[Red Envelope] Practical Tips No. 17: Trading System and Position Management Update

**Quickly hit like—it’s big gains nonstop; keep the questions coming—profits will soar! Hello everyone, I’m Wan Gu Chang Qing. **[Tao Xiba]
**

**
1. Preface
The market trend for A-shares recently can truly be described as making both those who hear it and those who see it feel hurt and cry—shareholders are living every day in the fear of thousands of stocks hitting downside. The win rate for short-term consecutive-limit-up breakouts is very low, and there are far more traps than opportunities that make money. In this situation, the first-limit-up arbitrage model is relatively suitable, because the entry position is low enough that even if it fails, the drawdown is relatively controllable.

In the earlier knowledge-post I shared, I already provided the formula for the first-limit-up arbitrage model, but the conditions are relatively strict. Although the success rate of the refreshed targets is relatively high, the number of targets that get refreshed drops drastically—you often can’t refresh any targets. To solve this problem, I made some optimizations and updates to the first-limit-up arbitrage formula (original name: bidding-price dragon). While doing my best to maintain the success rate, I increase the number of refreshed targets, thereby improving the profitability effect of this trading model.

2. The First-Limit-Up Arbitrage Model
The core idea of this model is to use the formula to filter for stocks that show good performance in today’s opening auction. These targets may trigger quantitative algorithmic crowding for shares or be selected by big funds. If they also rocket into a limit-up intraday, the next day they often have decent upside premium. The formula is: opening auction price increase greater than 3%, not hitting the limit-up during the auction, free-float market cap greater than 3.5 billion and less than 15 billion, the main board of Shanghai/Shenzhen is not a post-IPO stock and not ST, yesterday was not a limit-up, current price is greater than 95% of the highest price within the past three months, yesterday’s turnover rate is less than 30%

This formula has three key updates compared with the previous bidding-price dragon formula. First, the lower limit for auction price increase is reduced to 3%. This is because, based on my observations, some first-limit-up targets with good performance don’t open with too large an auction increase, so I lowered the requirement here. Second, the original condition of “a new high within the past six months” is changed to “current price is greater than 95% of the highest price within the past three months.” This change also lowers the standard. From the perspective of share-structure consolidation, “three-month new high” and “six-month new high” don’t make much difference. Third, the “new high” condition is changed to “current price is greater than 95% of the highest price within the past three months.” This is because the entry point for this model is on the limit-up board. With this change, once the stock hits limit-up, its price is basically at the new-high state—this both satisfies the new-high requirement and expands the range of targets being filtered.

After updating the formula, I’ll continue to explain the buy and sell points. First, this model has only one buy point: on the board. Buying early is not allowed. In the formula, the only stocks that have expectations are those that actually hit the board—if they don’t hit the board, they may very likely spike up and then bait people. As for the buy point on the board, everyone can choose to place orders manually or use conditional orders. Personally, I’m more in favor of using conditional orders. Because now it’s the quantitative era—human reaction is hard to keep up with algorithmic speed. And once you place too many limit-up orders, you can’t manage all of them. Also, manual operations involve emotional fluctuations, which are major issues. For how to use conditional orders, there are specific explanations in knowledge-post sixteen. If you’re interested, you can check it.

Next are the sell points. First case: if the stock hits the board and seals it, then the next day’s opening auction requires “stronger and even stronger”—meaning the next day’s auction price increase must be higher than the first day’s auction price increase; otherwise it’s considered not meeting expectations. For stocks that don’t meet expectations, after the auction ends, you need to place a price “fence” with a lower bound minus one tick to defend, to avoid a waterfall sell-off at the open. For how to use a price fence, if you’re not clear, you can look at knowledge-post eleven. For stocks that don’t meet expectations, even if the very first order at the open goes upward, any subsequent spike followed by a pullback is the point to fully exit—primarily to close quickly and settle.

Next, for the case where the stock meets expectations—i.e., the auction satisfies the “stronger and even stronger” conditions—then you also need to place a price fence with the lower bound minus one tick right after the auction ends to defend, again to avoid a waterfall sell-off at the open. If the very first order at the open goes upward and the price doesn’t fail to pull back, that can be treated as a sell point. If it breaks below the moving average, that’s the full exit point. Only if the stock quickly hits the board again on the second day or opens with a one-word limit-up would there be a reason to hold with a bigger mindset; otherwise, it’s always based on the idea of closing quickly. Because this model is a first-limit-up arbitrage model—its essence is to arbitrage.

Another scenario is when the stock explodes from the board (a “炸板”) on the day. Then the second day follows the same operation: after the auction ends, you place a price fence with the lower bound minus one tick for defense; a spike followed by a pullback is the point to fully exit. Also, because the formula conditions were relaxed, sometimes multiple targets appear. In that case, you need to select among them. Prefer those with auction guidance, higher auction price increase, and higher recognizability—the three conditions are arranged in that order, and then choose the best among the best.

3. Position Management
Position management is also a very important part. In the earlier knowledge-post eleven, I proposed a position management system based on the market’s up/down limit-hit ratios at the end of the auction. But it’s relatively complex, hard to execute in real practice, and the logic has some shortcomings. So this time, I propose a new position management system that is relatively concise and, logically, more reasonable. This position management system decides position size based on determinism. “Determinism” is the most valuable and hard-to-find thing in the stock market; using it as the standard for position sizing is absolutely appropriate. Theoretically, if there were 100% determinism, you could even put the whole position into the trade and be fine. Of course, in the stock market there is no 100% determinism—so position management becomes very important.

Based on determinism, position size is divided into three tiers: three-layer positions, five-layer positions, and seven-layer positions. For ultra-short-term trades, it’s basically a one-shot buy-sell. And since there is no 100% determinism in the stock market, the upper limit of position size per trade is set at seven layers, and the lower limit is set at three layers. If everyone is not even willing to go up to the three-layer position for a trade, it means you have zero confidence in that trade—you’re basically gambling. In that case, I suggest you don’t open the position. With that time, it’s better to buy a lottery ticket—the probability of making money might be higher.

These three tiers of position size correspond to three tiers of determinism. Then how should determinism be divided? Here I propose a standard: I believe determinism can be divided from three aspects: 1) the performance of the target stock, 2) the performance of the core stocks within the sector, 3) the overall market environment. Each condition met corresponds to one tier of position size. If all three conditions are met, that corresponds to the highest tier, the seven-layer position.

First is the performance of the target stock. Taking the operations on 4.3 as an example: the target stock that day was Tianjin Pharmac… In my early-morning “market musings,” I stated that the expectation for that day was for the stock to have an opening auction up more than 7 points, with an auction amount greater than 8B. Then Tianjin Pharmac… opened at 9.78 points, and the auction amount was 3B. This unquestionably satisfies the conditions—actually even beyond expectations. So this point is satisfied.

Next is the performance of the core stocks within the sector. At the time, besides the target stock Tianjin Pharmac…, there were three other core stocks in the sector: Meinonghua, Jiu’an Medical, and Wanbangde. All three opened lower that day. This condition is obviously not satisfied—core stocks in the sector basically had negative feedback. This means the sector’s price-action expectations won’t be good, and it will inevitably affect the target stock.

Then looking at the overall market environment. In my early-morning “market musings,” I had already stated that A-shares had a need to “step back with the second foot.” A low opening would be okay, but a high opening would basically be a “high-open then low-close.” So on that day, the overall market environment was undoubtedly not good. You can also tell from the auction. Besides PetroChina Capital, which had a high probability of blowing up the board, and Cu Wei Shares that added orders to “snipe” (steal chicken), the auction capital in the other directions basically wasn’t topping. With this situation, it would be very difficult for the overall market to do well that day.

Combining the above three conditions: using 4.3 and the example of the operation in the target stock Tianjin Pharmac…, the position size that this trade can justify is the three-layer position. Because besides the target stock meeting the conditions, the sector and the overall market do not meet the conditions—so position size can only be given as three layers. And here it’s important to note: the first point—performance of the target stock—is the most important. If this isn’t met, then even if the other two conditions are met, they are still not useful—you can’t determine position size using the latter two conditions when the first one hasn’t been met.

4. Closing Summary
This concludes the update content for this knowledge post. The core content has two parts: first, the update to the first-limit-up arbitrage model; second, the new position management method. First is the update to the first-limit-up arbitrage model: this is to solve the problem of the formula having overly high requirements, causing the targets often not to refresh. If a formula often can’t refresh targets, even if the success rate of the targets is relatively high, people probably won’t have the patience to use it. So I updated the trading model for this scenario.

Also, based on these two weeks of public operations, I can say my overall win rate is relatively high. In my impression, the number of failed times was relatively low, and the overall curve has been trending upward. But based on everyone’s feedback earlier, I clearly noticed a lack of position management ability among people—nobody knows in what circumstances they should give what position size. This can lead you to give a high position size in opportunities where determinism is low enough, resulting in overall losses. To solve this problem, I propose this position management system. Its core anchor is the determinism of the opportunity. And in future public operations, I will include the chosen position size, to help everyone better understand and put this position management system into practice.

5. Must-Read Every Day
Also, if everyone is interested in my trading system or has any questions, you can check the six posts below. The stock-picking formula, trading system, and answers to basic questions are all in them!

[Red packet] Knowledge Post sixteen, on being in cash (zero position) and trading models

https://www.tgb.cn/a/2ql81CLwIBn

[Red packet] Knowledge Post fifteen, Quantitative Trading System: Quantitative Trading

https://www.tgb.cn/a/2q9pLRC7uQN

[Red packet] Knowledge Post thirteen, The Path to Seek the Dao

https://www.tgb.cn/a/2pAw5lNyIZN

[Red packet] Knowledge Post eleven, Auction Trading System (more

https://www.tgb.cn/a/2pdneR1ZAWK

[Red packet] Knowledge Post ten, Software Trading System, little-known

https://www.tgb.cn/a/2p1TNOTUWWS

6. List of Thanks
Thanks to the friends below for cheering me on—thank you very much for your support.
@He Mu @zhuhuamin @Yu Mu Chen @H111c @Cheertt @hnlhyhx @不可一世小鳄鱼
Thanks to the friends below for their tips and for boosting/ramping—thank you very much for your support.

@Shi Guang Zhi Bi @Bu Wei Qing Kun @Yu Mu Chen @Xie Jian Zhou @Yi Yi

For those friends who recognize my trading approach—if you’re willing to deeply put into practice the three major trading systems I share, becoming a Gold Dust Club member would be a meaningful form of support. This is not only your trust in my content, but also a commitment to your own path as an investor—focusing on a proven method is often a key step toward stable profitability. As Gold Dust members, you will have a distinctive badge, making it easier for me to see your questions in the comments and prioritize interacting and answering. To join the Gold Dust Club, you need to accumulate 25,000 points of support, either by gradually accumulating or completing it in one go.

Gold Dust members, remember this moment—thank you for your strong support! @Beginner Crossing the Tribulation @I1獭兔 @Avatar and Nezha @Gu Hai Xiao Bai @Yu Mu Chen

If you’ve gained something by reading to here, please like and give support to the post. Your support is my biggest motivation for updates. Wishing you all a journey of uninterrupted success!