manifest as mid-term election disruption. What makes this question unusual: it is essentially asking forecasters to price tail risks to American democracy. The market exists partly as a hedge/statement market, not because the probability is meaningfully above ~97-99%. The key differentiator from superficially similar questions (e.g., 'who will win in 2028') is that this is about occurrence, not outcome — the bar is extremely low.
2. Research Parallel tool calls · legacy v2
ToolStatusTimeItemsSummary
kalshi_data OK 2.2s -
kalshi_orderbook OK 0.2s -
wikipedia_lookup OK 2.2s -
article_search OK 0.2s -
web_search OK 72.5s -
code_execution OK 0.0s -
3. Evidence Synthesis Sonnet · 13 items · legacy v2
#StrengthCredibilityDirectionSourceClaimPriced In
1 STRONG 92 ↑ UP code_execution No U.S. presidential election has ever been cancelled or postponed in 59 elections spanning Civil War, both World Wars, Great Depression, and COVID-19, yielding a 0% historical cancellation rate. Yes
2 STRONG 75 ↑ UP kalshi_data Kalshi market prices the 2028 election occurring at 92.90%, with high liquidity (depth 52,613) and very stable pricing (only -0.10% change over both 7-day and 30-day windows). Yes
3 MODERATE 70 ↑ UP kalshi_data Multiple related Kalshi markets (2028 winner by candidate and party) are actively traded with significant volume, implying broad market consensus that the 2028 election will occur. Yes
4 STRONG 85 ↑ UP web_search Legal experts confirm the president has no constitutional authority to cancel or postpone U.S. presidential elections; elections are administered by states which have a constitutional obligation to cast electoral votes every four years. Yes
5 MODERATE 72 ↓ DOWN web_search Trump made an offhand 'wartime' comment in August 2025 suggesting elections could be cancelled if the U.S. is at war, but the U.S. Constitution contains no wartime election exception and the U.S. has held elections during every major conflict including the Civil War. Yes
6 WEAK 45 ↓ DOWN web_search Senior Democrats including governors Newsom and Pritzker have issued warnings that the 2028 election may not occur or be legitimate, citing fears of federal agents at polling places and executive overreach, though no specific cancellation mechanism has been identified. Yes
7 WEAK 40 ↑ UP article_search No articles from October 2025 to April 2026 were found describing concrete catastrophic events (war, pandemic, natural disaster, constitutional crisis) that have materialized as imminent threats to the 2028 election. No
8 STRONG 95 ↑ UP code_execution In 59 U.S. presidential elections, none has ever been cancelled or indefinitely postponed — a 0.00% historical cancellation rate across wars, economic crises, and pandemics. Yes
9 STRONG 95 ↑ UP wikipedia_lookup The 2020 U.S. presidential election proceeded on schedule despite the COVID-19 pandemic, with record voter turnout, confirming the most recent precedent of election resilience under crisis conditions. Yes
10 MODERATE 65 ↓ DOWN web_search Trump has repeatedly flirted with seeking a third term in 2028 (expressly forbidden by the 22nd Amendment) and declined to directly rebut critics claiming he wants to cancel future elections, but no concrete legal mechanism or legislative effort to cancel the 2028 election has been identified. Yes
11 WEAK 50 ↓ DOWN web_search In 2022, Trump claimed that voter fraud 'allows for the termination of all rules, regulations, and articles, even those found in the Constitution' — rhetoric that has not translated into any actual legal or legislative action to suspend elections. Yes
12 WEAK 40 ↑ UP article_search No articles from October 2025 to April 2026 were found documenting active legal, legislative, or executive efforts specifically aimed at cancelling or suspending the 2028 presidential election. No
13 MODERATE 60 ↑ UP kalshi_data The 'Will the America Party contest the 2028 election' market at 22% (down -23% over 30 days) suggests market participants are actively pricing the 2028 election as a live event that third parties might enter, implying strong consensus it will occur. Yes
Information Gaps
  • No data on 2026 midterm election status — whether November 2026 midterms proceed normally would be the strongest near-term indicator of 2028 election safety
  • No recent (post-October 2025) news articles found covering any concrete threats or legislative activity related to the 2028 election
  • No data on current state of U.S. involvement in active military conflicts that could be used to invoke wartime emergency arguments
  • No information on any Supreme Court rulings or pending cases that could affect the electoral process timeline
  • No data on whether Congress has taken any action on electoral date legislation that could affect 2028 scheduling
  • No polling data on institutional trust in elections or military/judicial willingness to enforce constitutional constraints on executive overreach
Key Uncertainties
  • Whether Trump's third-term ambitions translate into concrete institutional action — currently only rhetoric, but a meaningful tail risk
  • Whether the 2026 midterms proceed normally — this is the critical leading indicator for 2028
  • Whether a major catastrophic event (novel pandemic, nuclear incident, large-scale cyberattack on infrastructure) could emerge in the 2026-2028 window
  • Whether Democratic alarm-sounding about 2028 reflects genuine inside information or is primarily political rhetoric
  • Whether the Kalshi price of ~93% represents genuine informed market consensus or is artificially depressed by statement/hedge traders pricing political anxiety rather than base-rate probability
  • Whether institutional guardrails (courts, military, state governments, Congress) would hold against a determined executive attempt to interfere with the election process
Market Context
Price: 0.929 · 7d: -0.001 · 30d: -0.001
Volume: LOW — 348 contracts/day average, modest for a major political question · Depth: HIGH — 52,613 depth with spread of $0.998, though the wide spread suggests thin active trading
Reliability: MEDIUM — High depth but low daily volume and extremely stable pricing suggest limited new information is being incorporated; market may be anchored rather than actively price-discovering. The 92.9% price is meaningfully below the ~97-99% level suggested by pure historical base rates, possibly reflecting a statement/hedge market discount or genuine political anxiety premium.
4. Independent Forecaster Opus · No market price · legacy v2 Did NOT see market price
Independent estimate: 97% (94% confidence)
base rate
97%
95% conf
evidence driven
97%
92% conf
contrarian
97%
95% conf
Spread
0pp
Low disagreement
sq1: Will the U.S. constitutional and democratic order remain intact through 2028, preserving the electoral process?
97%
Base rate: The U.S. constitutional order has preserved elections through 59 consecutive cycles over 232 years, including during existential crises (Civil War, WWII, pandemic). That's a 100% success rate. Current conditions: While there is unusual rhetorical pressure from the executive branch (Trump's third-term musings, wartime comments), there is no concrete legislative, judicial, or executive action toward cancellation. The constitutional framework distributes election administration to 50 states, requiring an unprecedented level of coordinated institutional collapse. The judiciary, military, and state governments remain independent actors. I push down slightly (~3%) from near-certainty to account for the genuine uncertainty of a 2.5-year horizon and the unusual nature of current political rhetoric, but the structural barriers remain enormous. Final: 97%.
sq2: Will there be no catastrophic event severe enough to cancel or indefinitely postpone a presidential election before November 2028?
96%
Base rate: No catastrophic event has ever cancelled a U.S. presidential election in 59 cycles. The Civil War (600,000+ dead), the 1918 pandemic (675,000 American dead during election year 1918 for midterms), WWII, and COVID-19 all failed to cancel elections. Trump's wartime comment pushes down slightly (~1%), but the Constitution has no wartime exception and precedent is clear. Democratic warnings push down negligibly (<1%) as they cite no specific mechanism. The absence of any identified catastrophic event in the Oct 2025-Apr 2026 window pushes back up slightly. The 2.5-year horizon does create real tail risk from novel catastrophes (nuclear war, catastrophic cyberattack), but even these would likely not cancel an election — they'd more likely result in an irregular or disputed election, which would still count as 'occurring' under this question's resolution. Final: 96%.
sq3: Has any U.S. presidential election ever been cancelled or postponed in American history?
1%
This is a factual question about historical record. The answer is definitively no — 0 out of 59 elections have been cancelled. This is well-documented. The probability of YES is essentially 0, but I assign 0.01 to account for extremely unlikely definitional edge cases (e.g., some might argue the 1864 election was effectively cancelled in Confederate states, though the Union held its election). This sub-question seems oddly framed for combination — it's asking about history, not the future. Since the answer is NO, and this feeds into a weighted average where YES = election occurs, I interpret this as: 'Does the historical record support that elections always occur?' — in which case p_yes should be very high. Reinterpreting: the historical base rate of elections occurring is 59/59 = 100%. P_yes (elections always occur) ≈ 0.99.
sq4: Is there currently credible political or legal activity that could result in cancellation or suspension of the 2028 election?
8%
This question asks whether credible activity EXISTS that could cancel the election. Trump's third-term rhetoric is real but lacks any identified legal mechanism. No legislative efforts, no executive orders, no court cases. The 22nd Amendment is clear. No articles from Oct 2025-Apr 2026 document concrete efforts. However, the rhetoric is more aggressive than any previous president's, and 2.5 years is enough time for rhetoric to evolve into action. I set this at 8% — there's a small but real chance that some form of credible political activity emerges, though 'credible' is a high bar. Note: this sub-question asks about activity existing, not about it succeeding. Even if such activity emerges, the election would almost certainly still occur.
Show full reasoning trace
base rate: 59/59 U.S. presidential elections have occurred on schedule (100% rate). This is the strongest anchor. The base rate alone suggests >98% probability.
evidence updates: Trump's third-term rhetoric and wartime comments push down by ~1-2%. Democratic warnings push down by <0.5%. Absence of concrete legislative/legal/executive action toward cancellation pushes back up by ~0.5%. The 2.5-year horizon creates genuine tail risk from unknown catastrophes, pushing down ~1%. The decentralized nature of U.S. election administration (50 states) provides structural resilience that pushes up ~0.5%. Net adjustment from base rate: approximately -2 to -3%.
combination method: Weighted average of sub-questions, with sq4 inverted (since YES on sq4 = bad for main question). The sub-questions are highly correlated — if constitutional order holds (sq1), catastrophes don't cancel (sq2), and no credible cancellation activity exists (sq4), the election occurs. The binding constraint is sq1/sq2, which are both very high probability.
final: The 2028 U.S. presidential election has an approximately 97% probability of occurring. The 3% residual risk accounts for: (1) ~1.5% chance of unprecedented authoritarian action succeeding in cancelling the election despite constitutional barriers; (2) ~1% chance of catastrophic event so severe it prevents any election from occurring (not just disrupting it); (3) ~0.5% for truly unforeseeable scenarios. This is consistent with the 59/59 historical base rate adjusted modestly downward for current political uncertainty.
ensemble: {'ensemble_count': 3, 'perspectives': {'base_rate': {'p_yes': 0.97, 'confidence': 0.95}, 'evidence_driven': {'p_yes': 0.97, 'confidence': 0.92}, 'contrarian': {'p_yes': 0.97, 'confidence': 0.95}}, 'spread': 0.0, 'individual_p_yes': {'base_rate': 0.97, 'evidence_driven': 0.97, 'contrarian': 0.97}}
5. Devil's Advocate Sonnet · legacy v2 Market price revealed
Math Error Overconfidence Missing Info Never Happened Fallacy
Challenges
  1. The forecaster's sub-question decomposition has a structural issue: sq3 ('Has any U.S. presidential election ever been cancelled?') is a factual/historical question, not a forward-looking probability. The forecaster recognizes this awkwardness but then reinterprets it on the fly, which introduces confusion into the combination method. The 'combination' is never actually specified mathematically — the forecaster just says 'weighted average' without showing weights or the actual calculation, making the 97% figure essentially asserted rather than derived.
  2. The ensemble shows zero spread (all three perspectives at exactly 0.97), which suggests the forecaster isn't genuinely stress-testing alternative viewpoints. A real contrarian perspective should produce a meaningfully different number. When the contrarian gives the same answer as the base-rate approach, the ensemble is adding no value.
  3. The forecaster may be underweighting the unique structural conditions of 2025-2028. While the base rate of 59/59 is compelling, no previous president has openly discussed cancelling elections or seeking unconstitutional third terms while in office. The 'never happened' base rate is powerful but the forecaster should more explicitly address whether the current political environment represents a structural break from historical precedent rather than just applying a small downward adjustment.
  4. The 2.5-year time horizon deserves more weight. The forecaster assigns only ~1% to catastrophic events and ~1.5% to authoritarian action, but over 959 days, compounding tail risks across multiple domains (constitutional crisis, nuclear conflict, severe pandemic, unprecedented natural disaster, cyber attack on critical infrastructure) could reasonably sum to more than 3%. The forecaster's own information gaps are substantial — no data on military conflicts, no Supreme Court case information, no data on congressional activity — and these gaps should push toward more uncertainty, not less.
  5. The forecaster notes that the Kalshi market is at 92.9% but argues this might reflect a 'hedge market discount' or 'political anxiety premium.' This is a reasonable explanation, but the forecaster doesn't adequately consider that sophisticated market participants might have information or models that justify the lower price. The market has been extremely stable (essentially no movement in 30 days), suggesting participants have converged on ~93% with some confidence.
  6. The direction balance of evidence (9 UP, 4 DOWN) may create a false sense of one-sidedness. Several of the UP items are essentially the same fact restated (the 59/59 base rate appears at least 3 times as separate evidence items), while the DOWN items represent genuinely distinct risk factors. The evidence is less one-sided than the count suggests.
Suggested adjustment: -2pp
The forecaster's combination method is not mathematically rigorous, the ensemble provides no real diversity of views, and the information gaps over a 959-day horizon should produce more uncertainty than reflected. The 59/59 base rate is powerful but several evidence items are redundant restatements of this same fact. Reducing by 2% to ~95% better reflects the genuine uncertainty from a long time horizon, substantial information gaps, and structurally novel political conditions (sitting president discussing election cancellation and unconstitutional term extension).
Market comparison: Forecaster: 97%, Market: 92.9%. The 4.1% divergence is notable. The market price likely reflects both genuine political risk assessment and some structural market factors (hedging, risk aversion). A 95% forecast would still be above the market but more appropriately accounts for the forecaster's reasoning weaknesses: lack of rigorous probability combination, zero ensemble spread, and substantial information gaps over a 2.5-year horizon. The forecaster has no genuinely private information that would justify strong divergence from the market.
6. Aggregator Opus · final calibrated forecast
95% (82% confidence)
The 59/59 historical base rate of U.S. presidential elections occurring on schedule is the strongest anchor, supporting a probability well above 90%. However, I ACCEPT the Devil's Advocate's critique that my ensemble showed zero spread (indicating insufficient stress-testing), that the combination method was asserted rather than derived, and that the 2.5-year horizon with compounding tail risks (authoritarian action, nuclear conflict, catastrophic events, cyber attacks) across ~959 days warrants more than 3% residual risk. I PARTIALLY ACCEPT Challenge 3: while no president has openly discussed cancelling elections while in office, the absence of any concrete legal/legislative mechanism and the strength of institutional checks (courts, Congress, states running elections, 22nd Amendment) still make cancellation extremely unlikely — but this is a genuinely novel condition deserving more than a 1.5% adjustment. I REJECT the implication that the market at 92.9% should pull me all the way down: the wide spread ($0.998) and low daily volume suggest this is an anchored market rather than one actively incorporating new information, and some discount likely reflects structural hedging behavior on prediction markets for distant events. My 95% estimate sits between the pure base-rate signal (~98%) and the market price (92.9%), reflecting genuine but small tail risks across multiple domains over 2.5 years.