Session Read 2026-09-25: 19 of 20 Pairs Higher

This post covers the completed daily session of 2026-09-25 across 20 USD-quoted crypto pairs sourced from Coinbase Exchange candles. The table rendered alongside this piece shows each pair's closing price, its open-to-close session change, its 7-day and 30-day percentage changes, and where the close sat inside the pair's own 30-day high-low range.

The session carried a broad-advance regime label — the coarsest description of what the data showed when the candles closed. The sections below work through what that label captures, what it costs to be present in conditions like these, and what a single session of this kind cannot establish.

What the Session Did

On 2026-09-25, 19 of the 20 tracked pairs closed higher than they opened. The median session change across the set was +4.38%, and the full table shows that most of the field clustered in a band broadly consistent with that figure. OP was the session's strongest mover at +11.26%, closing at a range position of 95% — meaning its close sat 95% of the way up its own 30-day high-low channel. APT (+7.91%), FET (+7.87%), and NEAR (+7.86%) followed, each sitting at range positions of 96%, 99%, and 92% respectively, indicating that their closes landed near the upper end of the territory each had traced over the prior 30 days.

The weakest pair was BTC, which closed -0.35% on the session — the only name in the set to finish lower. Despite that, BTC's range position of 74% places its close well above the midpoint of its 30-day range, meaning the session's single decliner was not sitting at a depressed level within its own recent history. ETH (+0.16%) and ATOM (+0.48%) rounded out the softer end of the field, while the layer-1 and infrastructure names dominated the upper half of the session table.

Dispersion — the gap between the strongest and weakest session change — came in at 11.61 percentage points. That figure is a mechanical property of the spread between OP's +11.26% and BTC's -0.35%; it describes how wide the field was on this particular day, not how wide it typically is.

Regime Description: What Broad-Advance Means Here

The broad-advance label is assigned when the preponderance of pairs close higher and the median session change is meaningfully positive. On 2026-09-25 both conditions were satisfied: 19 of 20 pairs advanced, and the median gain of +4.38% was not a marginal positive print. The label distinguishes this session from a mixed session — where breadth is split and gains and losses are roughly balanced — and from a compressed session, where nearly all pairs move but the magnitude of those moves is small enough that the field is tightly bunched near zero.

The 30-day figures visible in the table add context to the single-day read. Several pairs showed 30-day changes well in excess of their session move: NEAR's 30-day figure of +160.53%, FET's +46.70%, and APT's +52.63% indicate that the 2026-09-25 session arrived after an extended period of upward movement in parts of the field. Whether that context is relevant depends entirely on the question being asked; it is stated here as a description of where closes sat, not as a judgment about trajectory.

Execution Cost in a Session Like This

A broad-advance session with a median move of +4.38% and a field leader at +11.26% presents a specific cost picture. On Coinbase Exchange, maker fees for standard retail tiers sit at 0.40% per side (0.80% round-trip) as of the current published fee schedule; taker fees are higher. Against a session move of +4.38% at the median, an 0.80% round-trip represents roughly 18 basis points of the gross move per percentage point of gain — a non-trivial fraction but a smaller share of the move than it would be in a compressed session where the median might be ±0.50%.

For the pairs at the softer end of the field — ETH at +0.16% and ATOM at +0.48% — the round-trip cost structure looks materially different. A standard taker round-trip would consume the entirety of ETH's session move and more, meaning any position that entered and exited within the session on ETH would have faced a cost that exceeded the gross price change. Slippage compounds this: in fast-moving sessions, the bid-ask spread widens on smaller-cap names, and the executed price on a market order can diverge from the mid-price shown at order entry. The three mechanisms that allow a strategy to clear round-trip fees — higher frequency, larger size, or longer holding period — operate differently in each case, and the cost arithmetic shifts depending on which is in use.

What One Session Does Not Establish

A single broad-advance session is weak evidence about almost everything except what it directly measures: that 19 of 20 pairs closed higher on 2026-09-25, with a median gain of +4.38% and dispersion of 11.61 percentage points. It does not establish that the pattern will repeat, that the regime label will persist, or that pairs currently sitting at high range positions will remain there. The range-position column shows where each close sat within its own 30-day window; it describes the past 30 days and says nothing about the next period.

The regime classification itself is a label applied after the session closes. It describes the session that produced it; it is not a forecast of the session that follows. Broad-advance regimes end — sometimes immediately — and the structural failure of any analysis anchored to a single session label is that the label changes without warning. The 30-day figures for several pairs in this session are large enough that mean-reversion pressure is a plausible mechanical force, but "plausible" is not a prediction, and this piece does not make one.

The 2026-09-25 session was among the more uniform advances the 20-pair set can produce, with the breadth figure falling just one name short of a clean sweep — and that one name, BTC, declined by only -0.35% while sitting at 74% of its 30-day range.

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.