Markets are gripped by three distinct manias right now — in semiconductors, in space, and in options. The instinctive reaction to any mania is to recite the old wisdom that trees don't grow to the sky. That is true. But it is also dangerously incomplete, because manias can last far longer than anyone reasonably expects. The first discipline an investor must cultivate is holding both ideas at once: prices cannot rise forever, yet they can keep rising well past the point where any rational observer thought possible. This is precisely what separates a mania from a bubble, and why the distinction matters so much. In a mania, the market can climb higher than reason would allow before it finally turns.
The Rising Probability of a Correction
By careful analysis, the probability of a correction has risen considerably. Several independent signals point in the same direction.
The first is sentiment. It is extremely positive at the moment, and extreme positivity is a warning sign. It is not a precise instrument — it does not tell you that a correction begins today or tomorrow — but it tells you that a turning point is somewhere near.
The second is earnings expectations. Projections for the second quarter have just been revised upward. When expectations climb that high, they sometimes go unmet, and disappointment becomes its own catalyst for a pullback.
The third, and perhaps most structurally important, is concentration. AI-related stocks now make up roughly half of the S&P 500. When so much of an index rests on so narrow a base, the market becomes fragile. A stumble in a handful of names can cascade through the whole.
The fourth signal comes from the consumer. Consumer sentiment, as tracked since 1954, is now hitting its lowest levels ever recorded. The consumer continues to spend, which is encouraging on its surface, but the savings rate is falling and is heading toward negative territory. That dynamic cannot continue indefinitely; spending fueled by vanishing savings is borrowed time.
Finally, there is the political calendar. A midterm election lies ahead, and historically markets tend to dip in the run-up to midterms.
Taken together, these factors make the probability of a correction quite high — even as the underlying mania may carry prices higher for some time yet. The real question, then, is not whether a correction is coming, but what an investor should actually do about it.
The Answer Is Dynamic Hedging
The most reliable answer is a system of dynamic hedging. This is not an on-off switch, not a binary bet that the market will either rise or fall. Trying to call the exact top or bottom is a loser's game; nobody can time it consistently. Dynamic hedging is instead a slow, continuous process governed by proprietary signals that blend several forms of analysis.
A recent sequence illustrates how it works in practice. Cash was raised and the portfolio became heavily hedged just one day before a major geopolitical shock. Then, on March 30th — a day that only in hindsight turned out to be the low of the cycle — buying began, even though no one could have known at the time that it was the bottom. On April 1st and 2nd, profits were taken on the hedges. By April 7th, the call was to buy more. And now, with the market elevated, the process of building hedges has begun again.
The logic is straightforward. As the market rises and as the signals trigger, hedges are layered on progressively — more and more as conditions warrant. Long-term strategic positions are held throughout. Short-term positions are taken on as well, but with fairly close stops in the current environment. The result is a portfolio that continues to participate in the upside while being protected if a correction strikes. That is the entire game, and it is the discipline every investor ought to be thinking about.
The timing matters enormously. There is an old saying about closing the barn door after the horses have bolted — and that is exactly what most people do, reacting only after the damage is done. Right now the horses are still in the barn, content and feeding. The VIX is low, which means hedges are cheap — cheaper than they have been in a while. This is precisely the moment to build protection, not after the storm has already arrived.
Reading the Fed Realistically
A clear-eyed investor must also separate what the Fed should do from what it is likely to do. Looking purely at the data, the case could be made for raising rates. But an investor's job is not to set policy; it is to anticipate policy. The political reality is that there is strong pressure for lower rates, and a newly appointed and highly capable leader at the central bank cannot act alone. He must bring the rate-setting committee along with him, and that committee is unlikely to permit rate cuts. What he can probably achieve is preventing rate increases.
The widely discussed first step is the removal of the easing bias from the Fed's language. That can certainly happen — but it is a statement, not a substantive change in rates. The practical conclusion is that interest rates are unlikely to rise much from here. Perhaps a little more, but not a lot. This realistic read is why a hawkish-sounding chorus of officials and the data of recent weeks have not, in themselves, demanded a change in strategy.
Thinking in Zones, Not Lines
When it comes to where a correction might find its footing, the wise approach is to think in zones rather than single, precise support levels. It is an act of arrogance to claim you know the exact level at which the market will drop and then bounce. Over a long career, no one has been seen to succeed at that kind of precision. Zones, by contrast, acknowledge the genuine uncertainty of markets.
Three support zones are worth watching. The first is centered around 7,200, the second around 7,000, and the third around 6,800 and lower. Which zone the market reaches depends on the macroeconomic data. If the data stay as strong as they are now, a dip might reach only the first zone — probably toward the bottom of it. If the data show mild weakness, the second zone comes into play. If the data worsen beyond mild, into moderate deterioration, the third zone becomes the target.
Crucially, these zones are mapped out well ahead of time, not improvised after the fact. In the most recent episode, the March 30th low hit the exact lower band of the projected support zone and then bounced. That will not necessarily repeat, but it has happened often enough to validate the method.
The zones do not operate in isolation. They are paired with close monitoring of economic data to drive decisions. Suppose the market falls into the second support zone while the data have deteriorated only mildly, with no sign of further weakening — that is a signal to buy. But if the market hits that same second zone and the data appear set to slide from mild to moderate, then buying is held off and the hedges are maintained. The level alone does not dictate the action; the level combined with the trajectory of the economy does.
A Mosaic, Not a Single Lens
Underpinning all of this is a refusal to get hung up on any single form of analysis. The signals draw on macro data, fundamental data, quantitative data, and technical data, taking the best elements of each. Proprietary algorithms watch for the moment that might mark a bottom — and when that moment appears, as it did on April 1st and 2nd, the hedges are moved. Whether to then add more stocks and more exposure is treated as a separate decision, made on its own evidence.
This is the heart of the philosophy. Markets cannot be timed perfectly, so do not try. Instead, build protection while it is cheap and the skies are clear, scale that protection up as risk accumulates, and let a mosaic of data — not a single hunch — govern when to step in and when to hold back. Done patiently over years, this is how an investor stays in the upside of a mania while ensuring that the inevitable correction, whenever it comes, does no lasting harm.