How to use this site

1. What this site is

This is a research desk for a long/short equity book held for 20–60 days, built on the fundamental process taught in the ITPM Professional Trading Masterclass (PTM). It does not tell you what to buy. It collects the data, applies the course's rules as a first reading, shows the reasoning next to every number, and leaves the decision to you.

• Long/short: you buy stocks you expect to rise (longs) and sell short stocks you expect to fall (shorts), so the book can make money in both directions.
• 20–60 days: a trade, not an investment. Longer than about 60 days drifts into a "grey zone"; beyond 120 days it is an investment.
• A trade idea needs four things: fundamentals, timing, a catalyst and a structure. Missing one means it is not a trade idea yet.
• Free data, updated every day: an automatic job runs after the US close (22:23 UTC) and rebuilds every page. A few surveys (ISM, NFIB, China PMIs) are entered by hand because they are not free in machine-readable form.

The process in one picture

The course works top-down first (the economy sets how long or short the whole book should be), then bottom-up (which stocks), then puts the ideas together as a portfolio. Each box below is a page; click it to open it.

1. 1Macro DeskIs the economy pointing up or down? → the portfolio bias: net long, neutral or net short.
2. 2Macro ideasWhich industries does macro help or hurt? Which regions?
3. 3SectorsWhich sectors is the market buying? Where are estimates rising?
4. 4ScreenerWhich stocks stand out from their sector? Long and short candidates.
5. 5Trade ideaWork one stock through the template: targets, catalysts, timing, structure.
6. 6IdeasKeep a pipeline of 10–15 ideas and review them.
7. 7Portfolio LabCheck the whole book's risk before you trade.

Around this core: This week (your weekly to-do list), Regime, Global, Calendar and Evidence (context for the macro reading), and Data health (is the data fresh?).

Finding your way: the menu and the home page

• The menu at the top follows the same order: This week, Macro (Macro Desk, Regime, Global, Calendar, Evidence: the top-down part), Stocks (Macro ideas, Sectors, Screener, Ideas: the bottom-up part), Portfolio and Guide. Click or hover Macro or Stocks to open its list; the group of the page you are on is underlined.
• On the right: the pulse icon opens Data health, the moon / sun switches between the light and dark theme. On a phone the menu folds into the ☰ button.
• The home page is a summary of today: the portfolio bias with its last two years, the four blocks, what changed in the last month, the next reports and central-bank meetings, the regime and recession odds, the four process steps with live counts, and whether the data is fresh. Every card links to the page behind it.

2. Start here: your first 15 minutes

1. Open This week. It lists the seven weekly steps in order and marks the ones that need attention from this week's data. Use it as your to-do list.
2. Open the Macro Desk and read the big word at the top: Net long, Neutral or Net short. That is the automatic portfolio bias. Read "What changed" and the alerts next to it.
3. Open Macro ideas and look at the industries with a strong tailwind or headwind.
4. Open the Screener, pick a sector, click the type Positive outlier, tick "Volatility ≥ 1.5× VIX only", and sort by EG1.
5. Click a ticker. On its trade-idea page press Start a long idea and fill the notes. The idea is saved in your browser and appears on the Ideas page.
6. When you hold (or plan) several positions, enter them in the Portfolio Lab and check volatility, correlation and exposure.

3. Map of the pages

4. Macro Desk: the portfolio bias

What you see, top to bottom

1. As of: pick a past month-end to see what the desk said then, using only data that was known at the time. Reset overrides removes your own scores.
2. Portfolio bias: the big word and the meter from −1 (short) to +1 (long). A score of +0.25 or more is Net long, −0.25 or less Net short, in between Neutral (balanced).
3. Four blocks: US money markets, US surveys, cyclical prices/FX/equities, international. Each block is the average of its indicators; the bias is the average of the blocks, so a block with many indicators cannot dominate.
4. What changed in the last month: every indicator whose score moved, and the bias itself.
5. Read together: alerts: blocks that disagree, credit spreads widening in junk first, and the data vs prices alert (the data says long but the S&P 500 is falling, or the opposite).
6. Indicator cards (further down): each of the 19 indicators with its score, its state in words ("Expanding", "Junk widening first"…), the reason, the 10-year percentile, the date of the data and a strip of its monthly scores. Under Your reading you can override the score with −1 / 0 / +1; the bias recalculates at once and the override is saved in your browser.
7. Confirmers: coincident and lagging data (CPI, payrolls, jobless claims, industrial production, GDP). They confirm or contradict the leading indicators but are never scored.
8. Bias history: the monthly bias since 2000, with the S&P 500 drawn underneath (grey, right axis) so you can see how the reading lined up with the market.
9. Copy brief and Download CSV (top of the page): the view on screen as a short text to paste in your notes (bias, blocks, changes, alerts, indicators at extremes, regime and recession odds) or as a spreadsheet of every indicator with its automatic score and your override.

The 19 leading indicators

✎ = entered by hand (see section 14).

Click Chart, components & reading guide → on any card to open its indicator page: a chart (1Y to Max), the course's way of reading it, what to watch for, the first-reading rule and the history of its score.

Try it: how the bias is calculated

These are today's 19 automatic scores. Change any of them, as you would with an override on the Macro Desk, and watch the blocks and the bias move. Nothing here is saved.

Global, Calendar and Evidence

• Global: money markets by country (real rates, yield curves) and the regional equity markets side by side (US vs Europe vs Japan vs China), to answer "is the world up or down, and who is leading?". Context only; not scored again.
• Calendar: US data releases, Fed / ECB / BoJ / BoE meetings and the dates the hand-entered reports come out. The 20–60 day window is highlighted: these are the events that can move a trade you open today.
• Evidence: what happened to the S&P 500 after each past bias reading, with the number of months behind every figure. It uses real-time data vintages so the past is not judged with revised numbers. It reports results; the rules are never tuned to make it look good.
• At the bottom of Evidence, the Forward test checks the Screener live: every week all candidate stocks are written down with their price, and their return over the next 20, 40 and 60 trading days (against the S&P 500 and their sector ETF) is filled in as the days pass. The key number is the long-short spread (positive minus negative outliers). It needs many weeks before it means anything; the t-statistic tells you when (|t| above about 2).
• A second, pre-registered test runs in the same record: do stocks whose analysts are raising their estimates beat those being cut (H1), and do positive outliers with rising estimates beat the other positive outliers (H2)? Research suggests revisions matter more than how high the forecast growth is; this record will show whether that holds here.

5. Regime: the bigger picture

The Macro Desk scores leading indicators; this page shows the backdrop they move in. It is context only: nothing on it changes the bias.

• Regime map: across, growth (the Chicago Fed activity index, 3-month average; 0 = trend); up, whether core inflation is speeding up. Four regimes: Recovery (growth above trend, inflation easing), Overheat (above trend, rising), Stagflation (below trend, rising), Reflation (below trend, easing). The line is the last 24 months; follow its direction, because a reading near a line can flip with the next release.
• Two questions: the map shows the direction (is the economy speeding up or slowing, is inflation rising or easing). Next to it, Level (how hot) shows how far the economy is from the Fed's targets: inflation vs 2%, output vs its potential, unemployment vs its natural rate. Both can be true at once ("Recovery" with hot levels means hot but cooling at the margin). A yellow box warns when core CPI, published earlier, points the other way: the next release may flip the regime.
• Month by month: the regime as it could be read at each month-end since 1999, with only the data published by then.
• What did well in each regime: the average return over the next 3 months of five asset classes and of each sector against the S&P 500, after the months of each regime. The current regime's column is marked "now". Check n: a few dozen overlapping months is a small sample.
• Recession odds: the New York Fed's model turns the 10-year minus 3-month yield spread into a probability of recession 12 months later. Watch the rise; the table shows how high it got before past recessions, and the note shows when it gave a false alarm.
• Fed vs Taylor rule: where a simple rule (inflation and the output gap) would put the fed funds rate. Fed funds below the rule = easier policy than the rule suggests.
• Treasury curve: yields from 1 month to 30 years; drag the slider (or press "1 year ago") to compare with any week of the last 10 years.

6. Macro ideas: from the economy to industries and countries

• Net: every macro signal that reaches an industry adds +1 or −1 (for example rising copper prices +1 for copper miners; rising oil −1 for airlines), and so does the industry's place in the latest ISM growth or contraction list. +1 or more = tailwind, −1 or less = headwind, ±2 or more = strong (two or more drivers agree). Hover a driver to read why it matters.
• Macro-driven ideas: the Screener's long candidates inside tailwind industries and its short candidates inside headwind industries. A good place to look for ideas that agree top-down and bottom-up.
• International ideas (ADRs): companies based abroad with a US listing. EPS growth is compared across countries; P/E is hidden where the estimates are in another currency than the share price. Leader = top-quarter growth in both years; laggard = falling in both years.
• China check: China's official PMI is often too smooth to trust. The check compares it with what analysts are doing to Chinese companies' earnings estimates. "Disagree" means: trust the companies, not the official number.

7. Sectors: where to look first

The Screener works one sector at a time. This page puts the eleven sectors side by side so you can decide which ones to screen first. It is not a lesson step: it extends the course's method (sector averages, cyclical vs defensive) and is context only. Nothing on it changes the bias or the types.

• Rotation map: each sector ETF (XLK, XLE…) against the S&P 500. Across = how far it is ahead (right) or behind (left) over 3 months; up = whether that gap grew or shrank in the last month. Leading (ahead and gaining), weakening (ahead but fading), lagging (behind and still losing), improving (behind but catching up). Sectors tend to move clockwise.
• Sector cards: the colour shows the 3-month gap to the S&P 500 (blue ahead, red behind); the line is the ETF divided by the S&P 500. Below: median EPS growth, P/E and PEG of the sector's clean stocks, how many outliers it holds, the macro wind (the Macro ideas score of its industries) and the revision bar: the share of stocks whose EPS estimate analysts raised (blue) or cut (red) in 90 days.
• Growth vs valuation: the sectors on EPS growth and P/E. Bottom-right = fast growth that is cheap; top-left = expensive for little growth.
• Risk appetite: cyclical and sensitive sectors against defensive ones. Rising = the market is buying risk. The top tile says whether that agrees with the Macro Desk bias; when it does not, read both again before acting.
• How to use it: favour longs in sectors that are leading or improving, with rising estimates and a tailwind; look for shorts where the sector is lagging and estimates are being cut. Then open the sector in the Screener ("Screener: N stocks →") or the candidates at the bottom.

8. Screener: finding outliers

About 2,000 US-listed stocks above $2 billion, compared inside their sector. The course looks forward: what analysts expect for this year and next, not last year's accounts.

The 11 numbers

Try it: the 11 numbers from four inputs

Types: where a stock sits against its sector

Each stock gets one type, from its EPS growth ranked against the clean stocks of its sector (the percentile: 90 means it grows faster than 90% of them). The course gives the types; the cut-offs (top and bottom quarter) are our choice.

Data checks (lesson 28)

Tags such as Other currency, Feeds disagree, Extreme P/E or Few analysts warn about unreliable numbers. Red tags (and grey rows) keep the stock out of the sector averages and the types. Never trade on a number you have not checked.

Volatility filter

A stock must be able to move enough within 20–60 days: its realised (past 3 months) and implied (from options) volatility should be at least 1.5× the VIX. Tick "Volatility ≥ 1.5× VIX only" to keep only those; the column Vol ÷ VIX shows the ratio.

Evidence: accounts

The button Evidence: accounts switches the table to backward-looking numbers from the companies' SEC filings: margins, EV/EBIT, EV/EBITDA, debt ratios, free cash flow. They explain why money likes or dislikes a stock; they do not forecast.

9. The trade-idea page: one stock through the template

Every stock in the Screener has this page. The data fills what it can; you fill the rest. It follows the course's trade-idea template in five sections:

Price targets

The course builds two targets for each year. EPS-based (upper, bullish): the P/E stays as it is today and earnings grow, so the price grows with EPS. Sales-based (lower, realistic): the price-to-sales stays as it is and sales grow. Always a range, never one number. Try the course's own example (Zendesk, May 2020):

Is the target realistic for 20–60 days?

A stock with 40% annual volatility moves about ±13% in 40 days in a normal ("1σ", one standard deviation) case; about two moves in three stay inside that. A target that needs a move of more than 2σ is unrealistic for the horizon: take a nearer target or a longer view. On the page each target carries its σ tag: 0.8σ realistic, 1.5σ stretched, 2.6σ unrealistic.

Saving an idea

1. Press Start a long idea (or short). The page now shows note boxes in every section.
2. Write as you go; everything is saved in this browser at once ("Only in this browser"). The Template x of 6 filled checklist shows what is still missing.
3. Set the status (researching → ready → active → closed / dropped), the horizon in days and your entry price.
4. Optional: Publish to the repository opens GitHub with the idea as a file; press Commit changes. Every device then sees it. The repository is public, so publish only what you are happy to share.
5. When you get out, set the status to Closed: Exit price and date appear (today's price and date are filled in; correct them to your real fill). The trade then counts in the track record.

10. Ideas: your pipeline

• Aim: 10–15 open ideas, about 2 new ones a week (each takes 2–4 hours). The tiles show where you stand.
• Targets: conservative / optimistic for the current fiscal year, with your own assumptions; "to the conservative" is how far the price still has to go.
• Flags: Review after 7 days without an update, Stale after 14, Target reached (be active around targets), Targets against you, earnings within 7 days.
• Export / Import moves the ideas kept in this browser to another device as a file.
• Track record (bottom of the page): every closed idea with its return for the side, the S&P 500 over the same days, the stock's beta and the alpha (the return minus what the market move would have given at that beta). On top: hit rate, average win and loss, payoff ratio and average alpha. This is how you find out whether your process works: judge it after dozens of trades, not a few.

11. Portfolio Lab: risk of the whole book

1. Enter the Current book (ticker, long or short, size in $) and the account size. Add my active ideas fills it from the Ideas page.
2. Build a Proposed book (start from the current one) with the change you are thinking about, and compare the two in the table before you trade.
3. Check the targets: volatility 15–30% a year (aim 17.5–22.5%), average correlation between −0.3 and +0.3, a beta-adjusted net exposure in line with the macro bias, and the right number of positions for your account.

Two ways to backtest

• Non-rebalanced: your dollar sizes invested on day one and left alone. What your account would really have done.
• CEW (constant equal weight): every position weighted equally again each day. The quickest way to see which basket does better.
• Drawdown: how far the book fell below its previous high. The deepest one is the "max drawdown".
• Factor exposure: is the book a stock-picking book or a hidden style bet? Its daily moves are compared with four ETF factors (market, small vs large, value vs growth, momentum). A "style bet" (e.g. +40% momentum) means a turn in that style can hurt every position at once; the table shows the ETF trade that would neutralise it, and how much of the book's moves is truly stock-specific.

Try it: why correlation matters

Two positions of the same size. Move the correlation and the sides: a long and a short in stocks that move together cancel each other's risk; two longs that move together add it up.

Beta and exposure

Beta measures how much a stock moves with the S&P 500 (beta 2 = twice as much). $10,000 long of a beta-2 stock against $10,000 short of a beta-0.5 stock is 0% net exposure but +150% beta-adjusted: the book still leans long the market. The beta-adjusted net is the honest measure of how long or short you are.

Sharpe framing

The return a strategy needs = minimum Sharpe ratio × portfolio volatility. At 15–30% volatility: CFDs (Sharpe 1.5) need 22.5–45% a year, US stocks long/short (Sharpe 2) 30–60%, options (Sharpe ≈3.3) 50–100%.

12. This week: the weekly routine

The course's retail routine takes about 8–12 hours a week. The page lays it out in order with our suggested split (about 10 hours): enter the reports (0.5 h) → Macro Desk (1.5 h) → Macro ideas (0.5 h) → Screener (1.5 h) → two new ideas (4 h) → review open ideas and catalysts (1 h) → portfolio (1 h). Tick a step when it is done; the ticks start fresh every Monday.

13. Where the data comes from and when it updates

1. Free sourcesFRED / ALFRED, Yahoo Finance, Eurostat, OECD, University of Michigan, SEC EDGAR, central-bank calendars
2. Your hand-entered reportsISM, NFIB, China PMIs in data/manual/*.csv
3. Daily updateevery day at 22:23 UTC (after the US close), and whenever you change a manual file
4. This websiteevery page rebuilt; the last good copy is kept if a source fails

Data health

• ✓OK fresh; !Stale older than expected; !Fetch failed today's download failed and the last good copy is used (usually fixes itself the next day); ×No data nothing yet.
• For the hand-entered files: Up to date, New report out (a report is published and waiting for you), Awaiting data (missing for too long: the indicator is left out of the bias, never guessed).

14. Entering the hand-entered reports

Five files, one row per month. The Calendar and This week tell you when a new report is out.

1. On GitHub, open the repository → folder data/manual → the file (for example ism_manufacturing.csv).
2. Click the pencil icon (Edit this file).
3. Add one line at the end for the new month, in the same order as the header line, e.g. 2026-10,52.1,53.4,…. The month is written YYYY-MM; leave a value empty if you do not have it.
4. Click Commit changes… and then Commit changes.
5. The site rebuilds in a few minutes. If a value is wrong (wrong month format, out of range, duplicate month), the update run turns red, GitHub e-mails you, and Data health shows the line and what is wrong.

15. What is saved where

Clearing your browser's site data, or using a private window, removes what is kept "in this browser": export your ideas and books now and then.

16. Glossary

Every term used on the site. Type to filter.

17. Questions