Beta Notice: Proximarket is in beta testing. All credits have no real monetary value and are for testing purposes only.

How It Works

Open-ended text prediction markets with deterministic, objective payouts

How Payouts Are Determined

Proximarket uses semantic embedding models to mathematically measure how close your prediction was to the actual outcome. The closer your prediction, the larger your share of the total prize pool. Here's how it works:

Similarity Scoring

When a market resolves, the system compares your prediction text to the actual outcome using semantic embeddings. The cosine similarity is then squared to amplify the difference between accurate and inaccurate predictions, creating the final similarity score between 0 and 1.

similarity_score = (cosine_similarity(your_prediction, actual_outcome))²

Points Calculation

Your points are calculated by simply multiplying your similarity score by your wager amount. This straightforward calculation determines your share of the total prize pool.

your_points = similarity_score × your_wager

Proportional Payouts

The total prize pool is distributed based on your share of all points earned. Your final payout depends on both your prediction accuracy and your wager amount.

your_payout = floor((your_points ÷ total_points) × total_prize_pool)

Example: Google Trends Prediction Market

Market Question:

"What will be the top trending Google search in the US on December 15, 2025?"

Actual Result: "iPhone 15"

User Predictions:

Alice: "iPhone"Wager: 20 credits
Bob: "technology"Wager: 30 credits
Carol: "very"Wager: 50 credits

Similarity Scores:

Alice: High similarity (iPhone matches iPhone 15)Score: (0.78)² = 0.60
Bob: Moderate similarity (technology relates to iPhone)Score: (0.63)² = 0.39
Carol: Very low similarity (unrelated adverb)Score: (0.57)² = 0.33

Points Calculation:

Alice: 0.60 × 20 = 12.03 points

Bob: 0.39 × 30 = 11.75 points

Carol: 0.33 × 50 = 16.29 points

Total Points: 40.06

Total Prize Pool: 100 credits

Final Payouts (rounded down to nearest cent):

Alice:(12.03 ÷ 40.06) × 100 = $30.02
Bob:(11.75 ÷ 40.06) × 100 = $29.31
Carol:(16.29 ÷ 40.06) × 100 = $40.65
Total paid out: $99.98 (only $0.02 remains due to rounding down)

Key insight: Alice made a profit of $10.02 (won $30.02 from a $20 wager) because her prediction was accurate. Bob lost $0.69 and Carol lost $9.35 because their predictions were less accurate than their wager amounts justified. Carol's large wager amplified her loss - bigger bets mean bigger potential gains but also bigger potential losses.

Key Points

  • Objective Scoring: Embedding models provide deterministic, semantic similarity measurement
  • Accuracy Rewarded: Cosine similarity is squared to create the final similarity score, significantly favoring precise predictions
  • Proportional Payouts: Your payout depends on your share of total points earned
  • Strategic Balance: Higher wagers amplify your points when you're accurate

Note: The similarity scoring formula is subject to change and improvement as we conduct ongoing research into optimal prediction market mechanics. We may adjust the mathematical approach to better reward accuracy and create fairer market outcomes.