📊 Tuesday's Close: $628.05 vs a Score That Refuses to Move
Yesterday's post ended with the line: "These two facts will not both be true a month from now." Well, here we are one session later, and both facts got louder, not quieter.
QQQ closed at $628.05, up $10.66 on the day. That's another 1.73% on top of Monday's rip. The cushion above the EMA 70 (now $602.75) is $25.30 — ten days ago it was sitting at $9.44 when I wrote about consumer sentiment cratering. We've nearly tripled the buffer in a week and a half.
Today's score tape:
- 8:52 AM: Score 4.93 — pre-market, pinned to the floor
- 10:39 AM: Score 4.94, QQQ $623.64 — a tiny twitch higher, still Extreme Risk-Off
- 3:21 PM: Score 4.93, QQQ $627.68 — right back down into the close
Ref still holds at 4.94 (Apr 11, 8:49 AM). The 10:39 AM print to 4.94 wasn't even an escape attempt — it was a one-notch twitch that retraced by afternoon. The next upside exit is still 5.01 (ref 4.94 + 0.07, crossing into High Risk). There is no downside trigger — the score is already on the floor. Position: score says 100% SQQQ, EMA override says 100% QQQ, portfolio keeps riding.
📈 The Math of the Miss (And Why It's Still Winning)
Let me put actual numbers on the divergence because the abstraction is starting to lose its shape. The last time the score entered a bullish position was back on April 8 at 8:38 AM, when it printed 5.25 (Constructive) with QQQ at roughly $606. From there it walked the stairs down: Neutral, Cautious, High Risk, and finally Extreme Risk-Off on April 11 at 4.94. Every step of that descent was the score getting MORE defensive.
Over the exact same window, QQQ went from $606 to $628.05. That's +$22, or +3.6%, in four trading sessions. A pure-score portfolio — one that just listens to the raw signal without the EMA filter — would have gone from 80% long at $606 to 100% short at $610-ish, and then watched a 100% SQQQ position bleed while QQQ ripped $18 higher. That's not a miss. That's a face-plant. I'm not going to sugarcoat it: on the raw signal, this week has been ugly.
But the system isn't the raw signal. The system is the signal plus the trend filter. And the filter has been parked at "100% QQQ" the entire time. From $606 to $628, the actual portfolio has been long the entire +3.6% move. That's not a moral victory. That's real P&L. The score got schooled. The system did not. That distinction is the only thing that matters when you're the one writing the checks.
🚀 What Actually Moved the Tape Today
Tuesday was a continuation, not a catalyst day. No CPI, no Fed speakers worth quoting, no surprise headline out of the Gulf. Just more of what we've had for nine straight sessions: a market that has decided the Iran war, the inflation pipeline, the record-low sentiment print, and the stretched multiples all add up to "buy." Software led again. Semis came along. Small caps held their Monday bid. The breadth is actually decent — this is not a two-stock tape anymore — which is the one thing the bulls have going for them that they didn't have in late March.
Oil is still hanging around the upper $90s. The Strait is still a mess. None of the "reasons to be scared" went away today. The market simply refused to price them anymore. That's either exhaustion (the crowd ran out of reasons to sell and the path of least resistance is up) or conviction (the crowd genuinely believes the macro doom is priced in). I honestly can't tell which one it is, and I'm suspicious of anyone who says they can.
Here's the tell I'm watching: the score's own price-level math. The reason the score keeps dropping is that QQQ keeps crossing ABOVE key levels that subtract points. The further above the score's internal "fair value" QQQ gets, the more defensive the pure signal becomes. That's not a bug — that's how it's wired. But it means every dollar higher from here makes the pure signal look WORSE, even if the macro backdrop doesn't change at all. We're not going to get a clean 5.00+ score reading from a market melt-up. The only path back to a bullish pure signal is either a pullback in QQQ (price-level additions) or a material improvement in the macro data. Neither is in the forecast.
🛡️ Ten Days of Override, Ten Days of Being Right
I want to say this plainly because it's easy to get lost in the day-to-day narrative: the EMA override has now been the correct call for ten consecutive sessions. From April 3 (when the score first started sliding toward defensive) through today, every single day the override said "ignore the score, stay long," and every single day that was the right answer. QQQ is roughly +8% over that stretch. A pure-score portfolio would be deep in the red. The override-aware portfolio is green and getting greener.
This is the exact scenario the override was backtested for. The designers knew the macro engine would occasionally run hot in a confirmed uptrend — they looked at decades of data, saw that the trend wins most of these disagreements, and built the override to protect against that failure mode. This week is a textbook example of the filter working as designed. The raw model has an opinion. The trend disagrees. The framework defers to the trend until the trend breaks. Simple, mechanical, and — ten sessions in — unambiguously correct.
None of this means the pure signal is "wrong" in any philosophical sense. It's reading the macro honestly. The macro is just not what's driving price right now. And when macro stops driving price, you stop listening to the macro. That's why the filter exists.
🎯 My Take: The Bigger the Cushion, the Uglier the Deactivation
Here's the part that keeps me up at night, and I want to flag it before the tape lulls everyone to sleep: the $25.30 EMA cushion is a double-edged sword. On one hand, it's a huge buffer — QQQ would need to fall 4% in a straight line just to touch the EMA 70 at $602.75, let alone break it. That's a lot of room. On the other hand, the BIGGER the cushion gets, the LATER the override will deactivate if the market reverses, and the MORE violent the eventual regime change will be if it comes fast.
Think about it. If the override flips off at $602.75 when the pure signal is screaming 100% SQQQ, that transition goes from "long 100% QQQ" to "short 100% SQQQ" in a single session. Right now QQQ is at $628. If we got a fast -4% reversal — which is exactly the kind of move you see when melt-ups unwind — the portfolio would eat roughly $25 of downside on the QQQ leg before the override flips, and then potentially catch the next leg down in SQQQ. That's a best-case-plus-worst-case-stacked-together scenario, and the odds of it are non-zero when the macro is this hostile and the crowd is this complacent.
I'm not saying it's going to happen. I'm saying the framework has a pressure point that grows with every new high: a wider cushion means a slower override-off, which means more downside captured before the regime flips. The best outcome from here is boring — the macro data softens a little, the pure signal drifts back up to 5.00+, and we never have to test the override in reverse. The worst outcome is a one-day gap-down that makes everyone re-learn what "trend filter" means.
What I actually think is happening: the rally has legs into end-of-week as long as nothing explodes in the Gulf, and if we get a clean Fed speaker day or a soft PPI print we can probably tag $635. But the setup underneath is not what bull markets are supposed to look like. Record-low sentiment, a record-low score, oil at $99, and a melt-up anyway? That's a market running on short-covering and FOMO, not on fundamentals. Those rallies don't die of old age — they die on one headline, and nobody sees it coming until it's already happened.
⚠️ Bottom Line
Ten days. Ten wins for the override. A $25.30 cushion that's big enough to feel safe and big enough to make the eventual reversal genuinely dangerous. The pure score is at 4.93 and isn't going anywhere until either QQQ pulls back or the macro stops bleeding. Until then, stay long because the filter says so, and keep one eye on $602.75 — because the further the price runs from that line, the more of the rally you give back when it finally gets retested. The trade is still on. The conviction in the trade should not be.