The number that matters isn't the 30-day profit streak. It's $63,372.
That figure is the on-chain cost basis for Bitcoin's 1-to-3 month holder cohort. The 3-to-6 month cohort, meanwhile, sits at $73,190. Read both numbers slowly, because they describe something most coverage skipped entirely: the newest buyers acquired their coins cheaper than the buyers who came before them. That's an inversion, and it rewrites how you should read this dataset.
I pulled CryptoQuant's short-term holder snapshot the moment it crossed my terminal. The headline sold "30 consecutive days of partial profit." The subheading whispered "bull market reversal signal strengthening." My first instinct as a signal strategist is to distrust adjectives and interrogate structure. The structure said the opposite of the headline. In a genuine uptrend, cost basis ascends with time โ later entrants pay more, because price climbed and they chased it. Here the ladder runs backward. Newer coins are cheaper than older coins. That is not a market accelerating into a breakout. That is a market climbing out of a hole.
Context: what the data actually measures
Short-Term Holders โ STH โ are addresses holding Bitcoin for fewer than 155 days. They are the marginal bid: the speculators, the momentum chasers, the tourists who arrived with the last wave of volatility. Their aggregate cost basis is the average price at which their coins last moved on-chain, and it functions as the group's collective break-even line. When spot trades above it, the cohort sits in profit and is less likely to capitulate. When spot falls below it, they flip to loss en masse and selling pressure compounds. This is production-grade methodology โ the same skeleton Glassnode, CryptoQuant, and IntoTheBlock all run on. It is neither novel nor proprietary. It's plumbing, and plumbing is only interesting when it leaks.
It's leaking. CryptoQuant reports $168.2 billion in profit coins against $102.6 billion in loss coins across the STH cohort. That's a 62.1% profit share โ healthy-ish on the surface, but nowhere near the overwhelming dominance a true trend reversal demands. Then the tenure slices expose the buried story: 1โ3 months at $63,372, 3โ6 months at $73,190.
To understand why this inversion matters, you have to understand what the cohort represents. The 1โ3 month group bought during the summer lull โ a stretch of fear, low funding, and exhausted sentiment. The 3โ6 month group bought during the spring, when BTC flirted with $70k and the ETF narrative ran hot. The newest cohort is, in effect, a bet placed after the pain. The older cohort is a hangover from the euphoria. When the newer cohort sits in profit while the older one sits in loss, the market is telling you the pain already happened and the recovery is still early. That's a bottoming signature, not a breakout signature.
The 155-day threshold itself deserves a note. It's a heuristic borrowed from the historical average holding period of speculative capital, and it shifts meaningfully across cycles. In high-volatility regimes, coins that would have been STH frequently age into LTH faster, artificially shrinking the STH pool. That timing artifact can manufacture profit streaks without any price appreciation at all. It's the kind of thing that produces a green headline and an empty thesis.
There is also a provenance problem. The dataset's own summary cites "the longest consecutive profit streak since 2026" โ a year that, if this data reflects September 2024, hasn't happened yet. Map the cost basis backward and the numbers land almost exactly on 2024: BTC traded $65kโ$73k from March to June 2024, then slid to $55kโ$65k through the summer. The "2026" label is almost certainly a typo. But a typo inside a dataset is not a trivial thing. It's a crack in provenance, and provenance is the entire foundation of on-chain analysis.
The insight most people missed
A 30-day STH profit streak is a lagging confirmation, not a leading indicator. It reports what price already did, not what it will do next. The streak here was not produced by a surge of new demand. It was produced by two mechanical forces. First, a modest price bounce off a local low. Second โ and more subtly โ high-cost coins aging out of the STH bucket. Coins acquired at the April highs of roughly $70k-plus crossed the 155-day threshold in September and quietly reclassified as Long-Term Holders. They left the STH cohort as losers. The remaining STH pool therefore looked healthier by subtraction, not by appreciation. That's survivorship bias dressed up as strength.
Which brings the fragility into sharp focus. If spot sits just above $63,372, then the entire STH profit structure rests on a knife's edge. A 5% drawdown erases it. The 3โ6 month cohort โ carrying that $73,190 basis โ is already deeply underwater at current levels. They are the silent weight holding the other end of the scale. The word "partial" in "partial profit" is doing an enormous amount of work: it means roughly 38% of STH value is still locked in loss. That is not bull-market fuel. That is a bottoming process grinding through its final stages.
Compare this to what a genuine cycle reversal actually looks like on-chain. In real bull reversals, the 1โ3 month cost basis climbs above the 3โ6 and 6โ12 month bases, funding rates turn sustainably positive, exchange reserves drain, and stablecoin supply expands to fund the bid. None of those confirmations appear in this dataset. What appears is a single metric recovering from deeply negative territory โ a wounded cohort limping back to break-even. The absence of corroborating signals is itself a signal, and it's the one the headline chose not to mention.
The report also leans hard on a 2022 analogy: "similar conditions preceded the end of the bear market." I've watched this move before, and it's narrative anchoring, not analysis. No parameters are provided โ no MVRV comparison, no SOPR print, no funding-rate regime, no stablecoin-flow backdrop. An analogy without quantified inputs is not evidence. It's memory, and memory is the cheapest signal in the room.
Consider, too, how much of this is already priced. On-chain profit/loss distributions are public and refreshed in near real-time. I'd estimate 70% or more of the "STH recovered" narrative is already embedded in spot. That leaves the marginal information content thin โ thin enough that a single report rarely moves price more than 2%. Speed is currency, but precision is the vault, and the precision here points to a signal approaching its own expiry.
There's one more structural point worth flagging. Every figure in this dataset traces to a single provider. CryptoQuant is credible, but a data vendor publishing a bullish interpretation of its own numbers carries a subscription incentive. That isn't fraud โ it's alignment of interest, and it should discount the confidence you assign to the conclusion. Glassnode's STH metrics can diverge materially from CryptoQuant's because address clustering is proprietary and the 155-day threshold is a heuristic, not a law. One source is a sample. Cross-verified sources are a signal. The market doesn't care about your sentiment; it cares about your liquidity, and liquidity follows verified data, not conviction.
There's also the matter of what the report leaves out. No exchange netflow. No stablecoin supply. No funding-rate regime. No open-interest delta. No spot-volume confirmation. A bullish claim built on one metric, published by the vendor that sells that metric, is a claim that should arrive with a discount baked in. When the corroborating indicators are missing, the honest read isn't "bull confirmed." It's "insufficient evidence, monitor closely."
The contrarian angle nobody is publishing
Everyone is debating whether this is a bull reversal. The more useful question is: what happens if it's a trap? August 19 โ the date the report marks as the onset of net profitability โ may well have been a local low. Bounce off the low. Coins age out of the bucket. Charts print green. Narrative strengthens. And then the 3โ6 month cohort, anchored to that $73,190 basis, treats every rally as an exit ramp. Each bounce becomes a distribution event. This is the classic anatomy of a bear-market rally wearing a bull-market costume, and it is precisely the setup that punishes late longs.
The profit-taking language in the report betrays the tension. It describes holders "locking in gains while continuing to hold." Read that again. Locking in gains means selling. Holding means not selling. Both cannot be simultaneously true, and the phrasing conceals which force is actually winning. When a dataset resorts to contradictory descriptors, the underlying distribution is usually more divided than the summary admits.
The tell will be exchange inflow. If STH profit coins start moving to venues, the "holding" claim collapses within days. Watch funding rates too: if they stay persistently positive while spot stalls, leverage has gone one-sided and a long squeeze is loading in the background. The pivot is not a retreat, it is a recalibration โ and understanding whether price is healing or reversing is exactly that kind of recalibration. There is a difference between a cycle turning and a wound closing. The data here describes the wound.
One more dimension the sell-side framing omits: the transmission channel. STH profit states feed directly into how traders position derivatives. A cohort sitting in profit is read as "less likely to sell," which encourages leverage on the long side. That leverage, in turn, becomes fuel for the very drawdown that flips the STH cohort back into loss. The signal is reflexive. It changes the behavior it claims to describe. Anyone treating this metric as a passive readout is ignoring that the market is watching the same number and positioning accordingly.
The deeper risk isn't that the signal is wrong. It's that the signal is fragile and self-referential. STH profit states are among the most reactive on-chain readings available โ they move with spot in near lockstep, which means they flip with equal speed. A dataset that turns bullish on a bounce and bearish on a dip provides no forward edge. It's a mirror, not a window. And mirrors don't tell you where you're going; they tell you where you've been.
Takeaway: the lines that matter
Here is your forward frame, not a backward summary. Track three lines. First, spot versus $63,372 โ the STH break-even. Second, spot versus $73,190 โ the trapped cohort's ceiling. Third, exchange netflow. If spot holds above the first line and netflow stays flat while the second decays over time, the repair becomes a foundation. If spot slips below $63,372, the entire "reversal" narrative inverts overnight and 62% profit tilts back toward loss. The data doesn't tell you the bull is back. It tells you the market stopped bleeding. Those are different sentences, and the gap between them is where your money lives.
Compliance Check
This analysis concerns market data, not a token offering, so no securities question is implicated. The relevant regulatory lens is market structure: US spot ETF infrastructure now channels institutional flow into BTC through fully KYC'd vehicles, which means on-chain sentiment signals increasingly interact with regulated capital. Traders using STH metrics should note that ETF creation and redemption flows can distort the very cost-basis readings these models depend on. Verify provenance before you act. When a single vendor's bullish framing and a backward cost-basis curve disagree, the curve is the more honest witness.
