New Rules For Avoiding Falling Knives
Raising the bar for new trades during this momentum correction.
New Rules for Avoiding Falling Knives
Why we’re tightening entries—and how to navigate this tape whether you’re active or risk-averse
Monday’s session gave us one of those days where several stocks dropped 10–20% on no news at all. Others delivered strong earnings and still traded lower. If you’ve felt like the tape is punishing both good and bad stories lately, you’re not imagining it.
That kind of environment doesn’t mean we stop taking risk, but it does mean we need to be more selective about when we take it.
So in this post, I’m laying out a simple set of rules I’ll be using going forward to decide:
when to enter new trades, and
when (and whether) to add to existing ones.
You’ll also see where Portfolio Armor’s tools fit in if you’d prefer to limit your risk or avoid active trading altogether.
Before we get to the rules, a bit of context for newer readers.
What Portfolio Armor Actually Does
Portfolio Armor didn’t start as a stock-picking system.
It began as a hedging tool I built with the help of a post-doc in Princeton’s financial engineering program. Together, we developed an algorithm that scans the options market to find the least expensive way to protect a stock or ETF against a large decline—an optimal static hedge.
Later, I discovered that the hedging algorithm revealed something else:
It captured options-market sentiment.
Large changes in optimal hedging cost often reflected changes in how the options market viewed a name’s risk and reward. Once I paired that with historical price behavior, we had method for estimating 6-month returns for optionable stocks and ETFs.
The top ten names on that ranking turned out to generate significant alpha.
Track record (real, forward returns):
Since December 2022, we 6-month performance data for 125 weekly top ten names cohorts.
Average 6-month Portfolio Armor Top Ten Names return: +19.90%
Average 6-month SPY return: +10.06%
No backtests—all live, forward performance. You can see the charts for all 125 top names cohorts (and what the top names were in each cohort) here.
Today Portfolio Armor does three things:
Finds optimal hedges to limit downside risk using the cheapest reliable protection. That tool is available on the Portfolio Armor website and iPhone app.
Identifies high-potential names using options-derived sentiment blended with historical price behavior.
Constructs concentrated, hedged portfolios for investors who want competitive returns with strictly limited drawdowns—a true set-and-forget solution.
In this post, we’re focusing on #2:
How we decide when to enter and re-enter individual names.
On The Portfolio Armor Substack, I share the trades I make based on those rankings, usually structured as multi-leg options structures with defined downside and asymmetric upside.
Now let’s talk about why we’re adding rules.
Why tighten now?
Two weeks ago, in The Bears Are Back, I wrote about being more selective with new entries—especially in biotech, where even solid names were getting sold off. One change I made then was avoiding new pre-earnings trades in biotech names for the duration of this selloff.
Monday’s action broadened that concern beyond biotech.
When multiple stocks fall 10–20% on no news, or when a company reports strong earnings and sells off anyway, it’s a sign that the tape is not receptive. In that kind of environment, entries matter more.
That’s why this post introduces clear rules for:
new trades outside the Portfolio Armor Top Ten
and follow-on trades in names that have moved against us
This lets us stay active without forcing trades the market isn’t rewarding.
1. Rules for New Trades
A. Portfolio Armor Top Names (PA Top Ten)
These remain our strongest pool of ideas.
A sharp pullback often improves a stock’s ranking—because the system evaluates upside potential relative to historical downside. After previous pullbacks, PA’s top names have continued to perform well. For example, the Top Ten from March 20, 2025 returned +85.55% over the next six months, versus +17.27% for SPY.
Rule:
If a stock appears in today’s PA Top Ten →
eligible for a new trade, even in a messy tape.
I’ll still be cautious about pre-earnings trades in most sectors, but PA Top Names get more leeway.
B. Non-PA Names
Not all good ideas come from the algorithm.
Some come from:
Market Watchers: analysts/traders I follow on X who specialize in mispriced or overlooked stocks
Multibaggers: a subset of the Market Watchers accounts with documented, >100% gainers this year.
Chartmill: I run a number of fundamental and technical screens there.
These can be excellent trades—but in this environment, being early is expensive.
So for any non-PA name, we now require technical confirmation before starting a new trade.
Rules for new non-PA trades:
Chartmill Set-Up Rating ≥ 6
Indicates the stock has stopped falling in a straight line and is compressing.RSI between 30 and 70
Avoids buying breakdowns (<30) and overextended bounces (>70).
These filters should give us higher-quality entries instead of trying to guess where the bottom is.
2. Rules for Follow-On Trades (Adds)
If we already have a position and the stock has fallen, we need more than “it’s down a lot” to justify adding.
A. If the stock is a PA Top Name
Add only if:
It appears in today’s PA Top Ten again.
Because the ranking already incorporates drawdowns, a stock that stays in the Top Ten after a decline usually has strong forward return potential.
B. If the stock is NOT a PA Top Name
We require both:
RSI crosses up through 30 from below, and
Chartmill Set-Up Rating ≥ 5
This injects discipline into adds rather than averaging down blindly.
3. Where “New Blue Chips” Fit In
Some names —strong-execution, obvious-moat compounders—are tempting to add on weakness. But unless they’re in today’s PA Top Ten, they follow the same rules as any other non-PA name:
Set-Up ≥ 6
RSI 30–70
Simple, objective, and consistent.
4. If You’re Not an Active Trader
Not everyone wants to actively trade. If you’re not an active trader, but want to limit your downside risk, that’s where the two tools I mentioned above can help:
A) Hedge individual positions
You can use Portfolio Armor’s hedging tool (on the website or the iPhone app) to limit your downside on individual stocks or ETFs. It’s a straightforward way to stay invested while strictly limiting risk.
B) Let the system build a hedged portfolio for you
The hedged portfolio method is a set-and-forget solution for risk-averse investors:
You tell the tool how much you want to invest
You tell it how much downside you’re willing to risk
It builds a fully hedged, six-month portfolio for you automatically
It’s designed for people who want solid returns without stress. You can see historical returns here at different risk tolerances. Since 2022, portfolios hedged against declines of 20% or more have beaten the market, on average. You can construct portfolios hedged against declines as small as 2%, but of course your returns are going to be more modest.
In A Nutshell
This is a market where entries and structure matter.
For new trades:
PA Top Names → green light
Non-PA names → require Set-Up ≥ 6 + RSI 30–70
For follow-on trades:
PA Top Names → only if back in last night’s Top Ten
Non-PA names → RSI cross up through 30 + Set-Up ≥ 6
These rules don’t eliminate opportunity—they give us a clearer sense of when to take it.
When momentum turns up again, we’ll be positioned with better entries and better asymmetry.
This is great stuff. Thanks David for putting all this together!
One thing I am curious about is setting stops using volatility of each individual stock as a factor to widen or narrow the actual trailing stop percent (using something like 2 x ATR to approximate) so as to reduce getting stopped out?
A lot of the Top 10 names seem to be much more volatile than the general stock market. On the downside this would make entry more complicated and potentially means setting the stop so wide that bigger losses happen if things go way down...