Smart Money Intelligence
The Polymarket Smart Money Index tells you what the sharpest traders are doing right now.
The PSMI is a daily 0–100 index built by Polycopy that measures how active, how broadly profitable, and how capital-committed the top verified wallets on Polymarket are — measured against their own recent baseline. This page explains what it measures and how to read it. For today's number, see the live index.
Polymarket Smart Money Index
62
ACTIVE
Example reading — not live data
What is the PSMI?
A real-time pulse check on Polymarket's most successful traders.
Polymarket's prices tell you what the crowd believes. The PSMI tells you what the best performers within that crowd are actually doing with their money.
The index is computed from a cohort of roughly 1,000 parity-verified wallets — wallets whose profit and loss reconciles to their on-chain totals, with market makers and arbitrage bots excluded. Their behavior is distilled into four components — momentum, breadth, participation, and capital conviction — and combined into one 0–100 score published once per day.
The most important thing to understand: 50 is not the middle of an arbitrary scale, it is this cohort's own 90-day baseline. A reading of 50 means these traders are behaving exactly as they usually do. Above 50 is busier or more broadly profitable than their norm; below 50 is quieter. The index tells you what they are doing — not what any market is about to do.
Why it matters
Market prices are efficient. But they don't tell you who's trading.
A 65¢ contract means something different on a day when the platform's most consistently profitable traders are deeply engaged than on a day they have stepped back entirely. Prices are anonymous; the PSMI puts a name on the participation.
Market price alone
Tells you the crowd's consensus probability. Efficient but anonymous — you don't know if the latest volume is from whales or retail, conviction buys or hedges.
PSMI + market price
Adds who is in the market to what the market believes. You can see whether proven performers are engaged at all right now, and whether their results are broad or driven by a handful of positions.
The four components
Four components, one score.
The PSMI combines four weighted components (Momentum 30%, Breadth 25%, Participation 25%, Capital conviction 20%), each normalised so that 50 is the cohort's own 90-day baseline. Together they answer: how are these traders behaving relative to how they usually behave?
Performance momentum (30%)
Is recent profit unusual for them?
Takes the cohort's average daily profit over the past week and asks how unusual it is against their own 90-day distribution — a z-score, not a ratio. That distinction matters: a simple ratio of profit rates breaks when the sign flips, scoring a cohort that swings from losses into profits as zero. Above 50 means this week is running ahead of their normal range.
Breadth (25%)
Are the gains widely shared?
The share of cohort wallets that traded in the past week and finished it profitable. Half of them profitable reads as 50. This is the component that stops a couple of enormous winners from carrying the whole index: strong momentum with weak breadth means the profits are concentrated, which is a genuinely different situation from a cohort winning across the board.
Participation (25%)
How much are they trading?
The cohort's fills per day over the past week against their 90-day median, on a log scale so that a ratio of exactly 1.0 lands on 50 and doubling or halving sit symmetrically either side of it. Fill counts are used rather than volume because some warehouse volume measures count shares rather than dollars.
Capital conviction (20%)
Are they committing capital?
Net capital the cohort is putting into positions — buys minus sells in dollars — against their own 90-day distribution. Note what this deliberately is not: a buy/sell ratio. On Polymarket, buying NO is still a “buy”, so counting buys against sells tells you nothing about direction. Size of committed capital does carry information; which side it lands on does not. See the underlying large trades in the Polymarket whale tracker.
How it's calculated
From raw trades to a single number.
Each signal is independently normalized to 0–100, then combined into a weighted average. The result is a score that's comparable across time periods and market conditions.
Data warehouse
~1,000 wallet cohort
Parity-verified P&L
Daily fills & flows
MMs & bots excluded
Signal extraction
Momentum score
Breadth score
Participation score
Conviction score
Normalization
50 = own baseline
30 / 25 / 25 / 20
One daily close
PSMI
62
0–100
Each component is normalised against the cohort's own 90-day baseline, then combined using weights: momentum 30%, breadth 25%, participation 25%, capital conviction 20%. These are deliberately flatter than the earlier version, which put 40% on momentum alone. Profit is the noisiest of the four inputs — a handful of large settlements can move it sharply in a day — so leaning on it that heavily made the index jumpier than the behavior it was trying to describe. Spreading weight across four components means no single spike can push the score into an extreme zone without corroboration.
Reading the gauge
What the four zones mean.
The zones describe activity levels, not market direction. There is no bullish or bearish reading of the PSMI — prediction markets are thousands of separate binary questions, not one asset with a direction, so there is nothing for a cohort to be collectively bullish on.
80–100
Surging
Smart money is unusually active. This is one of the busiest, most broadly profitable stretches on record.
55–79
Active
Smart money is trading at a healthy clip, at or above its own baseline.
30–54
Watching
Smart money is engaged but cautious. Activity is running below its usual pace.
0–29
Quiet
Smart money is sitting out. Activity and conviction are well below this cohort’s own baseline.
The cut points are not round numbers, and that is deliberate. They are set from the historical distribution of the index so that each zone stays rare enough to mean something — SURGING is calibrated to appear in roughly the top decile of days. A threshold at a tidy 80 that never triggers describes nothing.
The data warehouse
Why it matters where the data comes from.
The PSMI is only as good as the data behind it. Polycopy's warehouse tracks the right wallets and the right signals.
~1,000
wallets in the index cohort, selected from a larger parity-verified pool
50+
market categories analyzed across politics, sports, crypto, and more
Daily
one canonical close per day, archived and citable by date
90d
trailing baseline every component is measured against
The wallets in the data warehouse are selected from the Polymarket leaderboard based on realized performance — not just volume or activity. This means the PSMI reflects the behavior of traders who have actually demonstrated skill, not just noisy activity.
The warehouse captures data that the public Polymarket API doesn't expose: niche-level P&L, win rates per category, trade sizing patterns, and temporal activity trends. This proprietary data is what makes the PSMI possible.
How to use it
Read it as context, not as a trigger.
It describes how engaged a verified cohort of traders currently is. It does not forecast outcomes and it does not tell you which trade to make.
Read the components, not just the score
An index of 60 built on high breadth is a different situation from a 60 built on one enormous winning position. The composite tells you something is unusual; the components tell you what.
Trend, not snapshot
A 55 that has climbed for three days is a different reading from a 55 that just fell from 75. The daily archive exists so you can check which one you are looking at.
Combine with Copy Score
The PSMI describes cohort-wide activity. Copy Score assesses individual trade quality. They answer different questions and neither substitutes for the other.
Practical example
You see a trade with a Copy Score of B (strong). You check the PSMI and it is at 72 with breadth also elevated — the verified cohort is trading well above its usual pace and gains are widely shared. That is a different backdrop from the same Copy Score on a day the index sits at 25 and the cohort is barely trading. The PSMI does not change the trade's quality or make it more likely to win; it describes the environment the trade is happening in.
PSMI vs market price
What the PSMI adds that prices don't.
A price tells you what the crowd believes about one question. The PSMI tells you how hard a verified cohort of proven traders is currently working, and whether their results are broad or concentrated — which prices cannot show you.
The cohort is far more active than usual with no obvious catalyst. Worth investigating which categories the activity is concentrated in.
Elevated activity around a scheduled event. Expected, and mostly useful as a comparison against how the cohort behaved around similar past events.
Platform volume is high while this cohort sits out. The activity is coming from somewhere other than these wallets.
Close to baseline. The cohort is behaving normally and the index is telling you nothing unusual is happening.
Key insight
The most interesting PSMI readings are divergences between the index and the rest of the platform — the cohort trading heavily while overall volume is flat, or sitting out a week when everyone else is piling in. Those gaps are where the index carries information a price cannot, because they say something about who is trading rather than what any single market believes. What a divergence does not do is imply the crowd is wrong or that a reversal is coming.
Limitations
What the PSMI cannot do.
Transparency about limitations is part of the design.
Not predictive
The PSMI describes what the cohort has done, not what any market will do. It has no established forecasting record and we make no claim that it has one. Even consistently profitable traders are frequently wrong.
Aggregate, not specific
A single number across thousands of simultaneous markets. It cannot tell you which markets the activity is in — the per-category sub-indexes exist for that.
Survivorship in the cohort
The cohort is selected on demonstrated profitability, so it is by construction a group that has already succeeded. It describes the behavior of winners, which is not the same as describing behavior that causes winning.
Daily, not intraday
One canonical value per day. A reading always refers to a specific date's close, which is what makes it citable — but it will not reflect something that happened this afternoon.
History is reconstructed
Readings before launch were computed retroactively. Where the historical cohort could not be rebuilt as it stood at the time, snapshots are flagged in the API and on the methodology page. Reconstructed history should be read as indicative, not as a live track record.
Not a trading recommendation
The PSMI is an indicator. It does not tell you to buy or sell anything, and nothing on this page is financial advice.
See today's reading
The live index carries the current score and zone, the full daily history, the component breakdown behind the number, and the published methodology.
Keep reading
Risk disclaimer
Trading carries real risk.
The PSMI and supporting signals are informational tools. They help you think more clearly about market conditions, but they do not guarantee performance.
Past performance of tracked wallets does not guarantee future results. Even the most successful traders experience losing streaks.
The PSMI is an aggregate indicator. Individual market outcomes may diverge significantly from the overall smart money trend.
Polycopy does not provide financial advice. Use the PSMI as context for your own decision-making, not as instruction.