Loading this week's brief...
The research desk

Every name, screened and filed.

A rules-based screen over the investable universe. Each name gets a label — Research, Monitor, Pass — from price, volume and structure. The label is a sorting heuristic, not a forecast: measured over 21-day forward returns it shows no edge over the benchmark. Click any row for the company behind it.

Market regime—
—
#TickerStrengthVolRiskLabel
← Back to screener
Academy

How to read the goblin.

The theory behind the tool — what the verdicts mean, how a name is weighed, and how to use signals without turning them into noise.

The idea in one line

Most of the market is noise. The goblin scans the whole universe on the same evidence — price behaviour, volume, structure, risk — and files one plain verdict per name so you spend your attention where it earns its keep. It never tells you to buy or sell; it tells you where the work is.

The three verdicts

RESEARCH — strength, volume and structure line up. Worth a closer dig. A starting point for your own work, not a recommendation.

MONITOR — the shape is forming but the trigger hasn't fired. Parked on the watch list with the exact condition that would move it up.

PASS — hype without follow-through, broken structure, or risk that swamps the reward. Filtered out before it costs you attention.

On any given day the vast majority of names are PASS — that is the point. A scanner that flags everything flags nothing.

The appraisal gem

Every name gets a five-axis gem. The wider the shape, the higher the overall appraisal score — and it doubles as a check on the verdict: a name can carry a verdict while its gem is pinched and low-quality.

Key concepts

Relative strength (RS) measures a stock against the benchmark, not against itself — a name up 5% while the market is up 10% is weak, not strong. Volume confirmation asks whether a breakout had real participation behind it or was thin. Extension is how far price has stretched above its intermediate trend — the further extended, the worse the reward-to-risk, which is why conviction is penalised, not rewarded, for it.

Reading the chart

Each appraisal page shows candlesticks over the available sessions with SMA20 / SMA50 / SMA200 overlays, a dated axis, and a hover readout. Green candles closed up, red closed down. The moving averages show trend and the "stack" of the averages (short above long) is a quick read on momentum.

How to use it responsibly

Treat a verdict as the start of your research, never the end. Read the goblin's note and the fundamentals, form your own view, size positions to your own risk, and remember the disclaimer isn't boilerplate: this is a research tool, not advice. The edge is discipline — letting a transparent, repeatable process filter noise so you make fewer, better-considered decisions.

For research and education. Not investment advice. Markets carry risk, including loss of capital.

Method

How the goblin appraises.

No predictions, no price targets. Every name is weighed on the same evidence — price behaviour, volume, structure, risk — and filed under one of three verdicts.

Research

Strength, volume and structure line up and the setup is worth a closer dig. The goblin flags it for work — not a recommendation, a starting point for your own.

Monitor

The shape is forming but the trigger hasn’t fired. Parked on the watch list with the exact condition that would move it up.

Pass

Hype without follow-through, broken structure, or risk that swamps the reward. Filtered out before it costs you attention.

The appraisal gem

Each name gets a five-axis gem. The wider the shape, the higher the overall appraisal score — a quick read on why a verdict landed, and a check on it: a name can carry a verdict while the gem shows a pinched, low-quality shape.

Scores are appraisals of price, volume and structure. Research inputs, not advice.

Track record

The goblin keeps receipts.

Forward returns are measured against QQQ, by verdict cohort. If the model trails the benchmark, this page says so plainly.

Loading receipts…
Mistakes & lessons

The goblin studies what failed.

Failed signals are evidence too. This page turns negative excess returns and written lessons into a visible repair loop.

Loading lessons…
EXPERIMENT

Momentum study

The signal-based allocator, kept running for comparison. It has produced no funded positions in recent months. Not the plan that gets acted on.

Loading plan…
Monthly fundamentals

Allocation

Rules-based, USD-only, monthly. No predictive edge is claimed.

Loading quality plan...
Actual holdings

Portfolio

Average-cost accounting. Prices from the engine's daily cache, not live quotes.

P&L is decomposed into price and currency effects; the price move is valued at the pooled entry FX rate (average USD cost / average native cost), and the currency component carries the interaction and cent-rounding residual.

Loading holdings...
About

While everyone’s shouting, the goblin’s counting.

Market Goblin is a research and signal tool for investors who’d rather weigh evidence than chase noise.

Most market tools sell certainty — price targets pulled from the air, hot takes, and screens that all surface the same crowded names. Market Goblin does the unglamorous thing: it digs through the whole universe, weighs each name on the same evidence, and files a plain verdict. Research, Monitor, or Pass.

The verdicts are deliberately not buy/sell language. They mark where the work should go, not what you should do with your money.

For research and education. Not investment advice. Markets carry risk, including loss of capital.