Folks, I have a confession to make…. I'm not perfect.

Shocking, I know.

Some of you may have seen a LinkedIn post I wrote a while back. I want to start there, because everything in it was true — I just haven’t told you the rest of the story.

In August 2024, Lululemon's stock had dropped about 50% over six months. Management had given vague guidance on growth in the men's category, and the market treated that as a red flag.

But when I looked past the headline number, I saw a company with a healthy balance sheet, exceptional margins, and returns on invested capital that didn't look like a business in decline. The concern was fair — but I wanted to verify it before deciding whether the market was right.

So I did something simple. I went to the mall.

Three separate visits to the Lululemon store, blending in as just another shopper. Each time, I counted the line: 10–15 people, roughly 30% of them men — exactly the demographic the market was worried about. Then I walked past the competitors. Their stores were empty.

That was my smoking gun. I bought in on August 12, 2024, at $250.26 per share.

Within five months, the stock was up 62%. The market had finally caught up to what the mall already knew.

I wrote that post to make a point: the market is inefficient, and you can find that inefficiency yourself if you're willing to do the unglamorous legwork. I even closed the post by saying the opportunity had passed — that LULU at its new price wasn't a buy anymore.

But, unfortunately, the story doesn't end there.

The first signal didn't come from a headline. It came from the price itself — and I missed it completely.

By June 2025, LULU had quietly given back its entire gain. The stock had fallen roughly 45% from its January peak, with no single news event driving it — just a slow, steady bleed that crossed back through my $250.26 cost basis and kept going. There was no alert. Nothing told me "the trade you bragged about six months ago is now underwater." I simply didn't notice.

Then, on June 25, 2025 — the same day LULU closed at $231.52, already below my entry price — Lululemon sued Costco over knockoff versions of its leggings.

At the time, I remember thinking: that's not a great sign.

If your moat is partly built on brand and pricing power, and a warehouse retailer is selling convincing dupes for a third of the price, that's not noise. That's a crack in the thesis. Dupes will always be a thing, but if LULU felt threatened enough by the dupes to sue Costco, that was a sign the dupes were working.

I noticed the lawsuit. I didn't notice that the position had already turned into a loser. And I acted on neither.

From there, it got worse. By September 2025, LULU hit a new 52-week low of $159.25 after a soft quarter and a guidance cut tied to tariffs.

As of last week, it's sitting at $114.23.

From my $250.26 cost basis, that's roughly a 54% loss — a loss that didn't just erase the gain I'd written about, it took the position deep into the red.

Measured against the January 2025 peak, it's closer to a 73% loss.

The research was right. The entry was right. What was missing wasn't one warning sign — it was a plan for what to do with either of the two I got.

I want to be honest about the "why," because I think it's more common than most people admit.

For the first signal — the slow slide back through breakeven — the honest answer is that I had no idea it was happening. No price alert, no periodic check-in, nothing that would surface "this position has quietly given back its entire gain" unless I happened to be looking at exactly the right moment. I wasn't.

For the second signal — the lawsuit — it wasn't conviction that kept me in. I didn't sit down, re-examine the news, and decide the thesis still held. I just... didn't look any further. Sometimes doing nothing is easier than making a decision under uncertainty, and even with the position already underwater, it didn't feel urgent enough to act on.

The deeper issue is the same in both cases. My entry process is genuinely good — I can point to specific, repeatable steps that led me to LULU, and to other winners. But I had no equivalent process for getting out. Two different signals, arriving on the same day, and neither one had anywhere to land.

If I had to boil this whole thing down to one sentence, it's this: I never defined an exit criteria.

That's the headline takeaway, and it's the one I'm carrying into the next trade. But sitting with this for a while, I think there are two smaller lessons sitting underneath it — and they're worth pulling apart, because together they explain why the exit criteria never got written in the first place.

1. LULU was a cigar butt, not a compounder.

LULU was never a long-term thesis for me — it was a mispricing thesis. The mall walk told me the market had overcorrected on a fundamentally fine business, and I bought the gap between price and value. That's a classic cigar-butt trade: pick it up, get one good puff, move on. The exit criteria for a cigar butt is baked into the thesis itself — once the price corrects back toward fair value, you're done, regardless of what happens next. I never wrote that down, so a trade that should have ended the moment the mispricing closed quietly turned into a hold-forever position, with none of the conviction that "hold forever" should require. A simple price-based trigger — "if this gives back X% of its gain, I'm out" — would have caught the first signal months before the lawsuit ever made headlines.

2. The circle-of-competence evidence is now empirical, not theoretical.

When I looked back across my full recommendation history, the pattern was stark. My biggest winners — Micron (+847%), Alphabet (+127%), Newmont (+144%), Diamondback (+43%), Occidental (+40%), Exxon (+39%) — sit almost entirely in tech, oil, and gold. The sectors where I have enough pattern recognition to read a headline and know whether it actually matters. My biggest losers — Novo Nordisk (-53%), Nike (-41%), Kraft Heinz (-32%), MercadoLibre (-30%), Constellation Brands (-20%), and now LULU — sit almost entirely outside it. This isn't a coincidence. It's proof that straying outside my circle of competence has a cost, and that cost is concentrated and well-documented.

Why this matters together: I bought LULU as a cigar butt, in a sector outside my circle of competence — which meant the first signal (the price-based one) had no trigger to catch it, and the second signal (the lawsuit) landed without the pattern recognition to tell me it mattered. Any one of these three things alone might have been survivable. All three together is how a 62% gain becomes a 73% loss.

Going forward, the rule is simple: define the exit at entry, not after. And if the position is outside my circle of competence, default to the cigar-butt exit — sell when the price-based thesis plays out, full stop, before a new and harder-to-read thesis has a chance to take its place.

Here's the strange twist in this story: the same pattern that got me into LULU — a macro narrative (inflation, consumer pullback) causing the market to oversell a fundamentally sound company — is a real, repeatable edge. It's part of a framework I've been building over the past several months, and it's the subject of the next edition.

And I'm not just writing about it this time. I'm using it. There's a live position on the board right now, entered using this exact framework, and you'll see the entry, the thesis, and — this time — the predefined exit triggers, written down before I need them.