On determining probability forecasts from betting odds
Erik Štrumbelj · International Journal of Forecasting
Main finding
Probabilities from Shin's model were more accurate than basic normalisation or regression in the samples studied; bookmakers differed as sources of probability, and exchange prices were not always the best source, especially in smaller markets.
Limitations
Results depend on the bookmakers, sports and periods studied; later work finds the power method comparable to Shin and the gap varies with the number of outcomes.
Study details
- Authors
- Erik Štrumbelj
- Published
- 2014
- Venue
- International Journal of Forecasting 30(4), 934–943
- Status
- Published
- Evidence type
- Peer-reviewed empirical
- Evidence strength
- Moderate (our assessment of how much weight the finding can bear)
- Sport
- Several sports, including football
- Markets
- Bookmaker fixed odds and a betting exchange
- Method
- Compares methods of converting odds into probabilities (basic normalisation, regression-based methods and Shin's model) by the accuracy of the resulting forecasts.
- Sample
- Historical odds samples described in the paper.
- Replication
- The general ranking is broadly consistent with later comparisons such as Clarke, Kovalchik & Ingram (2017).
Read the original
doi:10.1016/j.ijforecast.2014.02.008 (opens in a new tab)
Our summary is an original paraphrase for education. Verification: Bibliographic record checked against publisher, index or author listings. Checked 2026-10-25.