HFT Elite
Discovery & screeners
Discovery screeners built on persistent market data: 252-day IV rank, theta-harvest premium-selling candidates, mispricing vs external fair value, catalyst calendar by exposure, and a whale-activity tape.
Why discovery matters more than execution
Most active Polymarket traders lose more PnL to picking the wrong market than to picking the wrong order type. The native interface puts a single category dropdown between you and 4,000+ markets β there is no realized-volatility ranking, no concentration of recent prints, no view across multi-outcome markets. The Workstationβs screener suite is built on the same time-series infrastructure that powers the price chart, so you discover with the same data the chart shows.
IV rank
For each market we persist 1-minute OHLC bars and compute trailing realized volatility over a 252-trading-day-equivalent window. IV rank is the percentile of current realized vol within that window. A market at the 95th percentile is bouncing more than it has all year β thatβs where directional plays and OCO brackets pay best. A market at the 5th percentile is pinned β thatβs where you sell premium.
During Phase 1 the screener runs in preview mode (todayβs biggest movers from intraday hourly bars) until the aggregator banks enough history for the full 252-day percentile. Once the history is in, the column flips to true IV-rank without any URL or layout change.
Theta harvest
When a binary outcome trades at $0.95+ with thick depth and several days until resolution, the spread between market price and the eventual $1.00 payout is essentially premium that decays in your favour. The theta-harvest screener filters every aggregated market for close β₯ 0.95 in the last 30 minutes, returning the candidates a premium-seller would scan first. Pairs naturally with the portfolio Theta gauge on the Risk dashboard.
Spread screener
Tightest current bid-ask across recently snapshotted tokens. Useful for picking taker vs maker entries and for noticing when a previously-illiquid market has tightened up enough to size into. Uses the order_book_snapshots table the workstation populates whenever a viewer opens a market.
Catalyst calendar
Every position with a known resolution_date β populated from the Gamma API enrichment pass β surfaces here, sorted by next-to-resolve. Coloured by exposure size, so you immediately see which catalysts are about to settle the most book. The Polymarket equivalent of an options traderβs earnings calendar.
Whale activity
Recent trades from the public tape, filtered by minimum notional in USDC. Default cutoff is $1,000 over the last hour. Useful for spotting concentrated flow before it shows up in price, and for tracking specific addresses you suspect are running directional positions.
Mispricing vs external fair value (preview)
For sports markets, the workstation surfaces a mispricing screener that compares Polymarket implied probability against external sportsbook odds. For weather markets, it pulls forecast probabilities. For elections, it can layer rival prediction-exchange consensus. Initial sources ship internal-only while licensing is sorted; the screener architecture lets new fair-value adapters drop in without touching the UI.
One-click "send to ticket"
Every screener row deep-links to the workstation order-ticket for that token, with the price you saw in the screener pre-filled. The latency from "I see an idea" to "I have a stop placed" is measured in seconds, not minutes.
Screener catalogue
What each screener measures
| Screener | Inputs | Ranks by | Use case |
|---|---|---|---|
| IV rank | 252-day realized vol of 1m mid-price | Percentile within trailing window | Pick directional / OCO candidates in vol regimes |
| Theta harvest | Latest 1m close, depth threshold | Close β₯ 0.95 with thick depth | Sell premium against the residual gap to $1.00 |
| Tight spreads | Latest order_book_snapshot per token | (best_ask β best_bid) ASC | Identify markets you can size into without slippage |
| Catalyst calendar | Position resolution_date (Gamma) | Days until resolution Γ notional exposure | Know what is about to settle in your book |
| Whale activity | last_trades, configurable USDC threshold | Most-recent trades above threshold | Spot concentrated flow before it shows in price |
| Mispricing (preview) | Polymarket price vs external fair value | |p β fair| Γ notional | Cross-check sports / weather / election odds |
Sample output
Theta-harvest screener β annotated row
Real shape of a screener row, with what each column means and why it matters for premium selling. Numbers are illustrative.
token_id Β¦ close vol(USDC) seen
ββββββββββββββββββββββββββββββββββββββββββββββββ
0x4fβ¦ Β¦ 0.9612 $48,310 8s ago β high p, thick depth, fresh
β theta-harvest candidate
sell @ 0.96 β eat 4Β’ to parFurther reading
References
- Cboe Volatility Index (VIX) methodology β the canonical IV-percentile framework that IV-rank applies per-market.
- Sinclair, E. Volatility Trading, 2nd ed. β practitioner treatment of realized-vol percentile and premium-selling regimes.
- Manaster, S. & Mann, S. C. Life in the Pits: Competitive Market Making and Inventory Control β origin of the spread-tightness signal.
Related Workstation pages
FAQ
Common questions
What is IV rank for a Polymarket market?
The percentile rank of current realized volatility within a trailing 252-trading-day-equivalent window. The Workstation computes realized vol from persisted 1-minute mid-price bars; IV rank tells you whether a market is bouncing more or less than it normally does. High IV rank = directional / OCO plays. Low IV rank = premium-selling territory.
How does the theta-harvest screener work?
It filters every market the bar aggregator is tracking for a recent close at or above $0.95 with non-trivial volume. Those markets sit in the territory where you can sell premium against the residual gap to $1.00 with controlled risk. Pairs with the portfolio Theta gauge on the Risk dashboard.
What counts as a whale trade?
Default cutoff is $1,000 of notional in a single tape print over the last hour. The threshold is configurable on the API. The screener pulls from the last_trades table the workstation populates whenever a viewer opens a market.
How do you get the resolution date for the catalyst calendar?
Each positionβs resolution_date is enriched from the Gamma API. The same enrichment pass populates category and event_id, used by the Risk dashboardβs concentration heatmap. Backfill of historical positions runs when fill_tracker first touches the row.
What external sources feed the mispricing screener?
For sports, a sportsbook-odds adapter. For weather, a forecast-probability adapter. For elections, rival prediction-exchange consensus. All adapters ship internal-only initially while licensing is finalised; the architecture lets new sources drop in without UI changes.
Get this on your account
The Pro Workstation surface β and everything described on this page β ships on the HFT Elite tier ($149/month, 0.10% per-trade fee).
Upgrade to HFT Elite