I keep a mental list of the trades that hurt the most, and almost none of them are from being wrong about direction. They are from being right about the old regime a few days too long. The trend had already broken. I just had not accepted it yet, so I kept sizing up dips that were no longer dips, and I kept getting stopped in that flat, sideways grind where nothing works and everything looks like a fakeout.
That flat sideways grind is the transition zone. It is the gap between one regime ending and the next one being legible. A trend dies, but a clean range has not formed. Volatility is elevated but directionless. And it is where a strange amount of the year's damage tends to happen, because the tools that made you money in the trend are exactly the tools that bleed you in the chop.
Why the middle is so expensive
A trending market is forgiving. You can enter late, size wrong, and manage sloppily, and the trend covers for you. A ranging market is also fine once you know it is a range, because you fade the edges and take small profits in the middle. Both regimes reward a specific, coherent playbook.
The transition has no coherent playbook, and that is the problem. It looks enough like the old trend that you keep trading the old trend. Pullbacks that used to get bought now stall. Breakouts that used to run now reverse into your face. Your win rate on the same setup quietly collapses, but because any single loss looks normal, you do not notice the regime has changed until you have strung four or five of them together.
The other trap is that the transition is genuinely ambiguous. You cannot know in real time whether you are seeing a healthy pause in an ongoing trend or the first evidence that the trend is done. Anyone who tells you they can call that cleanly is selling something. So the honest goal is not to predict the turn. It is to survive the ambiguity without giving back the good part.
Reading the turn early
You do not need a perfect signal. You need a handful of independent tells that, when they start clustering, mean the character of the market has shifted even if price has not confirmed it yet. The ones I actually watch:
- Failed continuation. In a healthy trend, pullbacks resolve in the trend direction. When two or three pullbacks in a row stall or reverse instead of continuing, the trend has lost its sponsor. This is usually the earliest tell.
- Momentum diverging from price. Price grinds to a marginal new extreme while your momentum read makes a lower high. Divergence alone is noise. Divergence plus failed continuation is a message.
- Volatility rising without progress. Ranges widen, daily moves get bigger, but net displacement over a week is roughly zero. That is chop energy, not trend energy.
- Correlations decoupling. The names or assets that moved together during the trend stop moving together. When the leaders and the laggards break formation, the flow that drove the regime is fading.
None of these is a trade trigger. They are a size trigger, which is the whole point of the next part.
Cut size before you are certain
The mistake I made for years was treating conviction as binary. Either I was in the trend at full size or I was flat. So I stayed full until I was sure the trend was over, and by the time I was sure, I had already eaten the transition.
The fix is to make size a dial, not a switch. The moment the tells start clustering, you cut. Not to zero, just meaningfully. Half, a third, whatever forces the position to stop dominating your P&L. You are buying yourself the ability to be wrong about the timing of the turn without it mattering much.
This feels bad in the moment, because most of the time the trend resumes and you left money on the table. That is the cost, and it is a good cost. You are paying a small, frequent tax to avoid the rare, large loss that comes from being fully sized into a regime that has already ended. Over a long enough run, trading the transition at reduced size is the difference between a drawdown you shrug off and one that changes how you trade for months.
Rules for staying flat through the chop
The hardest discipline is doing nothing once you have cut. The transition is boring and frustrating, and boredom makes you invent trades. A few rules that keep me honest:
- Do not run the old playbook on the new phase. If the trend setup that worked for weeks just failed twice, stop taking it. The setup did not get worse. The regime it belonged to is gone.
- Wait for a range to actually form before you fade it. A transition is not a range. It is a range that has not defined its edges yet. Fading a level that has been tested once is guessing. Let two or three touches build the boundary, then trade the boundary.
- Shrink your timeframe expectations, not just your size. Whatever the new regime is, it starts small. Take the first few trades in it as probes, sized to learn, not to profit.
- Write down what would tell you the trend is back. If you cut and the trend genuinely resumes, you want a pre-committed condition to re-add, so you are responding to price rather than to the itch of having missed some of the move.
The reason to define regimes explicitly, ideally in a scorecard you check rather than a feeling you carry, is that it takes the ego out of the call. When I built regime reads into Blockcircle, the surprise was not that they predicted turns better than I did. It was that having an external, unemotional read made it far easier to cut size when my gut still wanted to hold, because the decision was already made before the pain showed up.
The transition zone never fully goes away. Markets are always somewhere between one regime and the next, and the middle is always murky. What changes is how much of your capital you leave exposed to that murk. Trade the trend at full size, trade the range at full size, and treat the ambiguous stretch between them as the place to be small and patient. Most of the money in regime trading is not made in the transition. It is saved there.