[Red Envelope] If Enlightenment is like climbing a mountain! In the era of quantification, follow and master quantification!

To start, let’s answer a few questions [Taogu Ba]

Currently, what percentage of the market trading volume is dominated by quantitative strategies? Over 50%? Or less?

If you’re unsure, I recommend following the live broadcast tonight and learning carefully!

In the future, what will be the trend of quantitative funds in the market? Will they become more and more prevalent? Or less?

If you can’t answer, I suggest the same—watch the live broadcast tonight!

If quantitative funds continue to grow in the future, will your trading become more difficult? Or easier?

To understand these things, we first need to know where the boundaries of quantitative ability lie.

I used to come from a traditional trading background and later worked at one of the top private equity firms in China.

I should say I am one of the most knowledgeable about quantitative strategies, both from the perspective of teammates and competitors.

Building on that, let’s talk about the current market’s quantitative programs.

First, what is quantitative trading?

It’s basically a piece of code that, based on a certain formula, selects stocks and then executes buy or sell operations.

This is the most basic form of quantitative program.

The original purpose of this code was similar to today’s AI—to replace simple, repetitive manual labor.

Yes, manual labor.

If a trader follows a set process without just clicking a button to sell, it’s actually a test of their speed.

The core idea of quantitative trading is to ensure transactions are executed quickly!

That’s why we often see some large, seemingly outrageous stocks, along with small stocks, showing extreme intraday movements!

The main reason is that, because quantitative orders are placed so quickly, opposing traders don’t have time to react.

The only ones likely to catch a quantitative order are other quants!

So, I’ve always disliked certain teachers who say they pick stocks based on a few conditions and then just trade, ending the day.

If you really learn this, under those conditions, can your trading speed match that of a quant?

If your costs are higher than a quant’s and you sell more slowly, who will survive in the market?

This is just the most basic level of a quantitative program.

Over time, quantitative strategies naturally evolve.

The most basic aspect is capturing news, public opinion, and even information on futures, options, and US stocks, then applying that to trading.

For example, my public area has always been a hotbed for quantitative activity.

Because there’s no anti-crawling mechanism, quants can easily access it.

Quantitative strategies are different from humans—they can’t understand text and convert it into output; they can only copy and paste to gather information.

If you can’t do that, you can’t do data scraping.

So, quants like to gather data from my public area because I’m capable enough.

But they can only gather from my public area.

Of course, that’s about the extent of what quantitative strategies can do.

Currently, most domestic quantitative models lack decision-making ability.

For example, after a black swan event on Monday, the market still showed a bullish trend!

But I predicted on Monday, using micro-cap stocks as an example, that the market would fall for three days!

Because micro-cap stocks truly reflect market sentiment. The index, on the other hand, was supported by the “Wang Wang Team,” so Monday’s performance was unexpectedly strong.

Why?

Why did the market continue to fall after the Wang Wang Team’s efforts on Monday?

Is it because their support wasn’t strong enough?

Actually, it’s related to quant strategies.

How would a quant know the market will fall for several days?

But quant strategies do consider historical data!

Because, based on typical emotional cycles, a pullback usually lasts about three days.

Starting from 2024, this pullback cycle has been extended, due to some teachers’ strong resistance, into phases of pullback, recovery, and another pullback, but the cycle remains roughly three days.

So, even extreme market reactions caused by external black swans are still considered to last three days by quant models!

Applause if you understand!

On Wednesday and Thursday, the market was at this level. Looking back now, it’s still okay, right?

Because quantitative strategies are based on algorithms, sometimes big-picture logic is easier to guess than individual decisions.

At least, quant strategies won’t do something like a certain trader’s ex-girlfriend getting married and then crashing the market…

But during trading, the harvesting and betting of quants are very hard to predict!

This leads us to the next point:

Quant strategies also face opposition and harvesting relationships.

Initially, we said that quant strategies are just buy and sell programs that execute based on certain conditions.

So I’ve always said that the fixed dragon-head approach isn’t fixed!

It’s not about what kind of leader you want to create, but what kind of leader the market lacks.

Only then can the leader strategy continue to hold a place in this market!

But quant programs need to be fixed because machines have limited thinking ability.

They only have yes or no answers. If they lose money, they lose money. As long as the probability favors profit, that’s enough.

Therefore, quant strategies must be designed with specific rules.

I’ve been fortunate to participate in some of this.

Back then, the market was mostly driven by chart-based funds—those that rely on technical patterns.

Quant strategies also liked to draw charts, then sell these patterns to retail traders who liked to follow them.

This is about mimicking shapes rather than exact forms, emphasizing momentum over appearance.

Because quant strategies can’t understand, for example, in a very poor or all-encompassing market, the idea of funds banding together to create leaders.

Before September 24, 2024, when China Communications and Shenzhen Huasheng were in a coalition, it was very profitable for traders. But for quant strategies, even the best domestic models like Fantasia were experiencing losses during that period!

That’s a long story, but the point is the same.

Whether it’s another quant fund or a trader coalition, different objectives lead to strategies failing or becoming invalid during intraday trading.

In such cases, the market conditions are unpredictable.

Even the programmers designing quant models don’t know what kind of funds they’ll encounter or what rare outcomes might occur (since quant is all about probability).

So, they only focus on big probabilities, ignoring small ones.

The more we rely on quant, the more important daily K-line battles become.

This year, we’ll try some experimental chart modifications in many areas.

Returning to ourselves, how can we resist, accept, or manage quant strategies?

Mainly because quant strategies don’t think.

For example, earlier this year, Seedance 2.0 exploded. Since it happened over the weekend, the stocks that performed well on Monday were mostly top-performing stocks with a straight-up opening.

Impossible to buy in.

Quant strategies, leveraging channels, volume advantages, and speed, placed orders early to top with a straight-up opening (market transactions prioritize large orders and time).

How do we deal with funds that can’t get in?

I roughly divide it into a few parts:

The purest are those who choose to arbitrage from the back rows—these are traders with limited cognition who can’t do better.

They hold no positions.

Then there are those who look for stocks that hit a 10cm top and then buy 20cm arbitrage.

This strategy can earn high marks in this round because, during this period, retail traders using this approach did make money! But if the holding period is shorter, this strategy might not work as well.

What did we do? Since ByteDance hadn’t gone public, we directly became a upstream provider of computing power for ByteDance—Dongyangguang!

This is an unavoidable market component. Once it gains momentum, the market naturally moves in that direction.

Institutional funds, with their advantage in information channels, pre-positioned early, waiting for quant to lift them. That’s the most comfortable way, but hard to replicate.

Finally, there are those small writers, who modify entries to induce quant to buy certain stocks.

This is comfortable but highly unethical and legally risky.

Initially, they might not have enough resources, but after multiple times, just imagine how much energy is behind hundreds of billions in quant funds?

And then there’s oil—this round, oil was also artificially driven up, baiting quant strategies and putting them on the fire.

From a physical perspective, domestic oil prices are adjusted periodically, roughly every ten days.

Logically, the conflict might end before the oil price is even adjusted.

Like the Russia-Ukraine situation last time.

Last time, the oil and gas market didn’t heat up, but this time it did!

What changed between the two conflicts?

Only one thing:

Some institutional funds, having experienced the last Russia-Ukraine conflict, learned their lesson and deliberately baited quant strategies into action.

Of course, as long as quant strategies engaged in oil and gas, they probably made money, so they don’t mind.

But this gives us a new idea:

Besides igniting and baiting quant, you also need logic and storytelling!

No matter if this logic can be practically implemented, at least it should have some basis—like the current oil and gas situation.

So, research reports are still essential!

We experienced similar trading last week:

Domestic Ascend ecosystem fermentation, liquid cooling in short supply, and a surge in Lan

In reality, Lan’s solution isn’t very cost-effective.

How much can actually be implemented is questionable.

But the profit comes from earning more than companies like Sichuan Runsheng, which dominates Ascend.

Of course, this week, our trading wasn’t very good.

Mainly because on Monday I saw the risk but was confident I could handle it, so I didn’t open new positions and just cleared everything.

In the following days, the rhythm was off, and we started to retrace.

But on Friday, I realized—all the stocks I bought seemed good?

The rhythm was wrong, and I got hit.

But at least the overall idea was correct.

Next time I encounter a similar situation, I’ll sell on Monday or Tuesday and buy the dip on Wednesday or Thursday—different results will follow!

The key is: when can I develop the ability to predict and grasp the market?