从Dota2国际邀请赛抢票衍生出的黄牛经济学

Artificial intelligence stocks have begun to waver slightly, experiencing a selloff period in the first week of this month. The Nasdaq has fallen approximately 2%, wiping out around $500 billion in market value from top technology companies.

Palantir Technologies dropped nearly 8% despite beating Wall Street estimates and issuing strong guidance, highlighting growing investor concerns about stretched valuations in the AI sector.

Nvidia shares also fell roughly 4%, while the broader selloff extended to Asian markets, which experienced some of their sharpest declines since April.

Wall Street executives, including Morgan Stanley CEO Ted Pick and Goldman Sachs CEO David Solomon, warned of potential 10-20% drawdowns in equity markets over the coming year.

And Michael Burry, famous for predicting the 2008 housing crisis, recently revealed his $1.1 billion bet against both Nvidia and Palantir, further pushing the narrative that the AI rally may be overextended.

As we near 2026, the sentiment around AI is seemingly starting to shift, with investors beginning to seek evidence of tangible returns on the massive investments flowing into AI, rather than simply betting on future potential.

However, despite the recent turbulence, many are simply characterising this pullback as "healthy" profit-taking rather than a fundamental reassessment of AI's value.

Supreme Court Raises Doubts About Trump’s Tariffs

The US Supreme Court heard arguments overnight on the legality of President Donald Trump's "liberation day" tariffs, with judges from both sides of the political spectrum expressing scepticism about the presidential authority being claimed.

Trump has relied on a 1970s-era emergency law, the International Emergency Economic Powers Act (IEEPA), to impose sweeping tariffs on goods imported into the US.

At the centre of the case are two core questions: whether the IEEPA authorises these sweeping tariffs, and if so, whether Trump’s implementation is constitutional.

Chief Justice John Roberts and Justice Amy Coney Barrett indicated they may be inclined to strike down or curb the majority of the tariffs, while Justice Brett Kavanaugh questioned why no president before Trump had used this authority.

Prediction markets saw the probability of the court upholding the tariffs drop from 40% to 25% after the hearing.

The US government has collected $151 billion from customs duties in the second half of 2025 alone, a nearly 300% increase over the same period in 2024.

Should the court rule against the tariffs, potential refunds could reach approximately $100 billion.

The court has not indicated a date on which it will issue its final ruling, though the Trump administration has requested an expedited decision.

Shutdown Becomes Longest in US History

The US government shutdown entered its 36th day today, officially becoming the longest in history. It surpasses the previous 35-day record set during Trump's first term from December 2018 to January 2019.

The Senate has failed 14 times to advance spending legislation, falling short of the 60-vote supermajority by five votes in the most recent vote.

So far, approximately 670,000 federal employees have been furloughed, and 730,000 are currently working without pay. Over 1.3 million active-duty military personnel and 750,000 National Guard and reserve personnel are also working unpaid.

SNAP food stamp benefits ran out of funding on November 1 — something 42 million Americans rely on weekly. However, the Trump administration has committed to partial payments to subsidise the benefits, though delivery could take several weeks.

Flight disruptions have affected 3.2 million passengers, with staffing shortages hitting more than half of the nation's 30 major airports. Nearly 80% of New York's air traffic controllers are absent.

From a market perspective, each week of shutdown reduces GDP by approximately 0.1%. The Congressional Budget Office estimates the total cost of the shutdown will be between $7 billion and $14 billion, with the higher figure assuming an eight-week duration.

Consumer spending could drop by $30 billion if the eight-week duration is reached, according to White House economists, with potential GDP impacts of up to 2 percentage points total.

You've been using a 30-pip trailing stop for as long as you can remember. It feels professional, manageable and relatively safe. 

But during volatile sessions, you see your winners get stopped out prematurely, while low-volatility winners drift back and hit stops that are relatively too tight. 

Same 30 pips, different market contexts, but inconsistent in the protection of profit and overall results.

The Fixed-Pip Fallacy?

Traders gravitate toward fixed pip trailing stops because they feel concrete and calculable. The approach is easy to execute, readily automated through platforms like MetaTrader, and aligns with how most people naturally think about profit and loss. 

But this simplicity masks a fundamental problem. 

A twenty-five pip move in EURUSD during the London open represents an entirely different market event than the same move during the Asian session. The context matters, yet the fixed-pip approach treats them identically. 

This becomes even more problematic when you consider different currency pairs. GBPJPY might have an average true range of thirty pips on an hourly chart, while EURGBP shows only ten. The same trailing stop applied to both instruments ignores the reality that volatility varies dramatically across pairs.

Timeframe introduces yet another layer of complexity. Take AUDUSD as an example: a ten-pip move on a four-hour chart barely registers as meaningful price action, but on a five-minute chart it represents a significant swing. The fixed-pip method treats these scenarios as equivalent.

The natural response might be to use something more sophisticated, like an ATR multiple. This accounts for your chosen timeframe, the instrument's normal volatility, and even session differences. But it brings its own complications. 

When do you measure the ATR? Do you use the value at entry, knowing it might be distorted by sessional effects? Or do you make it dynamic, which becomes far more complex to implement in practice?

Perhaps there's another way forward that doesn't rely on abstract measures of volatility but instead responds directly to the movement of price in relation to the trade you're actually in—accounting for your lot size and the profit you've already captured. 

Maximum Give Back: The Percentage Approach

Instead of asking "how do I protect profit after fifty pips," ask "how do I protect profit after giving back a certain percentage of open gains." 

Consider a maximum give-back threshold of 40%. When your trade is up one hundred pips, the trailing stop activates if price retraces forty pips from peak, locking in a minimum of sixty pips. 

But when that same trade reaches two hundred fifty pips of profit, the stop adjusts, and now it activates at a one-hundred-pip pullback, securing at least one hundred fifty pips. The stop distance scales naturally with the magnitude of the win you're sitting on.

This creates a logical asymmetry that fixed pip approaches miss entirely. Small winners receive tighter protection. Big winners get room to breathe.

The approach adapts automatically to what the market is actually giving you in real time, without requiring you to predict anything in advance. 

You don't need to maintain a reference table where EURUSD gets thirty pips and GBPJPY gets sixty. You don't need different standards for different instruments at all. 

The same 40% logic works whether the average true range is high or low, whether volatility is expanding or contracting. It survives regime changes without requiring recalibration because it's responding to the trade itself rather than to abstract measures of what the instrument normally does.

The market tells you how much it's willing to move in your direction, and you protect that information proportionally. Nothing more complicated than that.

Key Parameters to Specify in Your System:

• Maximum Give Back Percent: 30-50% is typical, but is dependent on how much profit retracement you can tolerate. 
• Minimum Profit to Activate: In dollar amount or an ATR multiple form entry. This prevents premature exits on tiny winners, e.g., if it has moved 5 pips at 40% that would mean you are only locking in a 3-pip profit.
• Update Frequency: Potentially every bar. More frequent, but there may be issues if there is a limited ability to look at the market (if using some sort of automation, this could be programmed).

Is Maximum Giveback Always the Optimum Trail?

As with many approaches, results can be highly dependent on underlying market conditions. It is important to be balanced. 

The table below summarises some observations when maximum giveback has been used as part of automated exits.

The major difference isn’t likely to be an increased win rate. It is about keeping more of your runners during high-volatility price moves rather than donating them back to the market.

It may not always be the best approach, as different strategies often merit different exit approaches.  

There are two obvious scenarios where fixed pips may still be worth consideration.

• Very short-term scalping (sub-20 pip targets)
• News trading, where you want instant hard stops

Integrating Maximum Giveback With Your System 

You may have other complementary exit filters in place that you already use. Remember, the ideal is often a combination of exits, with whichever is triggered first.

There is no reason why this approach will not work well with approaches such as set stops, take profits and partial closes (where you simply use maximum Giveback in the remainder as well as time-based exits.

Final Thoughts

To use fixed-pip trailing stops irrespective of instrument pricing, volatility, timeframe, and sessional considerations is the trading equivalent of wearing the same jacket in summer and winter. 

Maximum Give Back trailing adjusts to the ‘market weather’. It won't make bad trades good, but it will stop you from cutting your best trades short just because your stop was designed for average conditions. 

The market doesn't trade in averages but has specific likely moves dependent on context. Your exits should not be average either. 

近期，全球投资者越来越明显地感受到一个信号——美元的钱，开始“紧”了。这似乎有些不对劲，美联储10月底刚刚降息，理论上市场流动性应该更宽松，资金成本更低，但现实却是美元资金反而愈发紧张，资产价格开始同步下挫。这究竟是怎么回事？先看一个最直接的指标：纽约联储的正回购操作（Repo）和逆回购操作（Reverse Repo）。一个基本的原理是：Repo 上升、Reverse Repo 下降 = 市场资金紧张；Repo 下降、Reverse Repo 上升 = 市场资金充裕。10月末，联储的正回购操作量一度达到 500亿美元，这一数值在以往往往出现在极端流动性紧张的时刻。更关键的是，这种“抽水”并非短暂的月末现象——进入11月初后，正回购仍在持续，这意味着美元市场的“钱紧”已经趋于常态化。

（纽联储官网）

资金紧张的根源在于：原先支撑流动性的“美元蓄水池”——ONRRP（超额逆回购工具）几乎被抽干了。

在过去两年，美联储在执行量化紧缩（QT）和财政部大量发债的同时，市场的流动性压力能通过ONRRP缓冲。但如今，这个高峰时期曾超过 2万亿美元 的“蓄水池”已经见底。

这意味着，财政部每多发一笔债、每回笼一笔资金，都将直接以消耗银行准备金为代价。

更雪上加霜的是，美国政府的“关门”也在加剧这一紧张。当前，美国政府停摆已进入第36天，创历史纪录，市场普遍预测将会持续更久。虽然财政部仍在继续发行美债，但政府支出却被迫收缩——这造成了所谓的“只收钱不花钱”的状态。结果是财政部的 TGA账户余额 从关门前的 8000亿美元 飙升至 1万亿美元。

这相当于财政部把市场的钱吸走，暂时“锁进保险柜”，导致金融系统内的资金流动性被进一步抽走。

（美国财政部官网）因此，美联储降息≠流动性宽松。流动性收紧最先体现在资产价格上。鲍威尔释放“鹰派降息”信号，美元走强压制非生息资产（美股、黄金、数字货币），BT币已跌破10万关键技术与心理支撑位。股票方面，尽管AMD、英伟达等AI概念股业绩亮眼，但估值已高企；一旦预期无法持续超越，抛压立即显现。美股在强劲的企业盈利支撑下，仍然出现广度恶化、板块分化严重的迹象——少数科技巨头拉动指数，而多数板块早已疲弱。近期，美股盘中波动明显加大。大盘科技股盘前集体走低，Palantir、AMD 等年内翻倍的热门股出现回吐，而小盘股指数罗素2000则因流动性担忧大幅下跌。多家华尔街机构已开始提示短期调整风险：摩根士丹利与高盛均警告未来12–24个月内市场或回调超10%；美国银行称当前AI板块和消费板块估值“已被透支”；Piper Sandler认为，六个月的牛市之后，市场正在寻找一次“健康的修正”。总结来说，盈利没问题，但资金太紧。过去一周，比特币跌破10万美元关键支撑位，创下年内第二大单日跌幅；以太坊也同步重挫超过10%。黄金从高点跌落到4000美金/盎司，苦苦挣扎在整数位上。市场的“贪婪指数”迅速转为“极度恐惧”。

钱荒之下，现金为王。数字货币和黄金都是“无息资产”，当美元利率仍高、而流动性又紧时，这类资产当然最先被抛售。短期看，美元的“钱紧”确实是一个政策性扰动。一旦美国政府重新开门，财政支出回流，TGA账户下行，流动性将得到缓解。市场也在等待未来两周的经济数据，以重新定价12月的降息预期。另外，本周三还需关注美最高法院关于特朗普政府依据《国际紧急经济权利法》发起全球关税是否合法的审理——如果判定不合法，这将是对总统权限边界的重新划定，美债、美股反而可能双双下跌。在这样的阶段，建议分层思考：

• 长期投资者：若已具备股、债、黄金等多元化配置，无需恐慌。市场的短期波动反而创造了优质资产的加仓窗口。
• 短线交易者：在美国政府重新开门前，适当对冲股票头寸（如配置防御性板块、买入波动率或保护性期权），以应对政策真空期与流动性扰动。
• 资产配置层面：流动性紧缩往往是“转折期信号”。美联储一旦释放宽松信号，资产价格将快速反弹。保持现金、等待机会，或许比急于抄底更重要。

当财政扩张与货币紧张的矛盾同时存在，美国金融市场正经历一个微妙的临界点。短期的钱确实“紧”了，但正如历史无数次验证的那样：每一次“钱荒”，最终都以更猛烈的放水收场。

免责声明：GO Markets 分析师或外部发言人提供的信息基于其独立分析或个人经验。所表达的观点或交易风格仅代表其个人；并不代表 GO Markets 的观点或立场。

联系方式：

墨尔本 03 8658 0603

悉尼    02 9188 0418

中国地区(中文）  400 120 8537中国地区(英文） +248 4 671 903

作者：

Christine Li | GO Markets 墨尔本中文部

‍