Strategy — Described Approaches Instead of Guesswork

Most novice investors trade based on intuition. They see a news report about a company, hear a tip from a friend, notice a chart that "looks like it's going to rise" — and buy. Sometimes it works, mostly it doesn't. And because they don't remember why they bought, they can't learn from their mistakes. It's guessing disguised as decision-making.

The Strategy tab (/strategy) is the answer to this problem. Instead of random tips, it provides you with a library of specific, tested approaches — each with rules on when to buy, when to sell, and most importantly, hard numbers that show how well the approach has worked in the past.

★ Insight: A strategy is not "a stock to buy." A strategy is a repeatable recipe — a set of rules for selecting stocks and managing trades. A good strategy works across dozens of trades, not just once by chance.

---

What Strategy Is For: Proven Recipes Instead of Chance

Imagine two chefs. The first cooks "by feel" — sometimes the food is great, other times inedible, and they never know why. The second has a tested recipe: precise measurements, procedure, time. Their food is consistently good because the recipe has been through dozens of trials.

An investment strategy is such a recipe. It tells you: what to look for, when to enter, when to exit, and how well it worked (hard historical numbers, not promises). Your task is no longer to invent your own recipe from scratch — that's years of research work. Your task is to choose a proven recipe that fits your style and stick to it.

---

Guide: Long-term vs Swing

The first decision is choosing the horizon — how long you want to hold stocks:

	Long-term Strategies	Swing Strategies
Holding Period	months to years	days to weeks
What Drives Returns	company fundamentals	technical movement + momentum
Rebalancing	typically monthly	per trade
For Whom	patient investor	active trader
Examples	Top Composite, Multibagger	CANSLIM, VCP, breakout

★ Insight: There is no "better" horizon — there is only your horizon. If you check your portfolio once a month, swing strategies don't make sense for you: you'll miss exits. Choose the horizon that fits your life, not the one that sounds more exciting.

---

What Is Edge and Why Every Strategy Must Have It

Edge is the reason why a given strategy should earn more than passive index buying. Without edge, a strategy is just an expensive and complicated way to do what the index does for free.

Edge is measured as alpha (α) — how many percentage points the strategy added compared to the index (S&P 500 / SPY). If a strategy made +35% and SPY +25%, the alpha is +10 percentage points (pp).

But the alpha number alone is not enough. The key question: was that result real skill, or just luck?

---

How to Recognize a Proven Strategy from Marketing

This is the most important part. Because anyone can find a backtest that looks fantastic. The art is to recognize which one is real and which is an illusion. QMA uses three hard criteria.

1. The t-statistic — "is it skill or luck?"

The t-statistic measures how confident you can be that the return wasn't just a lucky
coincidence. The higher the number, the less likely it is chance.

Textbooks usually quote a threshold of 2. That applies only to a single pre-registered test.
Once you test hundreds of variants and pick the best one, the result is not a finding — it is the
result of a search, and the bar has to rise with the number of trials.

★ Insight: Flip a coin ten times and six heads surprises nobody. Flip it a thousand times and

pick the best ten-flip stretch — and suddenly you have a "result". Since 31 July 2026 QMA keeps a

permanent ledger of every trial run and derives the significance bar from it. The bar therefore

sits well above that textbook 2, and a number of previously "verified" results did not clear it.

2. n ≥ 30 — "Was it tested on enough trades?"

n is the number of trades. A strategy that "made +200% on 6 trades" means nothing — six coin flips can easily land heads six times.

• n < 30 — small sample. The result is exploratory, not proven.
• n ≥ 30 — minimum reasonable sample.
• n in hundreds — solid foundation (e.g., CANSLIM in QMA tested on n ≈ 805).

3. Does it hold out-of-sample?

A strategy can be "tuned" so perfectly to historical data that it looks brilliant — but only because it memorized exactly that data. Once new data comes in, it falls apart. This is called curve-fitting. Defense: build the strategy on one part of history and test it on another it hasn't seen.

★ Insight: The toughest out-of-sample test is live trading. You can embellish a backtest, but live data has nowhere to hide. That's why QMA tracks live performance alongside the backtest (via /predictor) — and takes live numbers as decisive.

---

Tier-grouping: ✅ / ⚠️ / ❌

QMA categorizes each strategy into one of three groups — the backbone of the entire tab:

Tier	What It Means	Criteria	How to Treat It
✅ Proven	Statistically supported advantage	t-statistic above a threshold derived from the total number of trials · sufficient sample · holds out-of-sample	Highest documented tier
⚠️ Experimental	Hint, but not proven	small sample OR low significance	Monitor, possibly small position
❌ Over-optimized	Great backtest, fails live	beautiful historical numbers, but declines live	Do not use. Trap for novices

★ Insight: Tier ❌ is the most valuable thing on the entire tab — and you won't see it anywhere else. Competitors will only show you their best backtests. QMA deliberately shows you strategies that look great but fail live, and warns "don't do it." Example: the Champion strategy lost ~7 pp out-of-sample — it was over-optimized, hence removed from the flagship position.

---

Long-term Legends: Catalog of Approaches

The /strategy/long-term section contains "legends" — approaches with a long public history
(top-rated companies by composite score, companies with insider buying, candidates for multiple
appreciation, recovering companies, companies buying back their own shares, high-quality compounders).
Each card describes the selection logic and links to the current tracking state.

🚨 CORRECTION NOTE (31 July 2026) — the performance figures that used to stand here are WITHDRAWN.

>

They rested on a historical scores table whose older part was shown on 15 July 2026 to have been

derived from later prices — that is, from the future. On 31 July 2026 the significance

threshold was also tightened: it has to account for every test we have ever run, not for the

variants of a single run.

>

The figures are not being replaced by smaller ones. The inputs were never valid as of the day

in question, so an honest replacement does not exist — the claim goes away entirely. Only

forward tracking since 18 February 2026 is trustworthy.

★ Insight: An honest strategy card does not show only a return. It shows the number of trades,

the t-statistic and, above all, whether this is a backtest or forward-measured operation. If any of

those is missing, it is missing on purpose.

Beware of look-ahead bias (backtest mistakenly uses future information) and survivorship bias (analyzing only companies that "survived"). QMA addresses these pitfalls diligently — for Multibagger, it explicitly acknowledges survivorship bias and provides realistic expectations (out of the Top 50 candidates, usually only a few become true multibaggers).

---

Multibagger Pro v3: Empirically Calibrated Engine

Shows how to do a strategy honestly. The original version was based on academic literature ("multibaggers have high ROIC, high margins..."). The problem? These numbers describe what a multibagger looks like at the end of the journey, not at the start, when you enter it. Academic filters thus eliminated future winners — classic survivorship bias.

Version v3 turned it around: the team retrospectively analyzed real winners (500%+ over 5 years) and compared them with losers at the entry state. It measured which traits actually distinguished them (by effect size — Cohen's d). Most "textbook" metrics didn't distinguish at all. Only four things worked: momentum/distance from the high (counter-momentum), 3-month return before entry, company size (sweet spot of smaller capitalization), insider buying.

★ Insight: Marketing would write "520% return!!!". QMA writes "that is how one past test window came out; the t-statistic did not clear the threshold derived from the number of trials, so this is an exploratory observation rather than a finding — and after costs and slippage substantially less of it remains". The second sentence may not excite you, but it will protect your money.

---

Swing Methodologies: For Active Traders

The /strategy/swing section gathers technical-momentum methodologies (days to weeks):

Methodology	Principle	Sample
CANSLIM	Growth stocks with accelerating earnings breaking out of consolidation (O'Neil)	n ≈ 805 → ✅
VCP	Gradual "tightening" of volatility before breakout	smaller → ⚠️
Momentum breakout	Breakout from consolidation with volume increase	smaller → ⚠️
Episodic pivot	Explosive move after a catalyst	smaller → ⚠️

Among the methodologies, only CANSLIM has a robust sample; others are often exploratory (n < 30), which QMA fairly marks with a yellow ⚠️ on each card.

★ Insight: Swing strategies require more discipline — exits are managed quickly (profit-target, stop, time limit). With a long-term strategy, you can miss a few days and nothing happens; with swing, lack of discipline costs you the entire edge.

---

Live vs Backtest: The Toughest Test

Each card shows two types of results:

• ○ Backtest — how the strategy would have worked in the past. Useful but optimistic.
• ● Live tracking — how it works now, on real data. The real proof.

The /predictor tool tracks live performance forward in time.

★ Insight: A rule that will save you money: when live numbers significantly lag behind the backtest, trust the live ones. A backtest is an opinion about the past; live data is a fact about the present.

---

Practical Procedure: From Strategy to Trade

1. Choose a strategy based on horizon and style.
2. Verify tier and statistics. Rule for beginners: start exclusively with ✅ proven. Ignore ❌, just monitor ⚠️. On every card, look at the number of trades and at whether the figure is a backtest or forward-measured operation.
3. Read the strategy details — entry/exit rules, when it works, backtest. Understand the logic.
4. Compare live vs backtest (/predictor).
5. Find specific stocks through Screener — the strategy tells you what traits to look for, Screener filters current stocks.
6. Record why you entered in a trading journal.

★ Insight: The connection Strategy → Screener → Journal is the whole point of QMA. Strategy provides the recipe, Screener provides the ingredients, Journal records the result.

---

Why No Strategy Works All the Time

No strategy earns in all market phases. Momentum and breakout love a strong bull trend and suffer in sideways. Value strategies lag in euphoria but shine when the market seeks quality. Counter-momentum entries are most valuable in corrections. Therefore, link Strategy with market regime detection on the Dashboard.

★ Insight: A weaker period for a strategy ≠ a broken strategy. Even a ✅ proven strategy has months when it lags — it's normal. The mistake is not that the strategy sometimes loses; the mistake is abandoning it exactly when it has a weaker phase.

---

Common Mistakes

Mistake 1 — Jumping between strategies. Edge manifests only over a longer sample. Choose 1–2 strategies and stick with them for at least dozens of trades.

Mistake 2 — Trusting ❌ over-optimized. A high backtest without live data support is a red flag, not an attraction.

Mistake 3 — Ignoring the t-statistic and the number of trades. +200% on six trades is much less reliable than a more modest figure over hundreds of trades. And both are less reliable than they look if hundreds of other variants were tested first.

Mistake 4 — Mismatch of horizon and life. Align the strategy horizon with how often you realistically monitor the market.

Mistake 5 — No journal. Without records, you won't learn whether the problem was in the strategy or your execution.

---

Summary

The Strategy tab transforms investing from guessing into a repeatable, proven process: choose a horizon, follow the tiers (✅/⚠️/❌), read the numbers honestly (return always with t-stat and n), trust live data more than the backtest, and connect it to practice (Strategy → Screener → Journal) with regard to market regime. The greatest value of the tab is not in showing you the best strategies — it's in teaching you to distinguish a proven advantage from a good-looking illusion.

---

Disclaimer: QMA is an analytical and educational platform, not an investment advisor. This article describes how strategies and their statistical evaluations work — it is not a recommendation to buy or sell specific securities. The numbers provided are backtest/historical and serve to illustrate the methodology. Past performance does not guarantee future results. Statistical significance (t-stat) reduces but does not eliminate the risk of chance/bias. Investing carries the risk of capital loss. Consider your own situation before making decisions, or consult a licensed advisor.