Today's Wolf Wave Cycle Index = 0.222 (0 = bear bottom, 1 = bull top)
• Wolf Wave Cycle Index: https://wolfyxbt.github.io/wolfy-wave-index/
This bear market cycle will bottom at block height 971,250, which corresponds to approximately ≈ 2026/10/15. The date may be off by a few days depending on Bitcoin's block production speed.
The bull market for Bitcoin has ended, and we have now entered a bear market.
The peak of this bull market occurred on 2025/10/06, with the price of Bitcoin at 126,000. I also do not wish to speak negatively about my career, but according to the "four-year cycle theory," the price of Bitcoin has indeed peaked. I just want to convey this fact to more newcomers who have entered this cycle.
In my opinion, the "Bitcoin halving" has the greatest impact on the price of Bitcoin; it is much larger than any other event (macroeconomics, regulatory policies, geopolitical issues, wars, pandemics, etc.). The four-year cycle theory is a cyclical theory that revolves around Bitcoin halving, and it is currently the most reliable cyclical theory in the cryptocurrency space, without exception.
The core of the four-year cycle theory is: "Bitcoin halving" dominates the rhythm of Bitcoin's bull and bear transitions, dividing the fluctuations of a Bitcoin bull and bear cycle into 4 parts, which are:
Bull Beginning >> Halving >> Bull End >> Bear Market
Before the "halving," there will be a "Bull Beginning," After the "halving," there will be a "Bull End," The "halving day" will be interspersed in the "middle region" of the entire bull market, The duration of the "Bull Beginning" and "Bull End" is basically similar,
After the "Bull End," there will be a bear market lasting 1 year, and then we will wait for the next "halving day" to drive the next "Bull Beginning" market, and so on, ad infinitum.
From the diagram below, we can see that the "Bull End" of this round has already concluded, and we have officially entered the "Bear Market" zone. Even if the date I mentioned may not be absolutely accurate, there is no need for us to continue betting in this zone.
When you feel anxious, put down your phone, walk to the nearby convenience store to buy a bottle of cola, stand by the roadside and drink it. Watch the red light turn green, then turn back to red. See how the delivery rider rushes to deliver an order and runs the red light. Notice how the old man by the roadside picks up plastic bottles that can be sold for two yuan per kilogram next to the trash can—then look at yourself too.
We don’t need lots and lots of money to live. As long as we have enough for three meals, warm clothes to wear, a place to live, wake up each day with hope, and be able to do the things we love—then that’s enough.
Send a private message to someone you don’t know and you send just one line: “Are you there?”
Personally, I think that’s quite disrespectful—if not outright rude. If you’re going to message someone you don’t know, it’s best to lay out your purpose clearly in one go, along with the main point you want to express.
Say less and write less. If the other person usually uses English, translate it into English and send it over—do your best to reduce the time and effort cost for them to read your message.
If you send just “Are you there?”, then do you expect them to reply with something like “Yes, what is it?” and then you’re done, while they still have to wait for your response?
So every four years, Bitcoin’s price will gather a new wave of upward momentum. That momentum will inevitably become overheated, and after overheating, it will inevitably form a reasonable pullback. This repeats over and over, endlessly.
This is remarkably similar to the Kondratiev (K-wave) cycle.
Based on this idea, I invented a Bitcoin-specific “K-wave cycle,” named the “Wolfy Wave Index.” Here, you can observe this index in real time: https://wolfyxbt.github.io/wolfy-wave-index/
The closer the Wolfy Wave Index is to 1, the closer it is to the bull top; The closer the index value is to 0, the closer it is to the bear bottom.
In an algorithm that was just open-sourced by X, there’s a “same-author decay” mechanism.
What it means is: if you’ve already seen the first tweet from author A on your For You home page, then when author A posts their second tweet, that tweet’s score weight on your For You home page will start decaying by a factor of x0.5. Similarly, the third tweet will continue to decay, but the minimum decay is only down to x0.25.
So, in other words, when content appeal is the same, your first tweet of the day will get the best traffic; then your second tweet will begin to be attenuated.
From now on, we won’t need to use third-party tools to check whether our accounts and posts have been rate-limited. 𝕏 will roll out a tool called “Under the Hood.”
This tool lets us check ourselves whether our accounts and posts have been tagged with rate-limiting. The data also supports downloading.
It’s currently being piloted on a limited basis. It will be randomly made available first to users whose accounts have been registered for at least a year, who posted 10 or more times last month, and then it will be gradually opened up afterward. https://x.com/cb_doge/status/2087981530148966583/video/1
X has just open-sourced the recommendation algorithm for the For You homepage.
These parameters are probably worth taking a close look at:
Positive actions:
• Copy link & share: +20 points • Reply: +5 points (extra +15 points for mutual-follow users) • Quote: +5 points • Share via private message: +5 points • Follow the author: +4 points • Share: +2 points • Repost: +1 point • Like: +0.5 points • Click the post: +0.4 points • Open the link: +0.2 points • Open the image: +0.05 points • Open the video: +0.05 points • Valid video watch: +0.05 points • Time spent continuously: +0.004 points • Click the author profile: +0 points
Negative actions:
• Report: -234 points • Tap “Not interested in this post”: -43.2 points • Mute the author: -58.8 points • Block the author: -31.2 points • Swipe past without stopping: -0.02 points
One thing to note: something that’s especially easy to misunderstand. It’s not that the “reader” truly performing these actions will directly cause your post to be boosted or penalized. The boost/penalty logic here is executed based on “prediction.”
That is, the algorithm will use the reader’s past data to predict what actions that reader will take on your post. If the algorithm predicts the reader will take positive actions, then your post will receive a higher score on that reader’s For You homepage. The higher the score, the higher the priority the post will be pushed to that reader’s For You homepage.
• The specific GitHub page where the parameters are located: https://github.com/xai-org/x-algorithm/blob/main/home-mixer/params/param.rs
This animated video has already racked up nearly 19 million likes on Douyin.
So where exactly is the magic in this animated video? It hasn’t slowed down yet—likes are still growing by tens of thousands every day. It feels like it could be on track to break Douyin’s all-time highest likes record. I checked: the current Douyin record is 26.33 million 🤓
Please do cooking, return cooking, use sex; Please give blessings, return blessings, use violence; Please give law, return law, use fuck. Thank you everyone!
The day ChatGPT was released, I was doing homework in the study room.
I couldn’t write a single word, because during that period I’d earned my first bucket of money in the crypto world—I was completely consumed with trading coins. Even when I went to the cafeteria to eat, I had to keep scrolling on 𝕏.
So I studied my homework while scrolling on 𝕏,
until I came across a tweet posted by @sama about ChatGPT. I clicked into the website, dropped the title of my paper into the chat box, and read its answer—I was stunned. In the early hours of the study room, I was the only one left, and I kept staring at the screen, blurting “holy crap” over and over.
Within my understanding, I couldn’t imagine humans could create something like this.
Back then, GPT-3.5’s answers were relatively short. I broke my paper into 10 big questions, then split each of those 10 big questions into 10 smaller ones. Finally, I fed each question to GPT-3.5 to generate answers, and then stitched all the answers together into a complete paper. I submitted it to my teacher.
After that, I replicated the same method for all my assignments for the next few weeks, and I finished them all at once.
There’s a saying that goes: people can only earn wealth within their level of understanding.
On the day ChatGPT suddenly burst onto the scene, some people saw NVIDIA, others saw SanDisk, some saw Palantir, some saw Micron, some saw SK hynix.
But all I saw was the homework I had to write tomorrow.