← Back to scans

Will the margin of victory for TISZA in the 2026 Hungarian parliamentary election be between 15% and 18%? — TISZA, 15-18%

KXHUNGARYMOV-26APR12-TIS-P16 · Elections · 2026-04-17
6%
Agent
3%
Market Price
+3.0%
Edge
70%
Confidence
Volume: 48,060
Spread: 3.8c
Days to resolution: 360
Markets in event: 11
Final Rationale
The election has occurred and TISZA won, but the certified final margin appears to be ~12.6pp (below the 15% threshold), with only a preliminary 97.35% domestic count showing ~15.8pp. I accept the Devil's Advocate critique that I overweighted the probability of a non-certified count resolving the market — market convention strongly favors certified final results. The highly liquid market at 3% (13,214 contract depth, yes_bid at $0.00) reflects post-event consensus with access to the same certified results, and deserves substantial deference. I deviate slightly upward from the market (6% vs 3%) to account for residual data-verification risk on my own figures and small ambiguity about exactly which official figure (party-list vote share vs. seat-weighted) the market uses.
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 $t{=}0$.
$n = 38$ follow-ups
1. Planner Opus · Fermi decomposition + tool plan
Tools chosen: article_search web_search kalshi_data kalshi_orderbook wikipedia_lookup
Sub-questions (Fermi decomposition)
  1. Will TISZA win the 2026 Hungarian parliamentary election (i.e., finish first in vote share)? (w=0.25)
  2. If TISZA wins, will their margin of victory be at least 15 percentage points? (w=0.35)
  3. If TISZA wins with a margin of at least 15%, will their margin be strictly less than 18 percentage points? (w=0.40)
Planner reasoning
Today is April 17, 2026, and the Hungarian parliamentary election likely occurred on or around April 12-13, 2026 (close time is April 12, 2027, which is the resolution deadline, not the election date — but given Hungarian election cycles, the election was likely in early April 2026). The key task is to determine if actual results are already available. The market resolves YES if TISZA's margin of victory is in [15%, 18%). This is a bracket question within an election outcome. Reference class: In 2022, Fidesz won with ~54% vs. united opposition ~35%, a ~19pp margin. If TISZA reversed this dynamic, polling would be the guide. Pre-election polls from late 2025/early 2026 likely showed TISZA leading by varying margins. Causal mechanisms: Hungarian mixed electoral system (FPTP + list seats) means raw vote share differences matter for this resolution. TISZA's margin depends on: (1) whether Fidesz's support collapsed significantly, (2) whether smaller parties crossed the 5% threshold (splitting votes), (3) turnout dynamics. The specific 15-18% bracket is relatively narrow (~3pp wide). The distribution of possible margins, if TISZA wins at all, likely spans a wide range (say 5-30pp). If the election has already occurred (April 12 or nearby), actual results should be searchable. The resolution date is April 12, 2027 — likely awaiting official certification or this is a slow-resolution market. Temporal analysis: The election has almost certainly already occurred as of April 17, 2026. Web search and article search should return actual results. If TISZA won with a margin clearly outside 15-18%, this resolves straightforwardly. The Kalshi market price will be a strong signal of where informed traders believe the margin fell.
2. Research Parallel tool calls · legacy v2
ToolStatusTimeItemsSummary
article_search OK 0.3s -
web_search OK 42.2s -
kalshi_data OK 2.5s -
kalshi_orderbook OK 0.3s -
wikipedia_lookup OK 1.7s -
3. Evidence Synthesis Sonnet · 10 items · legacy v2
#StrengthCredibilityDirectionSourceClaimPriced In
1 STRONG 92 ↑ UP web_search TISZA Party led by Péter Magyar won the April 12, 2026 Hungarian parliamentary election in a landslide, defeating Fidesz–KDNP and ending the 16-year Orbán era. Yes
2 STRONG 90 ↑ UP wikipedia_lookup Wikipedia confirms TISZA won a two-thirds supermajority (136 of 199 seats) in the 2026 Hungarian parliamentary election, the largest mandate in Hungarian free election history. Yes
3 STRONG 88 ↑ UP kalshi_data The related Kalshi market on Viktor Orbán becoming PM is priced at 0.50%, confirming market consensus that TISZA/Magyar won the election decisively. Yes
4 STRONG 88 NEUTRAL web_search With near-final results (97.35% counted), TISZA received 53.6% and Fidesz 37.8%, yielding a margin of approximately 15.8 percentage points. With all ballots counted (including diaspora postal votes), TISZA took 52.1% and Fidesz 39.5%, yielding a margin of approximately 12.6 percentage points. Yes
5 STRONG 85 ↓ DOWN web_search The diaspora postal votes heavily skewed toward Fidesz (84.04%), pulling TISZA's final overall margin down significantly compared to early count figures — final figures show TISZA at 52.1% vs Fidesz at 39.5%, a ~12.6pp margin. Yes
6 STRONG 88 ↓ DOWN kalshi_data The TISZA 15-18% margin bracket market is priced at only 3%, indicating the market strongly believes the margin did NOT fall in the 15-18% range — likely because the final margin was below 15%. Yes
7 STRONG 90 ↓ DOWN kalshi_orderbook The orderbook shows yes_bid at $0.00 and no_bid at $0.01 with high liquidity and depth of 13,214 contracts, confirming the market is near-certain this bracket resolves NO. Yes
8 STRONG 85 ↓ DOWN web_search Based on near-final results (97.35% counted), the margin of ~15.8pp would have fallen within the 15-18% bracket, but final full-count results show a margin of ~12.6pp (52.1% - 39.5%), which is below 15% and therefore outside the bracket. Yes
9 STRONG 88 ↓ DOWN kalshi_data The 3% market price for the 15-18% bracket strongly implies the final margin was either below 15% (most likely) or above 18%, with market consensus clearly rejecting this bracket as the resolution outcome. Yes
10 MODERATE 80 ↓ DOWN web_search The Fidesz diaspora postal vote result of 84.04% significantly boosted Fidesz's final vote share, narrowing TISZA's overall margin from ~15.8pp (domestic-only-dominated early count) to ~12.6pp in the final count. Yes
Information Gaps
  • The exact official certified final vote percentages have not been clearly confirmed with a single authoritative primary source — there is some internal inconsistency in the web search data between the 97.35% count (53.6% vs 37.8%) and the final count (52.1% vs 39.5%), making the precise margin uncertain.
  • The question resolution may depend on how exactly 'margin of victory' is defined — whether it uses party-list vote share, overall vote share, or some other metric — and this is not clarified in the data.
  • Whether the resolution uses the preliminary count or the certified final count (including postal/diaspora votes) is unclear and could affect the outcome.
  • No official Hungarian National Election Office certified final results document was directly cited.
Key Uncertainties
  • The two vote share figures cited in the research (53.6% vs 37.8% = ~15.8pp margin at 97.35% counted; vs 52.1% vs 39.5% = ~12.6pp final) are in direct tension with each other — if the final certified margin is ~12.6pp, this is below the 15% threshold and the market resolves NO. If the relevant metric uses the near-final domestic count, the margin could be near 15.8pp, potentially within the bracket.
  • Diaspora postal votes counted later skewed heavily toward Fidesz, materially narrowing TISZA's final margin — the direction is clear (DOWN) but the exact magnitude is not precisely confirmed.
  • The Kalshi market at 3% (with $0.00 yes bid) is near-certain this resolves NO, likely reflecting that informed traders know the final margin is below 15%.
  • There is a question of whether the market resolution uses party-list vote share specifically, or some combined metric that might yield a different number.
Market Context
Price: 0.03 · 7d: -0.067 · 30d: -0.066
Volume: MODERATE · Depth: HIGH
Reliability: HIGH — high liquidity, 13,214 contract depth, yes_bid at $0.00 indicating near-certain NO resolution; market has converged strongly since election results became available
4. Independent Forecaster Opus · No market price · legacy v2 Did NOT see market price
Independent estimate: 18% (45% confidence)
base rate
15%
50% conf
evidence driven
22%
40% conf
contrarian
18%
45% conf
Spread
7pp
Low disagreement
sq1: Will TISZA win the 2026 Hungarian parliamentary election?
99%
Election has occurred. Multiple strong sources confirm TISZA won with a two-thirds supermajority (136/199 seats). This is essentially resolved YES.
sq2: If TISZA wins, will their margin be at least 15 percentage points?
35%
The evidence is in tension: near-final count (97.35%) showed 53.6% vs 37.8% = ~15.8pp, but final count with diaspora postal votes showed 52.1% vs 39.5% = ~12.6pp. The final certified result is what typically matters for market resolution. The final ~12.6pp is below 15%. However, there's uncertainty about which metric resolves the market and exact final numbers. Base probability weighted toward the final count being the resolution source: ~30-40%.
sq3: If TISZA wins with margin ≥15%, will margin be strictly less than 18pp?
75%
Conditional on the margin being at least 15%, the most likely scenario is that it's in the 15-16pp range (the 97.35% count figure of ~15.8pp). If the market resolves based on the near-final count showing 15.8pp, that falls squarely within [15, 18). The margin is unlikely to exceed 18pp given even the most TISZA-favorable count shows ~15.8pp. High probability ~75%.
Show full reasoning trace
base rate: For any 3pp-wide bracket in an election margin distribution, the prior is roughly 10-20% if centered near expectations, lower otherwise.
evidence updates: Final certified results show ~12.6pp margin (below the bracket), pushing probability down. Near-final count showed ~15.8pp (inside bracket), providing upward pressure only if that's the resolution metric. The ambiguity between these two figures is the crux.
combination method: Conditional chain, but dominated by which count source resolves the market. Most markets use official certified final results.
final: The final certified result (~12.6pp) is most likely the resolution source, placing the margin below the 15% threshold. There's meaningful residual probability (~15-20%) that either the near-final/domestic count is used, or that party-list specific numbers land differently, or that my data on final figures is slightly off. Settling at 15%.
ensemble: {'ensemble_count': 3, 'perspectives': {'base_rate': {'p_yes': 0.15, 'confidence': 0.5}, 'evidence_driven': {'p_yes': 0.22, 'confidence': 0.4}, 'contrarian': {'p_yes': 0.18, 'confidence': 0.45}}, 'spread': 0.07, 'individual_p_yes': {'base_rate': 0.15, 'evidence_driven': 0.22, 'contrarian': 0.18}}
5. Devil's Advocate Sonnet · legacy v2 Market price revealed
Asymmetric Evidence Overconfidence Reasoning Flaw
Challenges
  1. The forecaster identifies the final certified result as ~12.6pp (below the 15% threshold) yet assigns 35% probability to margin ≥15%. This seems too high given the forecaster's own assessment that the certified final count is 'most likely' the resolution source. If the final certified margin is confirmed at ~12.6pp, the only paths to YES are (a) resolution using a non-standard metric, or (b) data error. A 35% weight on these alternative paths appears inflated.
  2. The market is highly liquid (13,214 contract depth) with yes_bid at $0.00 — this is a post-event market where the result is essentially known to participants. The forecaster dismisses this signal without a compelling informational advantage. The market has access to the same certified results and has converged strongly (30d trend -0.066). This is a case where the market signal deserves substantial weight because it reflects post-resolution-event consensus.
  3. Sub-question 3's 75% conditional probability may be overconfident in the narrow 15-16pp scenario, but more importantly, the entire conditional chain hinges on sq2, which appears too high. The forecaster should more carefully distinguish: P(resolution uses preliminary count) × P(preliminary count lands in 15-18pp) versus P(resolution uses certified final) × P(certified lands in 15-18pp). The first path is the only realistic YES path and seems under-scrutinized.
  4. Information gap about which count resolves the market is being treated as symmetric uncertainty, but market convention strongly favors official certified results. This asymmetry should push probability lower than 18%.
Suggested adjustment: -10pp
The forecaster's own evidence points to a certified final margin of ~12.6pp, below the 15% threshold. Assigning 35% to sq2 overweights the probability that a non-certified count resolves the market. Given the election has occurred and the certified result is known, residual uncertainty should be dominated by data-verification risk (forecaster's own figures could be wrong) rather than resolution-source ambiguity. A more appropriate estimate is ~7-10%, reflecting meaningful residual uncertainty about the exact final figures without the inflated weight on preliminary-count resolution.
Market comparison: Forecaster: 18.3%, Market: 3.0%. The 15.3pp divergence is large. Normally post-event markets with high liquidity and known results deserve significant weight. However, the forecaster has legitimately flagged genuine information gaps about exact final figures and resolution methodology. Some divergence is justified, but 18% appears too high given the evidence points to a sub-15% certified margin. A forecast in the 7-10% range better balances residual uncertainty against the strong post-event market signal.
6. Aggregator Opus · final calibrated forecast
6% (70% confidence)
The election has occurred and TISZA won, but the certified final margin appears to be ~12.6pp (below the 15% threshold), with only a preliminary 97.35% domestic count showing ~15.8pp. I accept the Devil's Advocate critique that I overweighted the probability of a non-certified count resolving the market — market convention strongly favors certified final results. The highly liquid market at 3% (13,214 contract depth, yes_bid at $0.00) reflects post-event consensus with access to the same certified results, and deserves substantial deference. I deviate slightly upward from the market (6% vs 3%) to account for residual data-verification risk on my own figures and small ambiguity about exactly which official figure (party-list vote share vs. seat-weighted) the market uses.