Request Access

§ 00 Pre-trade intelligence

Ready to audityour next move?

You already know why you’re right. Blindspot finds what you haven’t looked at yet — the contradiction in your own reasoning, the trade everyone else is already in, the thing that breaks the position. Before the capital goes out.

Status Private betaAccess By invitationCoverage Equities · FX · Rates · Commodities
Exhibit A — thesis intake Auditing
Submitted thesis

I’m long the euro against the dollar for the next six weeks. The ECB is done cutting and the Fed still has room to ease, so the rate differential narrows in Europe’s favour. Everyone I follow sees the same divergence, and positioning data suggests the market hasn’t priced it yet.

AssumedDirection rests on a policy path neither bank has committed to.
Contradicts“ECB done cutting” and “Fed has room” both require the growth gap to move your way. Only one currently is.
Blindspot“Nobody’s priced it” and “everyone sees it” cannot both hold. Crowding is the risk, not the edge.

§ 01 The problem

Nobody loses money on the risk they wrote down.

Losing trades are rarely the result of bad analysis. They are the result of analysis that stopped one question early — and of research tools built to confirm rather than to challenge.

01

Research confirms; it rarely interrogates

You search for a view you already hold, and the internet obliges. Every source that agrees feels like evidence. The sources that would have changed your mind were never in the result set.

02

The bear case is written by someone who wants you wrong

So you discount it. A bear case is only useful when it is constructed as carefully as your own — from the same facts, with the same seriousness, by something that has no position.

03

Crowding is invisible from the inside

The most dangerous trade is the one that looks like original thinking and is actually consensus. By the time a view is obvious enough to feel safe, the payoff has usually gone.

04

Nobody defines what would make them wrong

Without a stated invalidation condition, a losing position is indistinguishable from an early one. That ambiguity is where small mistakes are held until they become large ones.

Risk is not what you got wrong. It is what you never put on the page.

Blindspot — operating principle

§ 02 Blindspot discovery

We look for the absence, not the argument.

Most tools grade the reasoning you supplied. Blindspot maps the shape of what you left out — the variables your thesis depends on but never names. Four classes of omission, found systematically.

TYPE I

Unexamined assumption

A load-bearing belief stated as fact. Your thesis cannot survive without it, and you never tested it — usually because it was true the last three times.

TYPE II

Internal contradiction

Two claims in the same thesis that cannot both be true. They coexist because each was reached at a different time, from a different source.

TYPE III

Absent variable

A driver of this asset that your reasoning never touches — positioning, funding, liquidity windows, correlation regime, the calendar.

TYPE IV

Consensus mistaken for edge

A view you believe is contrarian and that is in fact widely held. We check your thesis against what the market is already saying and already carrying.

§ 03 The core product

Thesis Stress Test

You give it four things: the asset, the direction, the timeframe, and the thesis in your own words. It returns a graded case file. We score the reasoning — never the trade. No signals, no price targets, no conviction ratings.

AssetEUR/USD
DirectionLONG
Timeframe6 WEEKS
Case refBS-4417
VerdictNEEDS WORK
“Policy divergence turns in Europe’s favour. The ECB has finished easing, the Fed hasn’t started in earnest, and the rate differential compresses from here.”
Thesis integrity
42/ 100
Needs work

Grades the reasoning, not the position. A high score is not a green light — it means the argument is complete, not correct.

Evidence support72 / 100
Adequate
Internal consistency41 / 100
3 conflicts
Driver coverage3 of 9 addressed
Partial
Differentiation vs consensus30 / 100
Crowded
Invalidation clarity15 / 100
Undefined
Evidence weight — same source set, both directionsNet: bear +24
Bull 38 Bear 62

The strongest version of your argument

Reconstructed from your thesis and the supporting evidence that actually exists — stated more forcefully than you stated it.

  • Policy path. If the easing cycles are genuinely out of phase, the differential compresses mechanically, without needing a growth surprise.
  • Flow asymmetry. A region running an external surplus has structural demand for its currency that reasserts itself once the rate penalty narrows.
  • Starting point. The position begins from a level that already embeds a wide differential — you are not paying for the move you expect.
Illustrative output — format demonstration. Scores grade the completeness of the reasoning, not the merit of the trade. Not research, advice, or a recommendation.

§ 04 Research engine

Built to disagree with you.

The engine runs the same way an investigator works a file: gather everything, establish what is actually known, find where the accounts conflict, then ask what is missing from all of them.

FilingsCentral bank minutesPositioning dataRates & curvesSell-side researchEarnings callsMacro releasesFinancial press
Stage 01

Decompose

Your thesis is broken into discrete claims. Each one is separated into what you asserted, what you assumed, and what you left implicit.

Stage 02

Corroborate

Every claim is checked against primary sources. Anything that cannot be sourced is marked as unsupported rather than quietly carried forward.

Stage 03

Adversarial pass

The engine builds the strongest opposing thesis from the same evidence base, with no knowledge of which side you took.

Stage 04

Gap analysis

Known drivers of the asset over your timeframe are compared against what your thesis addresses. The difference is the blindspot set.

Stage 05

File

Findings are written to a case file with sources attached, so a conclusion can always be traced back to what produced it.

§ 05 Knowledge map

Every thesis is a graph. Most are missing nodes.

Blindspot renders your reasoning as a structure: the claims you made, the links between them, and the drivers you never connected to. The holes are the point.

Policy divergenceECB terminal rateFed easing pathRate differentialSpot directionCONFLICTPositioningCarry costEvent calendarCASE BS-4417 — 5 CLAIMS MAPPED · 1 CONFLICT · 3 DRIVERS UNADDRESSED
Load-bearing claim Stated link Contradiction Unaddressed driver

§ 06 Weekly intelligence

What Traders Missed This Week

One briefing, every Sunday, to every member. Five fixed sections — the same five each week, so you can read the change rather than the noise.

What Traders Missed This WeekSpecimen issue — format only
Biggest blindspot
The variable nobody modelled. Each week we name the single driver that was absent from the largest share of theses submitted to the engine — and what it would have changed.
Most crowded trade
Where conviction has become consensus. The position held by the most members simultaneously, read against positioning data. Crowding is reported as a risk measure, never as a signal.
Most misunderstood narrative
The story that outran its evidence. A widely repeated market explanation traced back to its primary source — with the gap between what was found and what is now being said.
Key macro risk
What's on the calendar that isn't in the price. The scheduled event most likely to invalidate the largest number of open theses over the coming fortnight.
Institutional positioning
Where the large money actually moved. Disclosed flows and positioning shifts, reported against what the same institutions were saying publicly at the time.
Section format shown for illustration. Issues contain sourced analysis; nothing in the briefing is advice or a recommendation.

§ 07 Access

Invite-only. Admitted in batches.

Blindspot is in private beta. We open a small number of seats at a time so each case file gets a real review rather than a queue position.

  • Unlimited thesis stress tests across equities, FX, rates and commodities.
  • The full case file — bull, bear, blindspots, contradictions, market view, invalidation.
  • Sunday briefing: What Traders Missed This Week.
  • Direct line to the team while the product is being built. Your files shape it.
  • Founding members keep their rate when pricing is introduced.
Access requestForm 01-A
Optional, but it moves you up the list.

One email when your seat opens. No newsletter, no sharing your address. You can ask us to delete your request at any time.