A holding drops twenty percent. What is the first reaction? For most people it is averaging down. Buy the dip, lower the cost basis, recover faster when it bounces. The impulse arrives fast and feels self-evident. But price movement, by itself, has never been a reason to add or trim. The real question is different: what is the money sitting on?
One system sorts position moves into three actions, with a single criterion — is the underlying getting better, or getting worse?
First, pull the weeds. Names that are weakening and overvalued get trimmed step by step. Not dumped at once — trimmed. Second, water the flowers. Lines of business that are booming and undervalued get added to continuously. Continuously — not in one lump. Third, build positions in a pyramid, in batches; going all-in in one shot is strictly off the table. As prices fall, keep adding — and the amounts get smaller as the fall deepens. One more rule across all three: no spreading money evenly across everything. Capital is allocated by quality, not by existing weight.
This sounds simple. Run it into instinct and each rule meets a strongest counterargument, each sounding reasonable. One at a time.
Averaging down lowers the cost basis; that is the plainest arithmetic there is. The deeper it falls, shouldn’t one buy more?
The arithmetic is right. The direction is wrong. Lowering the cost basis is real; the cost of averaging shows up elsewhere: the harder one averages down, the less room for error remains. In extreme volatility, no cushion left, one deep drop nails a full position. Rushing to load up is queuing up to die. The pyramid works the other way — the deeper the fall, the smaller each addition, keeping room for error for the case where the call is wrong. The instinct and the discipline twist in opposite directions: instinct buys more the lower it goes, discipline buys less. Where instinct feels best is precisely where risk is largest. Plenty of accounts die not on judgment but on averaging down, harder and harder, into a fall. So much for the instinct; back to the analysis.
If the game is long-term, hold and do nothing; adding and trimming repeatedly is itchy hands, working for the broker. Is that right?
Half right. Holding still is indeed a virtue — provided what is held deserves it. Pulling weeds is not high-frequency fiddling. The action targets exactly one kind of holding: the kind whose fundamentals have deteriorated and whose valuation has run up. Not yet deteriorated? Hands off. Deteriorated? No dawdling. Holding a flower still is correct. Holding a weed as a long-term position is the most expensive laziness. Between the hand that works for the broker and the hand that never moves lies a third hand — it moves when the situation demands, and idles otherwise. The craft of long-term investing is not stillness. It is knowing what deserves stillness.
Lurking in a valuation trough, at bottom, is bottom-fishing; guess wrong and one is trapped all the same. Does that objection stand?
It does not, because the trough method does not guess the turning point. The left-side contrarian standard reads: when the market falls systematically and a sector enters a historically cheap valuation trough, deploy in multiple slow batches, wait for capital to flow back, and no urgency to cash out early. The strength of this routine sits at both ends. The part guessed wrong gets diluted across batches, so the loss is bounded. The part guessed right never needs a forecast of when to exit — the capital comes back on its own. The goal was never to buy at the bottom. It is to buy low enough. That is the difference between guessing a single moment and spreading across a range of prices. One boundary remains: waiting for signals is right-side discipline, governing when to confirm. The trough is a left-side sizing standard, governing how to ambush. One manages confirmation, the other manages position. Each governs its own stretch; neither replaces the other.
Finally, the limits. This system does not solve everything. What to do after gains arrive — letting winners run, cashing out the principal in batches, keeping a zero-cost profit position for the long haul — another set of rules, covered elsewhere; one line here, no more. Position limits per stock or per sector are not here either. No stock picks, no price targets, no calls on the present market; the pyramid gives a direction, never a ratio.
What remains is one question. When things fall, the money has to go somewhere — onto the weeds, or onto the flowers?
