Thousands of prices move across sportsbooks every day. These prices reflect models, risk management decisions, market sentiment, and new information entering the system. We believe markets are not simply predictions — they are mechanisms for aggregating information. SMR studies the markets themselves: how prices form, how they move, and how reliable they are.
SMR began where serious market research always begins — with a question worth answering. We understood how prices form under uncertainty, how information moves through participants with different models and different objectives, and how structural conditions determine whether a price can be trusted.
Sports betting markets caught our attention for reasons any serious market researcher would recognize. They are among the most active real-time pricing environments in the world. Nobody was studying them as markets. The entire field was studying them as prediction problems.
We built SMR to change that.
We chose the most demanding testing environment available. March Madness generates more market activity in three weeks than most sports generate in a full season. Games tip every hour. Lines move continuously. Consensus forms and breaks in real time.
If the structural signals were real, they would hold here. If they were noise, this environment would expose that.
The signals held. MLB enters the dataset this week. Additional sports follow over the coming months.
Sports betting markets are information markets. Thousands of prices move simultaneously across dozens of participants with different models, different objectives, and different information. Consensus forms. Disagreement appears. Prices become more or less reliable depending on the structural conditions in which they were formed. We study those conditions. Not who wins. Whether the price can be trusted.
The CBOE VIX measures implied volatility in equity markets — structural stress made quantifiable. The MSR VIX applies the same logic to sports betting markets. A five-regime index measuring structural pressure across tracked markets, updated every 15 minutes. Not a prediction. A reading of market condition.
In financial markets, regime classification describes the structural environment in which prices are forming — stable, transitional, or stressed. STRATA applies that framework to sports betting markets. Four-tier structural classification for every tracked market, updated continuously. The condition of the market, not the expected outcome.
In financial market microstructure, the bid-ask spread measures information asymmetry. When it widens, participants disagree about value and prices are less reliable. Cross-book dispersion is the sports market equivalent. When books diverge on price, the information aggregation process has not completed. The research confirms: higher dispersion, higher forecast error. Consistently. Without exception.
Sports betting markets are information markets.
Every day, thousands of prices move across sportsbooks as games approach. These prices reflect models, risk management decisions, capital flows, and new information entering the system. Participants with different objectives — sharp bettors, recreational money, sportsbook risk managers — all contribute to a price that is continuously being formed, tested, and revised.
This is not fundamentally different from how equity markets, options markets, or fixed income markets work. The mechanisms differ. The underlying process — information aggregating into a price under uncertainty — is the same.
Our background is financial markets. We understand how prices form under uncertainty. We understand information asymmetry, regime theory, the relationship between participant disagreement and price reliability. We understand what it means when a market is stressed versus stable — and what that implies about whether the price can be trusted.
We applied that framework here. Not because sports betting is a financial market — it is not. But because it behaves like one in the ways that matter for research. Prices aggregate information. Consensus forms and breaks. Disagreement appears when uncertainty rises. Structural conditions vary. Reliability is not constant.
We do not consult lineups. We do not read injury reports. We do not follow the news cycle. We have no view on which team is better. We watch how prices form. We measure the structural conditions around those prices. We test whether those conditions carry information about reliability. The data answers the question. We publish what it shows.
Each experiment runs continuously as new games settle. When the dataset grows, findings either hold or shift. Either outcome is a post.
SMR compares implied probabilities to actual outcomes across thousands of games. Markets are informative across most probability ranges. Two structural conditions — extreme favorites and high disagreement — produce systematic forecast error.
Implied probability and actual win rate align closely across the broad middle of the probability range. The market is a reliable forecasting engine in aggregate — in most structural conditions.
Markets in the highest disagreement group show materially higher forecast error than the lowest disagreement group. The relationship is monotonic across all five dispersion quintiles. No exceptions in the current dataset.
At the high-confidence extreme of the probability distribution, the market has been systematically more confident than outcomes support. The pattern has held consistently since the research program began.
Michigan vs UConn Monday 9 PM ET. Dispersion compressed to the lowest range we track after the bracket was set. Ten books in tight consensus. What the research dataset says about how games in these structural conditions tend to resolve.
The two structural patterns confirmed in the quantitative data. Duke's market arc through the tournament — what the market told us game by game through price alone. MLB enters the dataset this week.
Are there structural conditions under which sports betting markets are more or less reliable? We applied financial market microstructure methodology to find out. Two early observations and what the growing dataset is confirming.
Sports betting markets are among the most active real-time pricing environments in the world. SMR studies the markets themselves — how prices form, how they move, and how reliable they are under different structural conditions.
The research publishes daily. Access level determines depth — from structural labels and weekly summaries to the full dataset and live terminal.
Public posts cover market structure concepts, selected observations, and introductory charts. The goal is to introduce readers to how sports betting markets behave as information systems.