Themes / Group
Inactive · established relationshipLatest completed snapshot · Aug 20 · 16:27 ET

MARKET-DISCOVERED GROUP

High-Performance Computing

THE TRADER READBroad relative move, confirmed beyond the cohort—but not by flow

8 of 12 measured SPY-adjusted paths align to the downside across 9.2 effective names. Common-bar coverage spans 12 of 16 cohort members. ABTC is also moving with filing-linked members. Capital-flow agreement is 18%, so the basket is not yet clean. No verified headline preceded the measured move.

Price breadth1 up · 15 down-1.9% median path versus SPY
Capital-flow alignment18%Price breadth is not yet confirmed by flow
Activity nowCalmActivity score 12/100 · lift 0.8×
Evidence depth5 episodes21 observed active sessions
EVIDENCE CLOCK

What happened, in timestamp order

4 timed events · 09:55–15:55 ET
TIMING READNo verified timestamped headline was found within 24 hours of the first measured price move in the available common-bar window. The move currently reads as market-led or unexplained.
  1. 5-minute inferenceCRWV registered the first material downside path

    -0.79% session path at the threshold · CRWV

  2. 5-minute inferenceGLXY registered a material upside path

    +0.43% session path at the threshold · GLXY

  3. 5-minute inferenceLUMN registered a material downside path

    -0.77% session path at the threshold · LUMN

  4. 5-minute inferenceHUT registered a material upside path

    +0.81% session path at the threshold · HUT

2 earlier filing context receipts

TUE, FEB 10APH filing documented shared operating contextcredit / rates affecting interest expense · date precision only

WED, AUG 5MARA filing documented shared operating contextcredit / rates affecting interest expense · date precision only

Headline times and retained activity confirmations are point-in-time evidence. Price order is inferred from aligned 5-minute bars. Filing evidence has date precision only and is context, not an intraday catalyst claim.
THEME CONSTRUCTION

Broad move or one-stock disguise?

SPY-adjusted 5m paths

Observed movement is broad across 9.2 effective names. IREN represents 14.9% of absolute path energy; 8 of 12 members align with the down relative path.

ConstructionBroad14.9% from the largest path
Effective names9.2 / 1212 of 16 cohort members had common bars
Direction breadth8/12 aligneddownside relative path · 2 counter-moving
Largest moverIREN 14.9%28.3% from the top two paths
IREN14.9%Aligned contributor · path -3.60% · basket -0.30pp
FRMI13.4%Aligned contributor · path -3.24% · basket -0.27pp
CIFR12.9%Aligned contributor · path -3.10% · basket -0.26pp
ALAB11.2%Counter-move · path +2.69% · basket +0.23pp
APLD10.4%Aligned contributor · path -2.50% · basket -0.21pp
CLSK9.9%Aligned contributor · path -2.38% · basket -0.20pp
CRWV8.7%Aligned contributor · path -2.09% · basket -0.17pp
CORZ7.3%Aligned contributor · path -1.75% · basket -0.15pp
GLXY5.6%Counter-move · path +1.36% · basket +0.11pp
LUMN5.4%Aligned contributor · path -1.30% · basket -0.11pp
HUT0.3%Flat / inactive path · path +0.07% · basket +0.01pp
APH0.2%Flat / inactive path · path -0.04% · basket -0.00pp
Top six names by movement share are shown. Movement share measures absolute 5-minute path energy; signed basket contribution is equal-weight. Neither is a portfolio weight or causal attribution.

01 · WHAT IS HAPPENING

The move, the flow and who is driving it

Price breadth, unusual activity and signed capital flow are separate observations
HOW THE PRICE MOVE SPREAD

Observed session sequence

5-minute bars · inferred
  1. CRWV moved lower firstFirst member to clear the material path threshold · path -0.79%
  2. GLXY registered a material higher path0m after the first price move · path +0.43%
  3. LUMN registered a material lower path5m after the first price move · path -0.77%
  4. HUT registered a material higher path5m after the first price move · path +0.81%
  5. APH registered a material lower path5m after the first price move · path -0.39%
  6. APLD registered a material lower path15m after the first price move · path -0.65%

Sequence inferred from aligned 5-minute price paths; it is not a causal ordering.

RELATIVE CO-MOVEMENT

How the members are behaving together

SPY-adjusted
Oct 7 · 09:55 ETMove visible 10:20 ETOct 7 · 15:55 ET
Group median -1.92%Strongest relative path ALABWeakest relative path IREN

Observed member paths adjusted against SPY. The dashed marker shows when the median move first cleared a material threshold; shading marks the latest 20% of the observed window.

02 · WHY IT MATTERS

Why now and where the move could travel

Context and lead/lag links are evidence, not proof of causation or a forecast
WHY NOW

Verified recent context

0 items

No verified recent catalyst is attached to this move. The relationship may be market-led rather than news-led.

WHY THIS GROUP EXISTS

Shared filing context

15/16 members covered

Moderate filing support: the strongest shared 10-Q/K outlook is operating setup (near-term outlook) across APH, APLD, CORZ, CRWV, FRMI, GLXY, HUT, IREN, LUMN, MARA, WULF.

credit / rates affecting interest expense

“Interest income, Interest expense and Other Three Months Ended June 30, Change (in thousands) 2026 2025 $ Interest income Interest income from loaned bitcoin $ 4,291 $ 6,795 $ (2,504) Interest income from cash and cash equivalents 5,972 2,836 3,136 Total interest income 10,263 9,631 632 Interest expense (6,264) (12,835) 6,571 Other 2,646 (5,509) 8,155 Interest income increased by $0.6 million compared to the prior year period, primarily due to a higher average balance of cash and cash equivalents, partially offset by interest income earned on loaned bitcoin under our digital asset management strategy.”

MARA 10-Q · 2026-08-06
demand / volume

“48.9 million , or 93%, for the three months ended March 31, 2026, compared to the same period in 2025, primarily due to an increase in overall unit shipments driven by higher demand for our Scorpio, Aries, and Taurus products, as well as higher overall average selling prices resulting from an increased mix of hardware modules and Scorpio products .”

ALAB 10-Q · 2026-05-06
restructuring

“The $1.1 million, or approximately 11% decrease, was primarily due to reduced spending following the exit of certain technology initiatives in the prior year period as part of our previously disclosed restructuring plan, partially offset by continued initiatives and development activities to support our strategic expansion.”

MARA 10-Q · 2026-05-11
CONFIRMATION NETWORK

Outsiders and cross-asset links

1 outsiders confirming

No qualified cross-asset lead/lag link is available for this group.

Filing-linked outsiders test whether the move is spreading beyond the established cohort.

ABTCMoving with group

RIOT · text peer

KEELNot confirming

HUT · text peer

DGXXNot confirming

CLSK · text peer

NUAINot confirming

FRMI · text peer

03 · WHO IS IN IT

Member participation

Unusual activity and daily price direction are shown separately
MARACLSKRIOTALABHUTGLXYAPLDWULFLUMNFRMIIRENCIFRCORZNBISAPHCRWV

6/16 active1 rising · 15 falling

MemberUnusual activityToday's moveRead now
MARAMARA Holdings, Inc.
Most unusual · 100/100
-5.7%
Active; unclassifiedActive in 9.5% of observed group sessions
CLSKCleanSpark, Inc.
Elevated · 98/100
-6.2%
Aligned path contributorActive in 14.3% of observed group sessions
RIOTRiot Platforms, Inc.
Elevated · 95/100
-2.1%
Active; unclassifiedActive in 52.4% of observed group sessions
ALABAstera Labs, Inc.
Elevated · 78/100
-2.0%
Counter-move / dispersionActive in 14.3% of observed group sessions
HUTHut 8 Corp.
Elevated · 75/100
-3.1%
Active; unclassifiedActive in 19% of observed group sessions
GLXYGalaxy Digital
Elevated · 73/100
-3.4%
Counter-move / dispersionActive in 0% of observed group sessions
APLDApplied Digital Corporation
Elevated · 70/100
-6.1%
Aligned path contributorActive in 57.1% of observed group sessions
WULFTeraWulf Inc.
Elevated · 69/100
-3.9%
Flow divergenceActive in 47.6% of observed group sessions
LUMNLumen Technologies, Inc.
Quiet · 64/100
+0.2%
Watch / laggardActive in 0% of observed group sessions
FRMIFermi Inc.
Quiet · 59/100
-5.9%
Watch / laggardActive in 0% of observed group sessions
IRENIREN Limited
Quiet · 58/100
-6.2%
Watch / laggardActive in 61.9% of observed group sessions
CIFRCipher Digital Inc.
Quiet · 54/100
-6.1%
Watch / laggardActive in 42.9% of observed group sessions
CORZCore Scientific, Inc.
Quiet · 49/100
-4.5%
Watch / laggardActive in 33.3% of observed group sessions
NBISNebius Group N.V.
Quiet · 48/100
-5.1%
Watch / laggardActive in 19% of observed group sessions
APHAmphenol Corporation
Quiet · 45/100
-1.2%
Watch / laggardActive in 0% of observed group sessions
CRWVCoreWeave, Inc.
Quiet · 28/100
-3.5%
Watch / laggardActive in 33.3% of observed group sessions
View exact member values
TICKERCompanyRole nowTodayObserved historyTrader interpretation
MARAMARA Holdings, Inc.Most unusual-5.7%9.5% of 21 sessionsMost unusual activity in this snapshot; performance direction may differ.
CLSKCleanSpark, Inc.Participating-6.2%14.3% of 21 sessionsPrice participation is not yet confirmed by per-name flow.
RIOTRiot Platforms, Inc.Participating-2.1%52.4% of 21 sessionsPrice participation is not yet confirmed by per-name flow.
ALABAstera Labs, Inc.Participating-2.0%14.3% of 21 sessionsPrice participation with distributing signed flow.
HUTHut 8 Corp.Participating-3.1%19% of 21 sessionsPrice participation is not yet confirmed by per-name flow.
GLXYGalaxy DigitalParticipating-3.4%0% of 21 sessionsPrice participation is not yet confirmed by per-name flow.
APLDApplied Digital CorporationParticipating-6.1%57.1% of 21 sessionsPrice participation is not yet confirmed by per-name flow.
WULFTeraWulf Inc.Participating-3.9%47.6% of 21 sessionsPrice participation with distributing signed flow.
LUMNLumen Technologies, Inc.Quiet / lagging+0.2%0% of 21 sessionsNot participating yet; a breadth test or a genuine divergence.
FRMIFermi Inc.Quiet / lagging-5.9%0% of 21 sessionsNot participating yet; a breadth test or a genuine divergence.
IRENIREN LimitedQuiet / lagging-6.2%61.9% of 21 sessionsNot participating yet; a breadth test or a genuine divergence.
CIFRCipher Digital Inc.Quiet / lagging-6.1%42.9% of 21 sessionsNot participating yet; a breadth test or a genuine divergence.
CORZCore Scientific, Inc.Quiet / lagging-4.5%33.3% of 21 sessionsNot participating yet; a breadth test or a genuine divergence.
NBISNebius Group N.V.Quiet / lagging-5.1%19% of 21 sessionsNot participating yet; a breadth test or a genuine divergence.
APHAmphenol CorporationQuiet / lagging-1.2%0% of 21 sessionsNot participating yet; a breadth test or a genuine divergence.
CRWVCoreWeave, Inc.Quiet / lagging-3.5%33.3% of 21 sessionsNot participating yet; a breadth test or a genuine divergence.

04 · HAS THIS HAPPENED BEFORE

Identity history versus measured outcomes

Structural recurrence is not the same as a validated trading outcome
POINT-IN-TIME LIVE OUTCOMES

Early observed follow-through

2/2 events completed
30 minutes-0.21%50% continuation · n=2Median MFE +2.23% · MAE -2.84%
60 minutes+0.21%50% continuation · n=2Median MFE +2.23% · MAE -2.84%
Session close+0.01%100% continuation · n=1Median MFE +3.70% · MAE -1.63%
Next-session close-0.68%50% continuation · n=2Median MFE +2.23% · MAE -2.84%
Only point-in-time live watcher observations and conservative same-day EOD observations are included. Later backfills and replay runs are excluded.
moderate structural support

5 completed episodes across 21 active sessions provide moderate historical support. These episode counts describe recurrence; the outcome ledger above is the only forward-performance evidence.

Structural support62 daysHow often the cohort identity was observed
Completed episodes5Separate periods of broad group activity
Active sessions21Completed session observations, not trades
2026-08-191 active session100% composition consistency through the episode
2026-07-22 → 2026-08-1011 active sessionsNo name was active through every recorded session
2026-07-14 → 2026-07-162 active sessions11.1% composition consistency through the episode
2026-07-07 → 2026-07-082 active sessions25% composition consistency through the episode
2026-06-24 → 2026-07-025 active sessionsNo name was active through every recorded session
MEMBER CONSISTENCY IN OBSERVED ACTIVE SESSIONS
IREN61.9% · 4 episodes
APLD57.1% · 4 episodes
RIOT52.4% · 5 episodes
WULF47.6% · 5 episodes
CIFR42.9% · 3 episodes
CORZ33.3% · 2 episodes
CRWV33.3% · 3 episodes
HUT19% · 3 episodes
NBIS19% · 1 episode
ALAB14.3% · 2 episodes
CLSK14.3% · 2 episodes
MARA9.5% · 2 episodes
APH0% · 0 episodes
FRMI0% · 0 episodes
GLXY0% · 0 episodes
LUMN0% · 0 episodes