The Sunday Edge · where the market sits, one stock spotlight, one principle.
The position nobody decided to make this large.
o one wakes up and decides to put a fifth of a portfolio into a single stock. It happens one strong quarter at a time. A position bought at three percent of the account can drift to eight, then fifteen, then more, purely because the stock kept climbing while everything else moved at a normal pace.
The math is not a mistake. It is what winning looks like from the inside. The problem is that nobody chose the fifteen percent. The market chose it, one strong stretch after another, and most investors only notice once a single stock can move their whole net worth on an ordinary Tuesday.
The four gauges above measure today against market history back to 1950. A stretched reading shows up more often than most people expect, and on its own it has never been a reliable signal to buy or sell. Read it as a reference point, not a verdict.
Whatever the live numbers above show when this is opened, the same discipline was in force when it was written: positions sized by rule, rebalanced on a fixed calendar, no reaction trades in between. The market does not need anyone to predict it. It needs a plan for what happens when one position outgrows the size it was given.
NVDANVIDIA Corporation
NVIDIA is the cleanest real world example of size creep available anywhere in the market. A position built years ago as one holding among many has, for anyone who never trimmed it, grown to dwarf almost everything else they own. Nobody who bought it early planned for it to become the single largest driver of their net worth. The stock did that on its own.
The business behind the ticker is not confused about why this happened. NVIDIA does not just build graphics chips. It builds the computing layer most of the world's artificial intelligence models are trained and run on, wrapped in a software platform called CUDA that took a decade to build and that developers do not casually rewrite around. Switching away from it costs time, money, and risk most technology teams will not take on for an unproven alternative.
That moat shows up in results that keep surprising people who assumed the growth had to slow by now. It also comes with a concentration problem of its own. A small number of cloud computing giants account for an outsized share of NVIDIA's revenue, and several of them are openly building their own competing chips. A company this dependent on a handful of customers is exposed if even one of them succeeds at replacing what it buys.
The risk worth naming plainly: a stock that has already had one of the largest runs in market history does not need a bad headline to fall hard. It only needs growth that is merely good instead of extraordinary, because extraordinary is what the price already assumes. NVIDIA is currently held in both Core 20 and Tepper Tactical, and neither model owns it on a prediction about where the stock goes next. Apex Momentum, one of our five models, is built specifically for this problem: it ranks positions by rule and rebalances on a fixed calendar, so a winner that grows past its target weight gets trimmed back down on schedule rather than by feel.
“The most important rule of trading is to play great defense, not great offense.”
Paul Tudor Jones
There is a name for what happens after a position doubles or triples while the rest of the portfolio moves normally. Behavioral researchers call it the house money effect. Once a gain feels large enough, investors start treating it as the casino's money rather than their own, and they take risks with it they would never have taken with the original stake.
The bias does not show up as a decision. Nobody sits down and decides to bet a fifth of the account on one stock. It shows up as a series of non decisions: not trimming after the position doubles, not trimming after it doubles again, telling yourself the sizing is not the point because the business is still good.
The business being good is true and beside the question. A great company at fifteen percent of a portfolio and a great company at four percent are the same business carrying two very different amounts of risk to the person who owns it. One rough quarter tests the whole account. The other tests one position.
The defense Paul Tudor Jones is describing is not a prediction about any stock. It is a rule that does not care how good the story sounds. Pick a maximum weight for any single position before the position exists, and rebalance back to it on a schedule, never on a feeling about where the stock goes next.
Winning is not the hard part. Staying sized is.
Get the next one in your inbox.
The Sunday Edge lands every Sunday morning around 8:30 AM ET. Free. No urgency, no spam. Measured, useful.
Subscribe freeEducational research from Advising Alpha. We are a publisher under Section 202(a)(11)(D) of the Investment Advisers Act of 1940, not a registered investment adviser. Past performance does not guarantee future results. Full disclaimer at advisingalpha.com/disclaimer.