This post covers the daily session that closed on 2026-08-23 across the 20 USD-quoted crypto pairs tracked by Skia Paper using Coinbase Exchange daily candles. The table rendered alongside this piece shows each pair's close, session change, seven-day and thirty-day change, and where the close sat inside its own 30-day range.

The session carried a broad-advance classification. What that label means, what it cost to be involved in conditions like these, and what a single session cannot establish are each taken in turn below.

What the Session Did

On 2026-08-23, 19 of the 20 tracked pairs closed higher. The median session change was +1.74%, meaning roughly half the universe moved more than that and half moved less. The full table details every pair; the narrative here focuses on the shape of the session rather than reciting each row.

The strongest pair was IMX at +9.66%, closing at 97% of its own 30-day range — that is, within 3 percentage points of its 30-day high. NEAR was the second-strongest at +7.30%, and STX followed at +6.56%; both also sat in the upper portion of their 30-day ranges, at 78% and 92% respectively. The sole decliner was LINK at -0.91%, which despite the session loss still sat 77% up its 30-day range, reflecting the strength of the prior weeks.

Dispersion — the gap between the strongest and weakest pair — was 10.57 percentage points. That figure describes how unevenly the session's gains were distributed: most pairs clustered in a narrow band of roughly +0.88% to +4.09%, while IMX, NEAR, and STX sat well above the median. BTC and DOT were the quietest advancers, each closing +0.88% on the session, while BTC's range position of 90% reflects that the prior 30 days had already been strong.

Several pairs showed notable seven-day momentum entering this session: STX carried a seven-day gain of +91.80% into the close, XRP +53.18%, and FET +37.15%, all of which are figures from the table and describe the state of those pairs at the close of the 23rd rather than any forward condition.

Regime: What Broad-Advance Describes

A broad-advance label is applied when the overwhelming majority of the tracked universe closes higher in a single session. Here, 19 of 20 pairs advancing is close to the theoretical ceiling of breadth. The label is descriptive of the session that closed; it is not a category with predictive content.

What distinguishes this session from a dispersed regime is that the gains were widespread rather than concentrated in a handful of pairs pulling the average upward. What distinguishes it from a compressed session is that the median move of +1.74% and a dispersion of 10.57 percentage points indicate meaningful price movement rather than a flat, low-volatility day. A mixed session, by contrast, would show roughly equal numbers of advancers and decliners; that was not the case here.

The range-position column in the table gives additional texture: many pairs closed in the upper half of their 30-day ranges, and several — IMX at 97%, STX at 92%, BTC at 90%, ETH at 88% — closed near their 30-day highs. That means the session extended moves that had already been running for weeks in those pairs, rather than reversing a prior decline.

Cost Conditions on a Broad-Advance Day

When a session produces a median move of +1.74% and a leader at +9.66%, the round-trip cost of participation — exchange maker/taker fees, bid-ask spread, and any slippage on entry and exit — becomes a smaller fraction of the observed moves than it would on a compressed or near-flat day. On a session where the median pair moved +1.74% from open to close, a typical spot taker fee of roughly 0.50% to 0.60% round-trip (the precise rate depends on the tier and venue; Coinbase's published fee schedule is linked below) represents a meaningful but not dominant share of the median move. For the outliers — IMX at +9.66%, NEAR at +7.30% — the fee fraction was proportionally smaller still.

The caveat is execution timing. A daily candle records the open and close, but a participant who entered at a price above the open or exited below the close would have captured less than the candle implies. Slippage is not visible in daily candle data. On high-momentum days, spreads in thinner pairs can widen, and market orders during rapid moves can fill at prices materially different from the mid. The session change figures in the table describe the candle, not any particular execution.

The strategy dimension that matters here is holding period: moves of this size over a single session are large enough that a position held for the full session would have seen fees as a modest fraction of the gross move. Strategies that rely on frequency — many round trips intraday — face a different arithmetic, since each trip incurs the full round-trip cost regardless of the daily candle size.

What One Session Does Not Establish

A single session is a sample of one. The broad-advance label describes what happened on 2026-08-23; it carries no statistical weight about what the session that follows will look like. A sequence of broad-advance days is not implied by one, and the regime label is not a signal.

The range-position figures show that several pairs closed near their 30-day highs. That observation describes where prices sat at the close; it does not indicate whether those levels will hold, extend, or reverse. A pair at 97% of its 30-day range has, by definition, been higher only rarely in the prior 30 days — but that fact alone says nothing about the next session.

The seven-day figures for several pairs — STX at +91.80%, XRP at +53.18% — are large. Large prior moves are a description of what occurred, not a foundation for extrapolation. Momentum as a factor has documented structural failure modes: it tends to perform poorly precisely when it has been most visibly strong, because the pool of new buyers willing to pay higher prices eventually exhausts itself. That exhaustion is not visible in candle data before it happens.

Finally, breadth of 19 out of 20 is near the ceiling of what the 20-pair universe can produce. A session with near-maximum breadth is, by arithmetic, less likely to be followed by another session of equal breadth simply because there is little room to exceed it — but that observation is about the distribution of the statistic, not a forecast of direction.

The 2026-08-23 session closed with the broadest possible near-unanimous advance across the tracked universe; the shape of that session — its dispersion, its range positions, its cost fractions — is what the data recorded.

Note: This explains how a mechanism works. It is educational and is not financial, investment or trading advice, not a recommendation, and not a signal service. Any figure describes a session that has already closed and says nothing about what happens next. Trading cryptoassets risks losing money, including all of it.