Historical Price Series and Their Peculiarities
Event contract history is not like equity history. Fixed bounds, a settlement point, irregular trading and a survivorship problem built into the data.
A price history from an event market looks like a financial time series and behaves differently in ways that break standard tooling.
Bounded
Prices live between zero and one hundred. There is no drift, no compounding, no log return that behaves sensibly near the boundaries.
Percentage changes are misleading: a move from 2 to 4 is a 100% change and two cents. A move from 50 to 52 is a 4% change and two cents. Most of the time the cents are what matter. Percentage points versus per cent.
It terminates
Every series ends at zero or one hundred. That endpoint is not a price; it is an outcome.
Including settlement in a volatility calculation produces a meaningless number, because the final move is a jump to a boundary that was always going to happen. Treat the settlement point as a separate object from the price path.
Irregular and often empty
Most markets do not trade most of the time. A daily series will contain long runs of unchanged values that represent nothing happening rather than a stable view.
Forward-filling makes this invisible, which is exactly wrong. Keep the gaps and keep a field recording how old each observation is. Mid price versus last trade.
Bid-ask bounce
In an active market with a two-cent spread, consecutive trades alternate between the bid and the offer. The last-trade series oscillates continuously while the market is unchanged.
Compute volatility from that series and you will overstate it substantially. Use mid-quote series for anything statistical.
The survivorship problem
Series are usually collected for markets that became interesting. The ones that never got liquid, got voided, or resolved ambiguously are underrepresented, and they are disproportionately the cases you most need in any risk analysis.
The fix is to record from listing, not from when you noticed. There is no way to reconstruct this later. Backtesting honestly.
What to store
Mid, bid, offer, and sizes, each with the venue's timestamp.
Last trade and its own timestamp, separately.
A staleness field derived at write time, so later analysis does not have to recompute it.
The final outcome, once known, joined to the series.
Resolution criteria as they were, not as later amended.
What it is good for
Calibration. Bucket prices, check realised frequencies, plot the curve. The single most valuable thing a history enables. Calibration curves.
Convergence studies. How does a price behave as settlement approaches, by category and by liquidity band.
Cross-venue relationship over time. Whether a gap is structural or fleeting cannot be answered from a snapshot. Persistent versus temporary gaps.
Event studies. How far and how fast prices move on a known release. Macro release markets.
What it is not good for
Anything that assumes continuous trading, unbounded prices or regular sampling. Most standard time series machinery assumes at least one of those.
Next
Educational content only. Nothing here is financial advice, a solicitation, or an offer to trade. Prediction markets can move against you and a position can lose its entire value. Where this guide describes a difference between venues it describes a theoretical cross-market edge in observed prices, which is not a profit: fees, spreads, capital lockup, transfer time and differing settlement rules all come out of it first.
Keep reading
Data and Automation · 3 min readWorking With Public Prediction Market APIsWhat public endpoints give you across venues, the fields that matter, and the habits that keep a data pipeline working and welcome.
Data and Automation · 3 min readStoring Market Time Series Without Drowning in DataMost of what a naive collector records says nothing happened. Sampling, change-only writes and sensible retention keep a useful history small.
Data and Automation · 3 min readBacktesting a Market Signal HonestlyAlmost every impressive prediction market backtest is built on prices nobody could have traded. The specific errors, and how to build one that tells the truth.
What Markets Price · 3 min readCrypto Price Markets and Why They Are Harder Than They LookThe underlying trades continuously on a dozen venues. That makes the settlement source, the timestamp and the reference price the whole ballgame.