How Fortress Thinks
Everything the Fortress engine does, explained for both an 18-year-old investing for the first time and a seasoned analyst who wants the formulas.
What is Fortress Intelligence?
Imagine you want to buy a phone but you don't know which one is actually good vs which one just has a nice ad. Fortress is like having a very experienced engineer friend who looks inside every phone and tells you honestly — this one is built to last, this one will break in 6 months.
Instead of phones, we do this for companies listed on the stock market. We run every company through a 6-layer quality engine, give them a score out of 100, and flag the ones that actually deserve your attention — not the ones with the loudest news.
Not a tip service
We never say 'BUY THIS NOW'. We explain WHY a company passed our quality checks.
Education first
Every stock in our list comes with a plain-English explanation you can actually understand.
Honest filters
Our scoring is automatic and rules-based. No one gets paid to promote a stock.
What's New — Engine v2.0
L2 Pricing Power
Added sector-relative margin premium scoring. Now checks if margins are above sector median — not just absolute thresholds. A company 30%+ above its sector peers earns maximum pricing power points.
L3 Relative Strength
Extended from 3-month only to multi-band: 3M (40%) + 6M (35%) + 1Y (25%). Historical consistent outperformance across all three horizons may indicate a structural trend, not noise. Past performance does not guarantee future results.
L5 Governance
Fully rebuilt. Now uses real ownership data: promoter/insider holding %, FII/institutional holding %, short interest ratio, and debt trajectory. The previous version used D/E as a governance proxy — this is the real thing.
L6 Valuation Gate
New layer (5 pts). Eliminates stocks in genuine bubble territory: P/E > 100 with growth < 25%, or EV/EBITDA > 60 with growth < 20%. Quality growth stocks at 30–60x P/E are NOT penalised.
Weights
Default weights updated: L1:25 + L2:20 + L3:10 + L4:25 + L5:15 + L6:5 = 100. L3 reduced from 15 to 10 (momentum is confirmation, not thesis). L6 new allocation.
Impact on existing scores
Stocks scanned before v2.0 used the old L5 proxy and no L6. Scores from previous scans are NOT comparable to v2.0 scores. Run a fresh scan to get accurate v2.0 results. The L5 rebuild in particular can change governance scores significantly — positively for founder-led companies, negatively for those with low insider ownership.
The 6-Layer Scoring Engine v2
Every company we scan gets put through 6 questions. Each question is worth points. A company needs at least 60 out of 100 to even show up on our radar. Click each layer to see what we're actually checking.
The analogy
Think of it like checking if someone has savings and no credit card debt before lending them money — AND checking whether their debt is going up or coming down every year.
What we actually check
- Does the company owe more than it earns? (Debt check)
- Does it actually generate real cash, not just accounting profit?
- Is it earning a good return on the money invested in it?
- What percentage of its profit actually becomes real cash in hand?
- Is the company's debt growing or shrinking year-on-year? (3-year direction)
PASSES this layer
Asian Paints, Pidilite — zero or very low debt, consistent cash generation, debt falling every year.
FAILS this layer
A real estate developer with 5x debt-to-equity, negative cash flow, and debt rising for 3 years.
Multi-Bagger Score
What is a multi-bagger?
A multi-bagger is a stock that multiplies your money — 2x is a double-bagger, 10x is a ten-bagger. The term comes from Peter Lynch (the most successful fund manager in history). Multi-baggers don't happen by accident — they have a specific structure.
The 5-Layer engine tells you if a company is safe. The Multi-Bagger Score asks a different question: does this company have the structure to multiply your money? A company can be safe but not a multi-bagger. The best stocks pass both.
Rocket
80–100 pts
Launcher
60–79 pts
Builder
40–59 pts
Crawler
20–39 pts
Grounded
0–19 pts
The 5 Components
How much room does the company still have to grow? We compare its size today to the total size of its entire market. A company that's captured only 0.1% of its market has 99.9% of the road ahead of it.
Why this is separate from the safety score
A company can be financially very safe but still be a boring, slow-growing business. Think of a utility company — safe as a bank, but it won't 10x your money. The Multi-Bagger Score checks for explosive growth potential separately from financial safety.
Coffee Can Mode
What is Coffee Can Investing?
In 1986, Robert Kirby described putting stock certificates in a coffee can and locking it in a safe for 10 years — no checking, no selling, no panic. The idea: find companies so fundamentally strong that you can ignore the market noise entirely. Saurabh Mukherjea (Marcellus Investment) adapted this for India, with two strict rules: Revenue growing at 10%+ AND Returns above 15% — every single year — for 10 years.
The key insight: most companies can have one good year. Very few can have 10 consecutive good years. The ones that do tend to have deep competitive moats — the kind that compound wealth quietly while everyone else is checking their portfolio every day.
4 clean years of Revenue CAGR > 10% AND ROCE > 15%. Extremely rare. This is the buy-and-hold-forever tier.
Nearly Coffee Can quality. One small lapse in 4 years. These are the stocks you put on a serious watchlist.
Growing into Coffee Can territory. The business is improving but hasn't yet shown 4 years of proven consistency.
Lumpy business. Revenue or returns are uneven year to year. Not suitable for the buy-and-hold Coffee Can approach.
Why 4 years, not 10?
yfinance (our data source) provides ~4 years of annual financial history. We run a 4-year consistency check as a proxy. A stock passing all 4-year checks is already in the top 5% of listed companies — it's ON TRACK for Coffee Can status, not necessarily already there.
Megatrend Tags
What is a megatrend?
A megatrend is a massive, multi-decade shift that reshapes entire industries — like the internet in the 90s, or smartphones in the 2000s. Investing in companies riding a megatrend is like surfing: even average swimmers look great on a big wave.
Fortress automatically assigns every stock a megatrend tag using the company's business description and industry. No manual work — the engine reads what the company does and classifies it. Click any megatrend below to understand what it is and why it matters.
India is replacing imported weapons with locally made ones. This is a 10-year policy shift — defence budgets never get cut.
The GEM Score System
The 5-layer engine tells you if a company is financially safe. The GEM Score asks a different question: is this a hidden opportunity — something genuinely good that the market hasn't noticed yet? It scores 0–100 across 4 criteria and assigns a tier (Diamond down to Quartz).
Diamond
80–100 pts
A company that is genuinely great, genuinely undervalued, AND showing early signs that others are starting to notice. These are rare. 2–5 per full scan.
Sapphire
60–79 pts
Very good. Missing one piece of the puzzle. Still worth serious research. Most reliable category for steady returns.
Emerald
40–59 pts
Interesting but needs a trigger — a specific event — before it becomes a buy. Put it on a watchlist.
Quartz
20–39 pts
Speculative. Only for money you are 100% okay losing. Like a small bet at a poker table.
The 4 Criteria
Is the stock cheaper than it should be? Not just in absolute terms, but compared to similar companies in the same industry.
Red flags that disqualify this criterion
- PE low due to one-time gain
- EV/EBITDA low because of high debt masking real leverage
Investment Genie — Portfolio Engineering
The idea in one sentence
Instead of trying to find the one perfect stock, Genie looks for a perfect mix of stocks that fits your specific level of patience and risk tolerance.
Genie uses a method called Cluster-Based Allocation. It divides high-quality stocks into four groups (clusters) and builds you a balanced map based on how much risk you can handle.
Defensive Cluster
Stable, boring, low-debt companies. Your portfolio's anchor during a storm.
Growth Cluster
Companies growing fast. They might be volatile, but they drive long-term returns.
Value Cluster
Genuinely good companies that the market has temporarily mispriced or ignored.
Megatrend Cluster
Companies riding a 10-year wave (Defence, Green Energy, Infrastructure).
Sovereign Alpha — The Self-Learning Engine
The idea in one sentence
Every stock pick Fortress makes is tracked automatically at 30, 60, and 90 days. If a pick was right, the system learns which scoring rules led to it and uses those more. If it was wrong, it learns to trust those rules less.
It's like a doctor who checks back on every patient to see if their diagnosis was right — and uses those results to become a better diagnostician over time. Most stock platforms never tell you if their picks actually worked. We track every single one.
Pick is recorded
When Fortress identifies a stock, the price is locked in as the starting point.
Reality check
At 30, 60, and 90 days the system automatically fetches the current price and calculates the return.
Learning
Which scoring rules predicted winners? Those get higher weight. Rules that didn't? Lower weight.
The fair game rule
If a stock went down because of a market crash or a surprise geopolitical event — that's not the system's fault. Fortress lets you tag these "unfair" outcomes so the learning engine is only judged on genuine prediction quality.
Market Weather — The Macro Context
Think of this as a weather forecast for your money
Just like you'd dress differently on a rainy vs sunny day — you should invest differently depending on what the economy is doing. This section tells you what the economy weather looks like right now.
Banks are charging a lot for loans right now. Companies drowning in debt are struggling. Companies with no debt are fine — or even benefiting.
Prices were rising fast. Now slowing down. This means everyday companies (FMCG, consumer goods) can stabilise their cost base.
The US dollar is powerful right now. Indian IT companies (paid in USD) are winning. Indian companies paying for imports in USD are struggling.
The economy has been growing for a long time. Smart money is moving to safer, essential companies (defence, healthcare). Avoid expensive growth stocks.
India's central bank (RBI) is holding steady. Indian banks are stable. India's economy is growing faster than most of the world.
Countries are charging each other more for imports (tariffs). Bad for companies selling to the US, but GOOD for Indian factories that global companies choose over China.
Sector Traffic Lights
Defence & Aerospace
Government always buys — budgets never stop. India replacing imports with local production.
Healthcare & Pharma
People always need medicine. Indian pharma sells to the whole world at low cost.
Indian Manufacturers (China+1)
Global companies moving factories from China to India. This is a 10-year shift.
Large IT Services
Good companies, but US clients may cut tech budgets. USD is helping, US slowdown may hurt.
Banks & Financials
Stable but not exciting. Get more interesting when RBI cuts rates.
Real Estate (high-debt builders)
Expensive borrowing is killing companies with big loans. Avoid builders with lots of debt.
Luxury / Premium Consumer
Late-stage economy = people are more careful with money. Luxury is the first cut.
Data Integrity & Transparency
Where does our data come from? No magic, just clean engineering.
Primary Engines
Stock Financials (NSE / US)
Yahoo Finance API (yfinance) + Alpha Vantage. We fetch full income statements, balance sheets, and cash flows to run L1–L6 layers.
Macro Indicators
FRED (St. Louis Fed) + RBI public data feeds for indices, yields, and commodity prices.
Update Frequency
NRI Disclaimer
Our US market data focuses on ADRs and large-cap tech. For HKEX, data is fetched via secondary providers and may have a 15-min delay.
What Fortress Does NOT Do
We do not predict prices
No one can tell you a stock will hit ₹500 by December. Anyone who does is guessing or selling you something.
We are not SEBI-registered advisors
Everything on Fortress is educational. It is not personalised investment advice. Always make your own decision.
We do not tell you when to sell
Entry is a science. Exit is personal. We give you the thesis — you decide when your thesis is broken.
We do not guarantee returns
Every investment has risk. Our scores tell you about quality — not about what the market will do next week.
We do not use tips or insider information
Every score is derived from publicly available financial data via yfinance and NSE disclosures.
We do not promote any stock
No promoter, company, or broker pays us to list their stock. Our scoring is fully automated and rules-based.
Key Terms Explained
How much has the company borrowed compared to what it actually owns. D/E of 0.5 means for every ₹1 of owned assets, it borrowed ₹0.50. Lower is safer.
Return on Capital Employed. If a company invested ₹100 crore and earned ₹20 crore profit, ROCE = 20%. Think of it as how efficiently they use every rupee.
The real money left over after running the business AND paying for all maintenance and growth. This is what the company can actually use to pay you dividends, buy back shares, or grow further.
Price-to-Earnings relative to how fast the company is growing. A stock at P/E 30 growing 35% (PEG 0.86) is CHEAPER than a stock at P/E 15 growing 5% (PEG 3.0).
On every ₹100 of sales, how much is left after paying for the product itself. A company with 60% gross margin earns ₹60 before paying staff, rent, or taxes.
What percentage of the company the founders/original owners still own. High promoter holding (> 50%) means they have skin in the game — they win or lose with you. This is now a core part of L5 scoring.
When a promoter borrows money and uses their shares as collateral. High pledging (> 30%) means if the stock falls, they're forced to sell — which pushes the price even lower.
Are the profits real? Some companies show profit on paper but never actually generate cash. Earnings Quality checks if the profit becomes real money in the bank.
When big fund managers and analysts haven't discovered a stock yet. When they finally do, they all buy at once — and if you're already in, you benefit from that rush.
When a company is doing great business but the stock price hasn't moved yet. The business is running fast but the stock is standing still — that gap is the opportunity.
A big economic force that helps a company without them doing anything special. Like how all Indian defence companies benefit when the government increases the defence budget.
Buying great companies and holding them for 10+ years without touching them — like putting a stock certificate in a coffee can and forgetting about it. Popularised by Saurabh Mukherjea.
Fortress's main discovery score (0–100) for finding hidden opportunities. Four criteria: Is it cheap? Is it ignored by big funds? Is the business healthy? Is the price about to catch up?
A rule you set before buying: 'If this stock falls 15%, I sell automatically, no questions asked.' It protects you from turning a small mistake into a disaster.
What percentage of the company's market value it generates in free cash every year. FCF Yield of 6% means for every ₹100 you invest, the company generates ₹6 of real cash annually.
A way to compare how expensive a company is vs how much money it actually makes from its core business. Cheaper than P/E because it removes the effect of debt and taxes.
A share of the company's profits paid directly to you as a shareholder in cash. If you own 100 shares and the dividend is ₹5/share, ₹500 lands in your account — no selling required. Think of it as the company paying you rent for owning it.
The company gives you free extra shares. If you have 100 shares and it announces a 1:1 bonus, you now have 200 shares. The price halves automatically — so your total value is the same. But it signals that management believes the stock is worth more.
One share becomes multiple shares. A 2:1 split means your 1 share becomes 2, but the price halves. Like breaking a ₹100 note into two ₹50 notes — same value, more accessible.
A stock that multiplies your money — 2x is a double-bagger (2 bags of money from 1), 10x is a ten-bagger. The term was coined by Peter Lynch, who turned the Fidelity Magellan Fund into the best-performing fund in history by finding these early.
A massive, decades-long shift that reshapes entire industries. The internet was a megatrend. EVs are a megatrend. India's defence indigenisation is a megatrend. Companies riding a megatrend get market tailwind even if they're average businesses.
Is the company's debt going up or coming down over the last 3 years? A company whose debt is falling every year is getting healthier — that change in direction often triggers a re-rating (the stock price catches up to the improving reality).
Beta Feedback — Help Us Build This
You are the product specification
Fortress is being built specifically for NRI long-term investors. Your real-world feedback — what works, what feels wrong, what's missing — is more valuable than any internal test. Every response is read by the team.
Does L5 feel right to you?
We rebuilt L5 with promoter holding + FII data. For the Indian stocks you follow closely, does the governance score match your real-world assessment? Tell us if it gets it wrong.
Are the tier thresholds right?
Rocket (80+), Launcher (65-79), Builder (60-64). Do the stocks you'd genuinely consider land in the right tiers? Or is the threshold too loose/strict?
What's the #1 missing signal?
Promoter pledging data isn't in yfinance. Neither is management commentary accuracy or dividend history quality. What would you add if you could?
Would you use this before investing?
The real test: have you used a Fortress scan result to validate (or reject) an investment thesis you already had? Did it add value or tell you something you already knew?
Things that are intentionally NOT in v2.0 (yet)
Tell us which of these matters most to you — that determines the priority for v2.1.