Kalshi Price Impact Analysis

This analysis explores the predictive power of Median Daily Prices on the final “Yes/No” outcome of KXFEDMENTION markets. By examining historical data, we identify key price thresholds and “traps” that traders should be aware of.

Dataset & Methodology

  • Ticker: KXFEDMENTION (Target Rate Probability on Fed Mentioning specific words).
  • Data Range: Daily price history for all events from Jan 2025 to Jan 2026. Which is the first year of the Kalshi Fed Mentioning market.
  • Metric: Median Price. We use the median of the daily “mean price” (or “close” if mean is unavailable) for each market to represent its sustained trading level.
  • Target: Final Result (Yes = 1, No = 0).

Key Analysis Findings

We employed three statistical approaches to understand the relationship:

  1. Linear Regression: A baseline model assuming a straight-line relationship which yielded a strong correlation ($r \approx 0.76$).
  2. Logistic Regression: A probabilistic “S-curve” model, which is statistically more appropriate for binary outcomes.
  3. Binned Empirical Win Rates: Grouping markets into 10-cent price buckets to observe the actual historical win rate.

Median Price vs Outcome Comparison

The “10-20 Cent Trap”

The most striking finding from the Binned Analysis is the non-linearity in the lower price ranges.

Price Range Win Rate Insight
0-10 cents 5.9% Low probability, as expected.
10-20 cents 1.4% CAUTION: Markets trading in this range actually performed worse than cheaper ones. This suggests a “trap” where prices are elevated slightly by speculation but rarely deliver.
20-40 cents ~25% A modest improvement, but still largely a losing bet.

The 50-Cent Inflection Point

Both the Logistic Model and the Binned Data confirm that 50 cents is the critical tipping point.

  • Logistic Threshold: The model predicts a >50% probability of “Yes” once the median price exceeds 50.46 cents.
  • Empirical Jump: The actual win rate jumps from 20% in the 40-50c range to 45-53% in the 50-70c range.

High Confidence Zone

Once a market sustains a median price above 80 cents, the outcome is nearly guaranteed.

  • 80-90 cents: 94.1% Win Rate.
  • 90-100 cents: 98.1% Win Rate.

TODO

  • Initial data collection and cleaning
  • Linear & Logistic Regression modeling
  • Binned Empirical Probability calculation
  • Monitor future events to validate the “10-20 cent trap”

Home

Last Updated: 2026-01-28

This site uses Just the Docs, a documentation theme for Jekyll.