Numbers That Matter: The 10 Metrics That Actually Describe a Prediction Market

Prediction markets are now large enough to generate the kind of statistics that sound impressive in headlines. Annual volume reaches into the tens of billions. New event contracts are listed in huge numbers. Mainstream media increasingly cite market probabilities as if they were natural indicators of reality. For a fast-growing sector, those numbers are helpful. For a serious trader, operator, regulator, or partner, they are nowhere near enough.

The core problem is simple: activity metrics are not the same as quality metrics.

A platform can post large annual volume while many of its individual contracts remain thin, expensive to trade, vulnerable to ambiguity, or legally fragile. A market can have a clean interface and impressive growth while still hiding weak depth, wide spreads, poor exit conditions, overconcentration in a handful of events, and unresolved venue risk.

That is why a responsible standard needs better numbers. Not the ones that sound biggest, but the ones that tell you what kind of market you are actually dealing with.

The Wrong Metrics Are Usually the Loudest Ones

Most public commentary about prediction markets still leans on four familiar talking points: total platform volume, number of listed contracts, media visibility, and user growth. None of those are meaningless. But all of them are incomplete.

Total volume does not tell you whether the next contract you open is actually liquid. Number of listed contracts does not tell you whether those contracts are well designed. Media attention does not tell you whether the venue is resilient. User growth does not tell you whether ordinary traders are entering a high-quality market or simply a highly engaging one.

To understand a prediction market properly, you need metrics that describe execution quality, product integrity, concentration, settlement robustness, and venue durability.

Popular metric	Why people like it	Why it is incomplete
Total annual volume	It signals size and momentum	It does not tell you how liquid or tradable a specific market is right now
Number of listed contracts	It signals breadth and product expansion	It says nothing about contract quality, settlement design, or demand concentration
User growth	It signals adoption	It does not tell you whether users are entering robust markets or fragile ones
Media citations of market odds	It signals cultural legitimacy	It can exaggerate the quality of prices that may still rest on shallow books or concentrated flow

Metric 1: Live Bid-Ask Spread

If one number should sit near the top of any responsible framework, it is the live bid-ask spread. The spread is the gap between the highest standing buy price and the lowest standing sell price. In practical terms, it is the immediate cost of crossing the market.

Polymarket's own documentation explains that the displayed midpoint is the average of the best bid and best ask, and that this midpoint is what appears as the market's implied probability. It also notes that if the spread is wider than ten cents, the platform displays the last traded price instead of the midpoint. That single detail tells you how much the visible number can diverge from executable reality when the book is weak.

A market with a wide spread is not necessarily broken, but it is expensive. A market with a persistently wide spread is often telling you that the price discovery process is much weaker than the screen suggests.

Metric 2: Visible Depth Near the Best Prices

The second essential number is not just the spread, but the amount of resting size available near the best bid and best ask. A quote is not very meaningful if there is only trivial size behind it.

Polymarket's documentation exposes order-book data directly, including bids, asks, and last trade price. Kalshi's own educational materials explain that liquidity is what makes it easy to buy or sell larger volumes at competitive prices and that market makers provide the resting orders that make those trades possible. A responsible trader should therefore ask not only what the best price is, but how much actual size is available before the book starts moving away.

Thin visible depth means the market may look fine for one contract and become much worse for any serious order size.

Metric 3: Effective Slippage for Real Order Size

Spread tells you the cost of crossing the best quote. Slippage tells you the cost of crossing multiple price levels because your order is larger than the available liquidity at the top of book. This is one of the most underappreciated numbers in the category.

A serious market should be evaluated not only by quoted prices, but by the average execution quality for realistic order sizes. If the best ask is 0.55 but only tiny size exists there, a larger order may fill at 0.56, 0.58, and 0.60 as it walks the book. The trader did not buy at 0.55 in any meaningful sense. The trader bought at a weighted average that may be much worse.

Effective slippage is therefore more informative than a clean-looking top-line quote.

Metric 4: Time Since Last Meaningful Trade

Historical volume often hides dead or semi-dead markets. A contract may show impressive cumulative turnover while still having very little live participation now. That is why time since last meaningful trade matters.

If hours or days have passed since the last real execution near the displayed price, the market is not truly active in the way casual users might assume. It may still settle fine in the end, but from a trading-quality perspective it is weak. This metric is especially important because platforms and third-party dashboards often foreground total volume while downplaying live inactivity.

A contract can have historical attention and still be a ghost market in the present tense.

Metric 5: Concentration of Volume in the Top Markets

Headline volume is more informative once you know how concentrated it is. If a platform's activity is dominated by a handful of election contracts, war contracts, or major sports events, then the average market on the venue may be much weaker than the platform-level number suggests.

This is one of the biggest reasons why the sector's top-line metrics can be misleading. Total volume can rise quickly while real tradable quality remains highly uneven across the market menu. A responsible platform should know what share of volume sits in the top 1 percent, 5 percent, or 10 percent of listed contracts. A responsible observer should ask for that data before assuming the venue is broadly deep.

Concentration is not automatically bad, but ignorance about concentration is.

Metric 6: Share of Broad Outcome Markets Versus Fragile Micro Markets

Not all contract categories deserve the same confidence. One of the most useful quality metrics is therefore category mix.

What percentage of activity comes from broad, aggregate outcome markets that depend on multiple participants or longer horizons? What percentage comes from narrow contracts tied to single actors, injuries, officiating, tactical incidents, or other structurally fragile triggers? This matters because the CFTC's own March 2026 advisory distinguishes between sports-related contracts tied to injuries, altercations, and officiating actions by a single person or small group, which may create heightened manipulation or price-distortion risk, and broader outcome markets based on aggregate performance over longer periods, which staff describes as more consistent with core regulatory principles.

In other words, category mix is not just a content question. It is a market-quality question.

Metric 7: Settlement Source Quality

Execution quality is only half the story. Settlement quality matters just as much. A contract that trades actively but resolves through weak, ambiguous, or fragile sourcing is still a low-quality instrument.

A responsible platform should therefore be judged by how often its markets rely on strong primary sources, how clearly fallback hierarchies are defined, and how many contracts still depend on vague wording or vulnerable single-source interpretation. This is especially important because the CFTC has now explicitly opened questions around settlement disputes and how event contracts should resolve when the underlying event is contested or unclear.

A market that cannot explain how truth becomes payout is not as robust as its price chart suggests.

Metric 8: Dispute and Escalation Rate

Another number that deserves much more attention is how often markets become contested after apparent resolution. If a meaningful share of contracts move into proposal disputes, escalation processes, or delayed finalization, that tells you something important about product quality and settlement clarity.

Polymarket's documented process shows that a disputed resolution can move from proposal to challenge to renewed proposal and even to UMA's Data Verification Mechanism, with rare cases resolving as Unknown/50-50. That is useful transparency. But it also highlights why dispute rate matters as a metric. If too many markets end up in procedural conflict, the venue may be producing more ambiguity than its users realize at entry.

A low dispute rate does not prove perfection, but a high one is hard to ignore.

Metric 9: Jurisdiction and Access Footprint

This is where venue risk enters the scoreboard. A prediction market is not only defined by what it lists. It is also defined by where and under which entity it can offer those products.

Polymarket's own materials already show how meaningful this is: Polymarket US operates under a CFTC-regulated structure, while the international platform is separate and not CFTC-regulated, and the international product currently lists multiple blocked countries plus close-only treatment for Poland. Kalshi shows the other version of the same issue, namely that even a federally regulated venue can still face direct state conflict over sports contracts in places like Massachusetts, Nevada, and Arizona.

That is why the jurisdiction footprint of a platform should be treated as a real metric. A market with expanding legal fragmentation is not the same kind of venue as one with stable access conditions.

Metric 10: Cash-Out Realism

One of the simplest and most useful quality metrics is also one of the most practical: how realistic is it that users can actually exit positions at prices near what the interface implies?

Kalshi's own help center addresses this directly. If a market has low liquidity, there may not be anyone willing to buy your position at the current price. The user may need to place a limit order and wait, lower the asking price, or simply hold until settlement. That means any serious venue-quality framework should ask whether marks are close to actionable exit conditions or whether the displayed price overstates how easy it is to convert a position back into cash.

A market that is easy to enter but hard to exit cleanly is not a high-quality market just because it looks active on the front end.

The Ten Metrics in One View

Metric	What it reveals	Why it matters
Live bid-ask spread	Immediate execution friction	Shows how expensive it is to cross the market now
Visible depth near best prices	How much real size sits behind quotes	Separates cosmetic quotes from real tradability
Effective slippage	Actual execution quality for realistic order size	Exposes hidden cost beyond the spread
Time since last meaningful trade	Live market activity versus stale volume	Helps identify ghost markets
Top-market concentration	How much volume is clustered in a few contracts	Shows whether platform-level volume is broadly representative
Category mix	Share of broad markets versus fragile micro markets	Signals structural quality and manipulation exposure
Settlement source quality	How robust the truth-to-payout mechanism is	Shows whether a contract is reliable beyond trading activity
Dispute and escalation rate	How often markets become contested after apparent resolution	Measures hidden ambiguity in contract design
Jurisdiction footprint	Legal stability and access continuity	Turns venue risk into a measurable part of quality
Cash-out realism	How close displayed prices are to actionable exits	Protects users from mistaking marks for liquidity

What This Means for Platforms

If prediction markets want to be treated as serious market infrastructure, they should get more comfortable publishing and discussing the quality metrics that matter, not only the growth metrics that market well. The strongest platforms in the next phase will not be the ones with the loudest volume graphics. They will be the ones that can explain how deep their books are, how concentrated their liquidity is, how reliable their settlement framework is, and how stable their jurisdictional structure remains under scrutiny.

That kind of transparency would not weaken the category. It would make it easier to trust.

What This Means for Traders

For traders, the lesson is even simpler. Do not judge a market by its headline popularity alone. Ask what the live spread is. Ask how much size is visible. Ask whether the market is broad or fragile. Ask how it resolves. Ask whether you can realistically exit. Ask whether the venue itself is stable in your jurisdiction.

These questions are not advanced or optional. They are the baseline for treating a prediction market like a real instrument instead of a probability-themed interface.

The Predict Responsibly View

Prediction markets have reached the point where they need better scorecards. Total volume and brand excitement were useful signals in the novelty phase. They are not enough in the maturity phase.

A responsible standard has to ask harder questions. How much of the volume is actually tradable? How much of the market menu is structurally sound? How many contracts resolve cleanly? How many rely on weak design? How much venue risk is hidden below the interface?

Those are the numbers that matter because they describe not only activity, but quality.

The Bottom Line

The real test of a prediction market is not how loudly it can announce growth. The real test is whether it can show that the growth sits on top of fair execution, resilient settlement, defensible market design, and stable venue architecture.

That is why the ten metrics in this article matter more than the usual headline numbers. They move the conversation from spectacle to substance.

And for a category that increasingly wants to be trusted, that shift is overdue.