First, let's correct a misconception: when many people hear "quantitative strategies," they think of black-box technology that only PhDs can understand.
This impression is only half right.
Among the nine mainstream schools of quantitative trading strategies, some can be handled by ordinary people working with AI, while others require hundreds of millions in infrastructure just to get a seat at the table. The problem is that most popular science articles either mix them all together in a confusing way or skip the most critical question: "Can an ordinary person actually do this?"
In today's article, I will use a simple framework—Traffic Lights—to go through all nine schools: which are Green Lights that ordinary people + AI can start with now; which are Yellow Lights that require extra investment but are worth learning; and which are Red Lights that ordinary people should give up on early—not because you aren't smart enough, but because the entry threshold is wrong.
No formulas, just the logic of what each strategy is "actually betting on."
First, an Iron Rule: Beware of "Backtested Perfection"
Before going through the nine schools, let me give you a warning.
There is a consensus in the industry: In 2026, if any strategy shows a backtested Sharpe Ratio (an indicator of how "steadily" it earns) exceeding 3, your first reaction should not be ecstasy, but suspicion—there is a high probability that something is wrong with the backtesting method (such as accidentally using future data or picking survivors when selecting samples).
Only institutional strategies that use real money, extreme leverage, and grab speed at the millisecond level can "reasonably" run such ridiculously high numbers. If an ordinary person backtests a strategy with a Sharpe of 5, they haven't struck it rich; they've calculated it wrong. Remember this rule so you won't be fooled by "beautiful backtests" when looking at the strategies below.
🟢 Green Light Zone: Ordinary People + AI Can Play Now
These three schools have simple logic, public data, and AI can directly help you implement them. This is where beginners should start.
- Momentum Strategy—Going with the flow, but replacing emotion with discipline
Principle in one sentence: Things that rise a lot tend to keep rising in the short term; things that fall a lot tend to keep falling. Academia has repeatedly verified this phenomenon in stock, commodity, forex, and bond markets—the reason is that information takes time to spread, and human nature likes to follow the crowd.
Can ordinary people touch it: Yes, and it's the top choice for entry. This is essentially "buying high and selling higher," but the key to the quantitative version is using fixed rules to replace emotions—for example, "buy when the 20-day moving average crosses above the 60-day moving average," rather than chasing highs based on feeling.
What AI can do for you: Tell the AI your momentum rules in plain language, and it can directly write backtesting code for you, allowing you to see historical performance in minutes.
Risk Warning: Momentum's biggest enemy is the "sharp turn"—a trend can suddenly reverse without warning, and at that point, momentum strategies will be hit hard.
- Mean Reversion—The rubber band snaps back
Principle in one sentence: If a price deviates too far from its historical average, there is a high probability it will be "pulled back"—like a stretched rubber band that eventually snaps back to its original position.
Can ordinary people touch it: Yes. This is the "opposite brother" of the momentum strategy—one bets on "trend continuation," the other on "extreme correction." The two take turns being effective in different time scales and market environments, making them a classic combination for building a portfolio.
What AI can do for you: Judging "what counts as deviating too far" requires some statistical skill (in plain language: calculating how many standard deviations the current price is above the historical average). AI can directly help you with this calculation and visualization.
Risk Warning: Mean reversion performs poorly in extreme one-sided markets—something "undervalued" can keep falling because it has no intention of returning to the mean.
- Breakout Strategy—Follow through when it breaks key levels
Principle in one sentence: When a price breaks through a key range of long-term consolidation (such as a one-year high), it often signifies the start of a new trend, and following this breakout is often profitable.
Can ordinary people touch it: Yes, this has the simplest rules. "Buy when it breaks the previous high, sell when it breaks the previous low"—the logic is so straightforward even a primary school student can understand it.
What AI can do for you: Help you scan a basket of stocks and automatically find targets that are "breaking through key levels," so you don't have to watch the screen yourself.
Risk Warning: The biggest trap is the "false breakout"—it breaks out briefly and then immediately retracts, trapping those who chased in. This is why breakout strategies are usually confirmed with trading volume.
🟡 Yellow Light Zone: AI Can Significantly Lower the Threshold, but Requires More Effort
These four schools are more complex than the Green Light zone. Ordinary people working alone will find it difficult, but 2026 AI tools have lowered the threshold to the point where it's "attainable if you study seriously."
- Pairs Trading / Statistical Arbitrage—Two people who are always in sync, but one suddenly gets distracted
Principle in one sentence: Find two assets that have been highly synchronized historically (like Coca-Cola and Pepsi). When their price spread suddenly widens—one rises while the other falls—buy the cheap one and short the expensive one simultaneously, betting that their spread will eventually shrink back to normal levels.
Can ordinary people touch it: The simplified version is touchable, but be careful. The institutional version of statistical arbitrage manages hundreds or thousands of positions simultaneously, pursuing "complete market neutrality" (not afraid of ups or downs, just eating the spread). Ordinary people play the simplified version—picking a few pairs of highly correlated assets and doing small-scale spread trading.
What AI can do for you: Judging whether "two assets really have a stable statistical relationship" requires mathematical tools (professionally called "cointegration tests"). AI can run this calculation process for you directly.
Reality Reminder: This type of strategy has a "capacity ceiling"—it earns very small spreads. Once the capital scale becomes large, your own trades will actually erase the spread. This is precisely the natural advantage of ordinary people: your capital is small, so you won't encounter this problem, whereas institutions are limited by their size.
- Factor Investing—Labeling stocks and selecting them by label
Principle in one sentence: Group stocks by certain common characteristics (labels like "cheap," "highly profitable," "recently rising") and systematically buy stocks with certain labels because historical data shows some labels outperform the market in the long run.
Can ordinary people touch it: Yes, and it is the most "academically formal" path. This path is supported by decades of public academic research, not metaphysics.
What AI can do for you: Using open-source tools like Qlib, ordinary people can run a complete process of "mining factors → testing → combining"—something only institutional quant teams did a few years ago.
Risk Warning: Factors that were once effective may gradually fail because too many people are using them (this is called "factor crowding"). A factor that works well today is not guaranteed to work tomorrow.
- News Sentiment Trading—Let AI help you read news 24 hours a day
Principle in one sentence: Market sentiment is quickly affected by news, earnings reports, and social media discussions. If you can read the sentiment behind this information faster and more accurately than others, you can get ahead.
Can ordinary people touch it: This is a school that only truly opened to ordinary people in 2026. In the past, processing massive amounts of text and judging sentiment required a team that only professional institutions could afford. Now, a trained open-source financial language model can be run by an ordinary person on a consumer-grade graphics card.
What AI can do for you: This is almost an AI-native strategy—letting AI read earnings call transcripts, regulatory filings, and news flashes in real-time to provide sentiment judgments. This used to be the most expensive part of this school; now it's almost free.
Risk Warning: AI's sentiment judgment is not omnipotent, especially when the information itself is contradictory or when "expectations have already been priced in."
- Machine Learning Strategy—Let AI find patterns itself, rather than you setting rules for it
Principle in one sentence: In the previous strategies, rules were thought of by humans first and then executed by computers. This category is the reverse—throw massive amounts of data at a model and let it find complex patterns that the human brain cannot easily discover.
Can ordinary people touch it: Yes, but be prepared: this is the one most likely to "deceive yourself" among the nine schools. The more complex the model, the easier it is to "memorize" patterns in historical data that don't actually exist (professionally called "overfitting")—the backtest looks like a painting, but it falls apart in live trading.
What AI can do for you: Current open-source tools have standardized the process of "training a decent model," so ordinary people don't need to write code from scratch.
Iron Rule: The more complex the model, the more rigorous the "out-of-sample testing" required (verifying the model with new data it has never seen). If you don't know how to do this step, the risk of machine learning strategies is greater than the reward for you.
🔴 Red Light Zone: Ordinary People Should Give Up Early; It's Not a Matter of Ability, It's a Matter of Qualification
Frankly, for the last two schools: Ordinary people shouldn't waste their time. This isn't about IQ; it's about the entry ticket.
- Market Making—Being a middleman to earn the spread, but the opponents are the fastest institutions in the world
Principle in one sentence: Simultaneously post two quotes, "I am willing to buy" and "I am willing to sell," earning money through tiny spreads. Essentially, it's providing liquidity to the market and acting as a middleman.
Can ordinary people touch it: No. The winning factor in this game is speed and capital scale—whoever's quoting system reacts one millisecond faster can grab that spread before others. This requires institutional-level technical investment. Ordinary accounts and network latency don't even qualify for registration.
- High-Frequency Trading (HFT)—An arms race measured in microseconds
Principle in one sentence: Capturing fleeting price differences between different trading venues on an extremely short time scale (microsecond level).
Can ordinary people touch it: Absolutely not, and you shouldn't feel bad about it. This track requires: renting server rooms next to the exchange (professionally called "colocation"), customized network hardware, and execution systems at the dedicated chip level. This is not a gap that can be solved by "learning more Python"; it's a gap in physical distance and hardware investment. Even if you are a world-class mathematician, without that infrastructure, you still can't get to the table.
The mindset ordinary people should have: When you see the words "High-Frequency Trading," skip them immediately. Don't be envious; that is a completely different game. Your battlefield is in the Green and Yellow Light zones.
One Chart to Understand: Which One Should You Learn Now?
If you are a complete beginner, the suggested order is:
Step 1: Choose the simplest one from the Green Light zone (Momentum or Mean Reversion) and use a pre-built backtesting tool to personally run through a complete process—the focus is not on making money, but on understanding "how a strategy turns from an idea into a result."
Step 2: Once the Green Light zone is smooth, move to the Yellow Light zone—Factor Investing is the most worth learning because its academic foundation is the most solid and AI tools are the most mature.
Step 3: News Sentiment Trading and Machine Learning strategies can be tried as advanced attempts, but you must stick to the iron rule that "a backtested Sharpe over 3 should be suspected." Don't deceive yourself.
Red Light zone: No need to learn. Just know it exists and why ordinary people can't touch it.
Three Insights for Ordinary People
First, "complex" does not equal "valuable"; matching your resources is what's valuable.
Red Light strategies aren't at the end because they are "more advanced," but because they require resources (capital scale, hardware, speed) that ordinary people naturally lack. The first principle of choosing a strategy is not to choose the "most powerful" one, but the one that "matches your existing resources."
Second, what AI is doing is making "information processing," which used to be the most expensive part, cheap.
Among the nine schools, the biggest changes are in "News Sentiment Trading" and "Machine Learning Strategies"—they used to be institutional exclusives, but now, because of AI, ordinary people have the qualification to enter for the first time. This reminds us: any field that was once "monopolized because processing information was too expensive" is worth re-examining—AI may have already brought down the ticket price.
Third, "simple" strategies are actually a natural advantage for ordinary people.
In the section on statistical arbitrage, an unintuitive fact was mentioned: institutions can no longer "play" certain strategies because their capital scale is too large. Ordinary people have small capital and are more flexible in opportunities with limited capacity. Not everything is "the bigger the better"; in some tracks, being small is precisely the advantage.
Finally
Nine schools, three colors.
Green Light zone: you can start today. Yellow Light zone: worth serious investment in learning. Red Light zone: not your battlefield; don't feel any psychological burden.
True intelligence is not learning all nine schools, but clearly knowing which light to start under.
Those who stubbornly stick to high-frequency trading, fantasizing about competing with institutions using a laptop, are the ones truly wasting their talent—because they chose the wrong track, not because they lacked ability.
Start with one Green Light and go through it thoroughly; it's much faster than hesitating in front of nine lights at once.
