The amount of analysis available to you now is greater than at any time in human history.
And yet most people have less clarity about what’s really going on than they did five years ago.
What has changed is the scale. When analysis was expensive to produce, there was a natural filter. The people producing it must have known something, because the cost of getting it wrong was reputational and financial. Now this cost is practically zero. Anyone can generate a macro that looks like it came from a Goldman office in five minutes. The noise increases exponentially while the actual signal remains approximately constant.
The most insidious thing is that the noise no longer seems like noise. This looks like a signal. In the past, bad analysis was obviously bad. It is now refined, structured, uses the right terminology and cites the right data. The tools most people use to produce it are optimized for proper sound. Whether the result is actually correct is another question entirely.
Distinguishing the two is the whole game now. The same systems that flood markets with noise can be used to eliminate them. This is what I have spent the last two years proving – publicly, on X, with every call timestamped and nothing deleted, simultaneously on geopolitics, energy, macro, crypto and the broader markets.
The account grew from nothing to over 140,000 followers organically, without paid promotion or a name attached. Signal Core on Substack, which hosts the full forecasting operation, has become the best-selling crypto publication on the platform in nine months. In a market drowned in noise, the signal alone was enough.
The moment
The signal/noise problem happened at the worst possible time.
The next twelve months will reshape the financial, technological and geopolitical order more than the last decade combined. Digital assets are integrating into the traditional financial system at a pace that would have seemed impossible eighteen months ago. Regulatory frameworks that have been blocked for years are being rewritten in real time. AI is transforming the way capital is allocated. Geopolitical orders are realigning. Monetary policy is at an inflection point. The labor market is being restructured before our eyes.
These are fundamental changes, happening simultaneously and adding to each other. And that’s exactly when the ability to see clearly collapsed. There has never been more at stake and never less clarity about what is really happening.
The problem of convergence
It’s actually worse than a noise problem.
AI makes everyone converge on the same wrong answers simultaneously. When a thousand people use these tools to analyze the same event, they don’t get a thousand different perspectives. They get minor variations of the same output by default. Not only do the tools fail to produce a signal, they create false agreements.
Before AI, if five analysts said the same thing, it meant something. Now, if five hundred accounts say the same thing, that could just mean they all used the same tool.
What this looks like in practice
In January this year, the prevailing view was that a direct confrontation between the United States and Iran was unlikely. Diplomatic channels were still open. The market did not price in a significant risk of conflict. Oil was trading as if nothing had happened.
The structural picture tells a different story.
More than a month before the strikes began, indicators were already pointing toward a more likely than not confrontation. We reported this publicly on X on January 13 while the crowd was still rejecting the risk. When the strikes hit and oil nearly doubled, the move caught most of the market off guard. The signal was there. The crowd just wasn’t looking at him.
The entries we observed were not exotic. Public statements, internal economic pressure in Iran and the absence of certain de-escalation models. Anyone with access to the open Internet could see the same things. The advantage lies in synthesis: reading these inputs as a single converging system rather than as separate streams of information. This synthesis is the hardest part. The entries are just the entries. The bottleneck has never been technology. It’s how technology is used.
This is the model. The information was available. The tools to treat it were available. What was missing was the ability to read the signal before a crowd formed around a misreading.
The rare resource
Most people use AI to generate. Very few use it to see.
Signaling is when you can look at a situation that is confusing the entire market and see the structure underneath. It’s when you can hold a position that every flow is telling you to give up, and hold it anyway, because you can see something they can’t.
The challenge for most people is not generating a signal themselves. It’s about recognizing who actually has it. Most analysis is hedged to the point of insignificance – strategies to avoid liability disguised as analysis.
The old filter to overcome this problem was credentials. He no longer predicts who sees clearly. Many of the most important calls in recent years have been missed by traditional institutions and picked up by people working outside them. What matters now is whether anyone actually sees what’s going on – recognizing the patterns the crowd misses, naming what’s real before it becomes obvious, and being right often enough that it persists over time. Once you see clearly, you start operating on a different schedule than the rest of the market.
What comes next
We are entering an era where signal is the most valuable and least understood asset in the market. Investors, builders and dispatchers who understand this first will have a structural advantage that will grow over time. Those who continue to consume the flood without questioning it will continue to agree with the crowd. And the crowd will continue to make mistakes at the most important moments.
Finding rooms where the real signal still appears is becoming more and more difficult. Most sites that claim to aggregate market information only amplify what the models are already spewing.
The Miami Consensus 2026 is one of the few that still functions as a filter rather than an amplifier. The people who show up have skin in the game. Their disagreements are real. Their chords weren’t made according to the same five patterns that everyone else uses. This type of part is increasingly difficult to find elsewhere. That’s why I’ll be here to lead a small, invitation-only session on what large-scale signal mining actually looks like.
The advantage does not belong to who has the most information, the fastest tools, or the loudest platform.
It will belong to the one who sees clearly when everyone is drowned in noise.
It is currently the rarest resource on the markets.
And it’s only getting rarer.
