七年之九：Uptober 究竟为比特币、以太坊、Solana 和 XRP 带来了什么

“Uptober”神话的真相：历史数据揭示四季度表现

每年10月1日，“Uptober”（十月牛市）这个词都会再次出现。这一说法认为，十月是加密货币市场中表现稳健强劲的月份，而第四季度则是强势季节。我们统计了过去九年（从9月30日收盘价至12月31日收盘价）中交易量最大的四种代币的数据，并单独分析了十月的表现。结果呈现出两极分化：对于比特币而言，该说法在十月成立，九年中七年上涨；对于整个第四季度而言，勉强成立；而对于XRP，则完全不成立。

需要明确的是，这并非预测。我们计算的是过去某种情况发生的频率，以及各年份之间的差异幅度。如果基于季节性规律做出决策，你实际上是在依据一个从未在任何单一年份完全实现的平均值行事。

“Uptober”的主张及其衡量方式

这一概念背后是一种季节性理论：由于资金流动、税务截止日期和市场预期的年度重复，某些月份的平均回报率往往高于其他月份。在统计学中，季节性意味着时间序列的值系统地依赖于日历，而非仅由偶然因素决定。

要使这一假设可检验，需采用两个指标，且顺序至关重要。首先是胜率（Hit Rate）：即在多少年内该时期结束时处于盈利状态？其次是Median (Median), which is the median of data for all years. Here, the median is more valuable than the average, because individual extreme years will arbitrarily increase or lower the average without affecting the median. This is common in the cryptocurrency market, and its specific impact will be shown below.

Bitcoin rose in October: Seven out of nine years, with a median of 10.9%

Looking at October alone, Bitcoin's trend is as clear as the hype says. From 2017 to 2025, there were seven years in which October closed above the opening; only 2018 and 2025 ended in declines. The median was positive 10.9%.

• 2017:+48.5%
• 2018:-4.6%
• 2019:+10.3%
• 2020:+28.1%
• 2021:+40.0%
• 2022:+5.5%
• 2023:+28.5%
• 2024:+10.9%
• 2025:-4.0%

The winning odds in seven of nine years are indeed convincing, but this is by no means a guarantee for the tenth year. Loss in two out of nine years means that on average, this theory goes against every four to five October months, and the most recent one did.

The performance of the entire fourth quarter was much weaker than in October alone

The situation reversed. If the measurement endpoint is postponed from October 31 to December 31, Bitcoin will remain profitable in only five of the nine years, and the median will shrink from positive 10.9% to positive 5.4%. In short: The gains that October often bring are often erased in November and December.

• 2017年：+219.5%
• 2018年：-44.1%
• 2019年：-13.7%
• 2020年：+168.9%
• 2021年：+5.4%
• 2022年：-14.9%
• 2023年：+56.8%
• 2024年：+47.5%
• 2025年：-23.3%

因此，若依据整个季度而非单月来调整决策，历史上来看相当于抛硬币的概率，略微偏向正面。关于本季度具体的关键日期，请参阅我们的第四季度加密日历概览。

为何44.7%的平均值具有误导性

九个比特币季度的算术平均值为正44.7%。这一数字常出现在季节性表格中，但若作为未来季度的预期价值，则毫无意义。该数值几乎完全由两年贡献：2017年（+219.5%）和2020年（+168.9%）。剔除这两年后，七年的平均值仅为约正2%。

这两个异常年份均紧随比特币减半之后（2016年和2020年），并处于资金大量流入的阶段。从两个特殊年份中提取统计数据，描述的是这些特殊年份，而非日历效应。正5.4%的中位数才是更诚实且平淡无奇的数据。

以太坊：六个强劲十月，但中位数仅为0.7%

在以太坊身上，可以看出仅凭胜率是多么不可靠。九年中六年的十月收盘为正值，听起来不错。然而，中位数仅为正0.7%，因为获胜月份的涨幅普遍微弱，而亏损月份的跌幅明显：2018年十月下跌15.2%，2025年下跌7.2%，相比之下，两个获胜年份的涨幅分别为0.6%和0.7%。

在整个季度内，关系发生反转：九年中五年盈利，但中位数达到正22.5%，为四种代币中最高。以太坊在第四季度盈利的频率较低，但一旦盈利，力度更强。关于两大龙头币在本季度的对比分析，请参阅我们的比特币与以太坊直接对比报告。

XRP打破模式：九年中仅三年盈利

在XRP身上，“Uptober”理论并不成立。十月仅三年结束于盈利状态，中位数为负0.9%。整个第四季度同样如此，中位数降至负12.9%。九年中六个季度以亏损告终，其中2019年至2022年连续四年亏损。

此处的平均值显示出截然不同的图景，即正111.9%，这纯粹展示了陷阱所在：该数值 practically 仅来源于2017年第四季度（+899.5%）和2024年（+240.1%）。若期待XRP具备季节性强势，你实际上是在用两年对抗六年。

索拉纳仅有五年历史，波动剧烈

Solana has only continued to trade on large exchanges since 2021, so the data series only covers five fourth quarters. Three of them ended up in profit. The huge fluctuations make any calculation of averages questionable: a 376.1% increase in the fourth quarter of 2023, compared with a 70.0% decline in the fourth quarter of 2022, which coincided with the FTX collapse.

Five observations are statistically insufficient to draw seasonal conclusions. This is not a criticism of the token, but the limitations of the data. For tokens with shorter history, quarterly statistics mainly describe major events that fall into the quarter.

Revelation for 2025: All four tokens recorded negative growth in the fourth quarter

The just-concluded fourth quarter violated this theory on all four tokens, and it was very obvious: Bitcoin fell 23.3%, Ethereum fell 28.4%, Solana fell 40.4%, and XRP fell 35.4%. In October 2025 itself, these four tokens are also negative.

This does not overturn the seasonal rule, because a single year can never disprove it. But it puts it in context: a recent case-and the one that everyone remembers-is a counterexample to slogans. When you read in October 2026 that "October was strong in history", last year's October was caught in the middle, and last year was not strong.

In early October 2026, the performance of the four tokens compared with previous years

10月3日，比特币约为84,900美元，以太坊约为2,684美元，索拉纳约为120美元，XRP约为1.49美元。与9月30日相比，季度初几天几乎没有太大变动：比特币高出约1.6%，索拉纳约1.5%，以太坊和XRP基本持平。

作为对比：在2017年和2021年那些强劲的十月里，比特币在最初几个交易日就已经大幅领先。平静的开局本身并不能说明月底的结果，但它表明，仅仅因为到达了某个日期，季节性行情并不会自动启动。各个市场的当前价格可直接在交易所查看，例如Bitstamp上的XRP市场。

季节性不是定律：统计数据如何看待单一年份

九个观测值是一个小样本。在7胜2负的胜率下，九个案例并不能清晰地分辨出背后是日历效应，还是短期序列中随机性产生的聚集现象。此外，这九年并非相互独立，因为它们都处于两个市场周期之内，而在周期内部，季度的走势往往相似。

在实践中，这意味着季节性统计数据是谦逊的理由，而非择时的依据。这类统计数据有助于抑制由口号引发的过高期望，但不能用于证明买入时机的合理性。任何定期定额投资的人都可以完全避开这个问题；具体实践请参考我们的储蓄计划提供商对比。

Holding period and taxation: Double consideration of the German calendar

For German investors, the fourth quarter also carries a second level of very specific calendar factors, which has nothing to do with seasonality. According to Article 23 of the German Income Tax Law, cryptocurrencies held as private assets are subject to the one-year holding period rule: if they are sold for a profit within one year after purchase, the profit is subject to personal tax rate. Once one year has passed, it will be tax-free. A tax exemption of € 1,000 per year applies to gains during the holding period, covering the total of all private disposal transactions; beyond this limit, the full amount is taxable.

This leads to a calculation that is more important than any seasonal consideration: If your purchase date falls in December of the previous year, selling in the weeks before the anniversary may cost you more than the amount you can save by hedging price fluctuations. Conversely, losses during the holding period can offset gains from other private disposal transactions in the same year. Tax advisers are the right person to consult for your specific situation, rather than relying on calendar statements.

Uptober: Your next step suggests

Replace slogans with your own data.

For the tokens you hold, check the winning odds and median, not the average. Bitcoin has risen in seven out of nine years, with a median of 10.9%;XRP has risen in three out of nine years, with a median of minus 0.9%. This difference is more decisive than slogans.

Check holding periods before quarterly arithmetic.

从交易历史中提取头寸的买入日期和买入价格，并注意哪些即将接近周年纪念日。税务截止日期是固定量，而季节性只是概率。

设定买入节奏而非买入日期。

如果你想在季度内加仓，固定金额和间隔比固定日期更好。不同提供商的成本差异显著，详见我们的加密交易所对比报告。