Theses

The desk's register of calls — each thesis is either a trade on traditional markets or a Polymarket bet, with its entry, its target, its current state, and a full map of what supports it and what could break it. A Market thesis reads two ways: as a bet you can replicate on Polymarket, or as a position on the underlying environment — betting NO on "Iran hands over its uranium by May 31" is a way of saying the war isn't over, and that conviction runs through how you read the rest of the workspace.

Hawkline gives you research-backed trade ideas and convictions, not a portfolio to copy. You decide how much to risk, and when to act.

• You've read a brief or spotted a convergence zone and you're after the trade that follows: filter by MARKET for convictions backed by a Polymarket market, by TRADE for directional calls on an asset.
• To test your own conviction: "Supports the call" and "Cuts against" are laid out flat — you see immediately what is known, and where the argument can break.
• On a closed thesis, to understand how it played out — the closure note and the log give you the desk's read day by day.

Three screens. The top strip restates the state of the register (open, closed, markets/trades filter) and the last update. The list sits below, grouped in two sections — first the Markets theses (backed by a Polymarket market: replicable, but read first as a conviction about the environment), then the Trades theses (directional calls). Each row is a card summarizing the position. A click opens the detail page: chart, state of play in figures, the analyst's memo, the evidence ledger, what would change the view, and the log.

The top strip

The header filters

A thesis card

A header, a title — the question in one line — a body that depends on the type, and a footer summarizing the material available.

The body of a Trade thesis

The body of a Market thesis

The state chip — the card's center of gravity

To the right of the body, a stacked block condensing the current verdict. Three lines: an arrow ▲ ▼ ◆ followed by the raw move since entry (sign included), a state label, and a detail line.

Important, on the sign. The raw move is the one you read on the chart, with no correction for direction. A short winning 5% on a falling price shows −5.00% — but the chip stays green and reads WINNING. Color and label follow the call's direction; the number follows the price.

The detail line varies: N% to target on a target trade, Convergence X% → Y% on a Polymarket fade, Expects YES on a resolution, Resolved YES / Resolved NO once settled.

The card footer

N+ / N− evidence · N watch · N log — the material available to go further. + counts the pieces supporting the call, − those cutting against it. Open chart ↓ tells you the click opens the detail page.

The detail page

Opens full screen, and is reachable by direct URL through the ?thesis=… parameter — useful for sharing one specific call.

Workbench — chart + state of play

The Call — analyst memo

A prose block with a rule down its side: the desk's memo. It isn't a summary; it's the formulation the rest of the argument answers to.

The research panel

Three tabs, each with a counter. Only non-empty tabs are shown.

The detail page footer

Back to the list, and on a Market thesis, an Open on Polymarket ↗ link to the underlying market's page.

• A Market thesis you want to dig into: Markets opens the full Polymarket flow (top movers, related markets, whale tape) — useful for checking whether the call lines up with recent smart-money or not.
• A Trade thesis on an asset you follow: Trend matrix places the asset in its regime, and gives the factor-by-factor read (macro, positioning, risk, flow).
• On a settled Market thesis, Open on Polymarket ↗ at the foot of the detail page opens the original market.

• Horizon reached does not close the position. The banner says the originally expected window has elapsed. The thesis stays open until a closure has been written — a position running past its horizon without being closed does happen.
• A Market target call can overshoot the target. Edge captured then passes 100% — the position has over-shot, which is still a gain for the call.
• The evidence ledger grows; it doesn't get rewritten. A published exhibit stays exactly as written — the desk doesn't reword it and doesn't pull it. What the desk does do is add to it over the life of the position, in both columns: a closed thesis can carry exhibits dated well after it opened. It isn't a photograph of the open, it's the case as it accumulated.
• An exhibit's timestamp is the date of the fact it reports, not the date it was added to the ledger. So an exhibit filed mid-position can carry a date earlier than the open, and the numbering follows ledger order rather than chronology. To read an older thesis: the Evidence ledger gives you the accumulated case, the Log gives you the desk's dated commentary on what moved.