Session Read 2026-09-07: 11 of 20 Pairs Higher
This piece covers the daily session that closed on 2026-09-07 across the 20 USD-quoted crypto pairs tracked by Skia Paper's session desk, sourced from Coinbase Exchange daily candles. The table rendered alongside this post carries the six standard columns: pair, close in USD, session change, 7-day change, 30-day change, and range position.
The session was classified as dispersed — a label applied after the close to describe what the data showed, not a characterisation of what follows. Eleven of the 20 pairs closed higher; nine closed lower. The median session change landed at +0.26%, a figure close enough to flat that the headline breadth count tells more of the story than the median does.
A ready-made Notion finance tracker for spending, bills, debt payoff and investments. Set it up in minutes.
What the Session Did
The defining feature of the 2026-09-07 session was the distance between its extremes. ICP gained +9.24% on the session and DOT added +8.57%, while at the other end NEAR shed -5.39% and LINK fell -3.69%. The gap between the strongest and weakest pair — the dispersion figure — came to 14.64 percentage points. That is a wide spread for a single calendar day across a basket of 20 pairs.
Range-position readings, which place each pair's close on a scale from 0% (30-day low) to 100% (30-day high), were broadly elevated. FET closed at 94% of its 30-day range; DOT at 88%; AVAX and ETH both at 89%; IMX at 87%; ICP at 86%. Even several of the session's decliners — BTC at 84%, SOL at 81%, NEAR at 80% — sat well above the midpoint of their own recent ranges. That means the session's losers were giving back ground from elevated positions rather than making new lows. XLM at 59%, XRP at 57%, and ADA at 56% were the only pairs whose closes sat below the midpoint of their 30-day ranges.
On the 30-day horizon, STX stood out with a +108.89% move over the period, a figure large enough that the session's +3.08% reads as a continuation of an already extended run rather than an isolated spike. LINK (+53.55% over 30 days), NEAR (+43.30%), and SOL (+36.64%) also carried substantial 30-day momentum into the session close, though all three declined on the day itself.
| Pair | Close (USD) | Session | 7d | 30d | Range pos. |
|---|---|---|---|---|---|
| ICP | 2.98 | +9.24% | +24.32% | +35.68% | 86% |
| DOT | 1.06 | +8.57% | +26.47% | +30.38% | 88% |
| FET | 0.182880 | +4.50% | +18.93% | +33.71% | 94% |
| XLM | 0.193629 | +3.21% | +9.23% | +17.84% | 59% |
| STX | 0.277400 | +3.08% | +10.17% | +108.89% | 85% |
| APT | 0.639900 | +2.61% | +19.61% | +6.05% | 64% |
| AVAX | 8.08 | +2.16% | +11.72% | +24.73% | 89% |
| ATOM | 1.63 | +2.06% | +10.81% | +17.77% | 74% |
| HBAR | 0.081860 | +0.36% | +11.24% | +18.79% | 78% |
| IMX | 0.132590 | +0.26% | +6.42% | +18.28% | 87% |
| OP | 0.110400 | +0.13% | +18.01% | +24.04% | 70% |
| DOGE | 0.090480 | -0.47% | +9.24% | +28.60% | 68% |
| ETH | 2,489.62 | -0.98% | +0.91% | +29.98% | 89% |
| ADA | 0.220450 | -1.22% | +11.48% | +10.62% | 56% |
| BTC | 79,091.97 | -1.55% | +0.67% | +21.85% | 84% |
| SHIB | 0.000005 | -1.63% | +7.51% | +17.75% | 58% |
| XRP | 1.40 | -1.91% | +1.21% | +34.42% | 57% |
| SOL | 103.79 | -2.59% | +0.73% | +36.64% | 81% |
| LINK | 12.75 | -3.69% | +12.62% | +53.55% | 83% |
| NEAR | 2.31 | -5.39% | +19.93% | +43.30% | 80% |
The Dispersed Regime: What It Describes
Skia Paper applies five coarse regime labels to each completed session: broad-advance, broad-decline, compressed, dispersed, and mixed. The dispersed label is assigned when the spread between the session's strongest and weakest pair is wide relative to the basket's typical daily range of outcomes. On 2026-09-07 the 14.64-point dispersion qualified; a compressed session, by contrast, would show a narrow spread with most pairs clustered near the median.
What distinguishes dispersed from mixed is primarily the magnitude of the spread rather than the direction of breadth. A mixed session tends to show a roughly even split between gainers and losers with moderate moves on both sides. Here, the split was also close to even (11 up, 9 down), but the amplitude of the outliers — particularly ICP and DOT on the upside — pulled the spread wide enough to override the mixed label. The median of +0.26% reflects that the centre of the distribution was nearly flat; the dispersion figure reflects that the tails were not.
Execution Cost in the Context of This Session
Coinbase Exchange publishes a tiered maker-taker fee schedule; at standard retail volume tiers, taker fees run to approximately 0.60% per side, producing a round-trip cost in the range of 1.20% before spread and slippage are added. For pairs that moved in the single digits on 2026-09-07, that round-trip floor is a meaningful fraction of the session's move. A pair that gained +2.06% (ATOM) or lost -1.22% (ADA) on the session leaves little margin between the gross price change and the cost of entering and exiting within the same candle.
The outliers at the tails of the distribution — ICP at +9.24%, DOT at +8.57% — represent moves large enough that a round-trip fee is a smaller proportional burden, though slippage on less liquid pairs can widen the effective cost beyond the published schedule. Pairs near the median (+0.26%) offer essentially no room between the gross move and a standard round-trip cost; the fee would exceed the price change entirely. Dispersed sessions create a wide range of cost-to-move ratios across the basket simultaneously, which is itself an observation about the structure of the day rather than a guide to action.
What One Session Does Not Establish
A single daily candle is a sample of one. The dispersed classification describes the shape of the 2026-09-07 close; it carries no predictive content about the session that follows. Regime labels are applied after the fact, and the label that fits today's data has no mechanical relationship to tomorrow's label.
The elevated range-position readings across much of the basket describe where prices closed relative to the prior 30 days — they do not indicate whether that position is a ceiling or a floor. A pair sitting at 94% of its 30-day range (FET) has simply closed near the top of that window; whether the window shifts up, down, or compresses over the next 30 days is not encoded in the figure. Similarly, STX's +108.89% 30-day change is a description of what already happened in that window; it says nothing about the direction or magnitude of the next window.
Dispersion itself is not a persistent state. Sessions classified as dispersed are followed by sessions of every other regime type; there is no structural reason a wide spread on one day propagates into the next. The failure mode of reading too much into a single dispersed session is precisely that it treats a descriptive label as if it were a signal.
The 2026-09-07 session closed with its gains and losses distributed unevenly across the basket, its median barely above flat, and most pairs still sitting in the upper half of their own 30-day ranges — a snapshot of one day in a market that will produce a different snapshot tomorrow.
Sources
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.