The same complexity idea on the bare up/down sequence, ignoring move size (sign-binarized). ยท row #20 ยท custom ยท rolling-200 on close ยท Pinto 2025 (arXiv:2504.01974); sign-binarized length-m word complexity
WATCH
Verdict (worst valid cell)
0.94
Max โฯโ vs panel
0.24
Max Rยฒ
1.00
Min h_norm
210
Valid cells
201 / 9 / 0
PASS / WATCH / BAN
10 / 10
Slices done
fast
Cost tier
How this verdict was reached
9 cell(s) landed in the 0.85โ0.95 WATCH band and none were banned. The worst cell is โฯโ 0.937 vs bar_petrosian_fd. The whole-candidate verdict is the single worst cell (a feature must be orthogonal in every regime, not on average), so the candidate is WATCH.
What this feature measures
In plain terms: The same complexity idea, but on the bare up/down sequence of the price โ throwing away HOW BIG each move is and keeping only its direction. How much organised structure lives in the pattern of ups and downs.
Analogy: Strip the chart down to a string of โฒ/โผ and ask how patterned that string is, versus a random sequence of ups and downs.
High vs low: HIGH = the up/down pattern is structured. LOW = the up/down pattern is flat or indistinguishable from random.
Tunable knobs (Axis-2): Word length = m=3 (PR#513) and sign (up/down) binarization โ both parameter-free. (Binarization is assumed sign-based per Pinto 2025; flagged for a faithfulness check before promotion.)
How we test it
We slide a 200-bar window across real ClickHouse bars and compute this feature at every bar, for each of 21 (symbol ร threshold) cells across 6 crypto symbols, repeated over 10 historical 2-month windows ("slices") spanning 2021โ2026 โ 210 independent tests.
In each test we ask one question only: does this feature carry information the live production panel does not already have? (This is Axis-1 orthogonality โ no profit/signal claim is made here.)
Three checks per cell: rank-correlation vs the single closest feature (โฯโ), joint reconstructability from the whole panel (Rยฒ), and whether the feature actually varies (h_norm).
Read-only ClickHouse ยท against the live panel + the landed Petrosian/Katz fractal features.
What each number means
Max โฯโ (rank correlation): how strongly the feature tracks the single most-similar existing feature, in its worst cell. 0 = unrelated, 1 = a rename. Lower is better. โค0.85 good ยท 0.85โ0.95 watch ยท >0.95 banned (duplicate).
Max Rยฒ (joint redundancy): how much of the feature can be rebuilt from all existing features together. 0 = fully new, 1 = fully reconstructable. Lower is better. Same thresholds.
Min h_norm (liveness): how much the feature actually moves (normalised entropy of its own values). Near 0 = basically constant / collapsed = useless. Want โฅ0.05.
Worst-valid-cell rollup: the candidate inherits the verdict of its worst of 210 cells โ deliberately strict, so a feature only PASSES if it is orthogonal everywhere.
Per-slice (regime) orthogonality
Each row is one 2-month historical window. "Worst โฯโ" is the most-correlated existing feature in that regime's worst cell โ lower is better.
Slice
Valid
P / W / B
Worst โฯโ
Top competitor
slice012026-01 โ 02
21
21 / 0 / 0
0.81
bar_katz_fd
slice032025-07 โ 08
21
21 / 0 / 0
0.73
bar_petrosian_fd
slice052025-01 โ 02
21
21 / 0 / 0
0.69
bar_katz_fd
slice072024-07 โ 08
21
21 / 0 / 0
0.79
bar_katz_fd
slice092024-01 โ 02
21
21 / 0 / 0
0.83
bar_katz_fd
slice122023-04 โ 05
21
21 / 0 / 0
0.79
bar_katz_fd
slice142022-10 โ 11
21
20 / 1 / 0
0.88
bar_petrosian_fd
slice172022-01 โ 02
21
21 / 0 / 0
0.69
bar_katz_fd
slice202021-04 โ 05
21
16 / 5 / 0
0.92
bar_petrosian_fd
slice212021-01 โ 02
21
18 / 3 / 0
0.94
bar_petrosian_fd
Most-similar existing features (the candidate's nearest neighbours)
Which already-shipped features this candidate most often resembles. If one feature dominates here and โฯโ is high, the candidate may be re-deriving it.
bar_katz_fd โ most-similar existing feature in 105 cell(s)
bar_petrosian_fd โ most-similar existing feature in 68 cell(s)
trade_intensity โ most-similar existing feature in 22 cell(s)
Resource utilization
Hard cap: 2 CPU cores (CPUQuota 200%), 2 GB RAM, spare-only โ the loop runs at the lowest priority (CPUWeight 10, nice 19, ionice idle) and only starts when โฅ2 cores and โฅ2 GB are free, so it never competes with production (ClickHouse / the live sidecar).
This sweep: wall time 8m 12s ยท peak RAM 1915 MB of 2048 MB.
Data volume: 6,333,627 bar-rows processed across 210 cells. 28 of 210 cells were sub-sampled to the most-recent 100,000 bars (largest raw cell โ 1,058,962 bars). Sub-sampling keeps giant cells tractable and does not change the correlation estimate.
Safety: the sweep completing means it stayed within 2 GB โ had it exceeded, the cgroup would have killed only the loop, never the box, and the cap-worked event would be recorded (never raised).
Worth flagging to the operator
๐ด โฯโ 0.937 crosses 0.85 in โฅ1 cell (closest existing feature: bar_petrosian_fd) โ that cell is WATCH.
โน๏ธ 28 cell(s) sub-sampled to the 100,000-bar cap (estimate unaffected).
โน๏ธ #20 binarization is assumed sign-of-increment (parameter-free); confirm faithful to Pinto 2025 (arXiv:2504.01974) before promotion.