Research campaign · Iteration 1 · unassessed
DeMark analysis for SP-500 | RecessionAlert — exploratory proxy campaign
Can public ^GSPC/SPY/IEF proxies test disclosed TD-Setup Class-A/B/C rules without native RecessionAlert series parity?
- Expressions
- 24
- Logged trials
- Not recorded
- Independent events
- Not assessed
- Evidence
- unassessed
What the research found
The 2014 DeMark SP-500 article discloses modified TD-Setup Class-A/B/C buy and sell rules with RA buy>=8/sell>=9 thresholds and a 1-5 bar reaction hint, but no native setup/summation series or portfolio map.
Twenty-four labeled proxy variants on ^GSPC/SPY/IEF closes were executed; all are exploratory inferred reconstructions, not RecessionAlert-native series.
SPY/IEF overlay on RA buy>=8 lag2 yields the best tested Sharpe (0.446); long-history ^GSPC buy setups underperform buy-and-hold.
Diagnostics confirm lag>=1 causal baseline, six fidelity failure modes, and mixed crisis-window behavior; lag-zero controls are non-causal only.
Disposition is exploratory_tested_proxy with no playable indicator promotion.
Mechanism and falsifiers
Not recorded in this iteration.
Not recorded in this iteration.
Confidence and limitations
Exploratory inferred/proxy only; not native RecessionAlert DeMark chart series.
No playable indicator promotion without native parity.
Post-2020 Trendex successor semantics not tested here.
Not recorded in this iteration.
Compare expressions
Download evidenceExploratory results. Check each period, proxy and cost assumption before comparing. — means not recorded.
| Expression / family | CAGR | Sharpe | Max drawdown | Test period | Assessment |
|---|---|---|---|---|---|
| demark-canonical-buy9-lag1-gspccanonical-td-setup-9 | -0.4% | -0.09 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-ra-buy8-lag1-gspcra-threshold-buy8-sell9 | -0.5% | -0.09 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-class-b-summation-lag1-gspcra-class-b-summation-cross | 0.3% | 0.06 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-class-a-sell-lag2-spyra-class-a-setup-flip | 1.1% | 0.25 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-canonical-buy9-lag0-gspccanonical-td-setup-9 | -0.5% | -0.10 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-canonical-buy9-lag0-gspc-h3canonical-td-setup-9 | -0.5% | -0.10 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-canonical-buy9-lag1-gspc-h3canonical-td-setup-9 | -0.4% | -0.09 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-canonical-buy9-lag2-gspc-h3canonical-td-setup-9 | -0.5% | -0.11 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-canonical-buy9-lag1-gspc-h1canonical-td-setup-9 | -0.2% | -0.05 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-canonical-buy9-lag1-gspc-h5canonical-td-setup-9 | -0.3% | -0.04 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-ra-buy8-lag2-gspcra-threshold-buy8-sell9 | -0.8% | -0.14 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-class-b-summation-lag2-gspcra-class-b-summation-cross | 1.0% | 0.21 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-reaction-hold1-classb-lag1-spyra-class-b-summation-cross | 0.3% | 0.10 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-reaction-hold3-classb-lag1-spyra-class-b-summation-cross | -0.8% | -0.17 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-reaction-hold5-classb-lag1-spyra-class-b-summation-cross | -1.4% | -0.23 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-class-a-sell-lag0-spyra-class-a-setup-flip | 1.2% | 0.28 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-class-a-sell-lag1-spyra-class-a-setup-flip | 0.9% | 0.24 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-canonical-sell9-lag1-gspccanonical-td-setup-9 | 0.9% | 0.22 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-ra-sell9-lag1-gspcra-threshold-buy8-sell9 | 0.9% | 0.22 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-class-c-summation-lag1-gspcra-class-c-summation-flip | 1.8% | 0.18 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-canonical-buy9-lag1-spycanonical-td-setup-9 | -0.4% | -0.07 | — | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-overlay-classb-lag1-spy-iefspy-ief-risk-off-overlay | 2.8% | 0.38 | -25.8% | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-overlay-canonical-lag1-spy-iefspy-ief-risk-off-overlay | 2.8% | 0.37 | -23.6% | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
| demark-overlay-ra-buy8-lag2-spy-iefspy-ief-risk-off-overlay | 3.3% | 0.45 | -28.3% | Not recorded |
exploratory inferred public proxy; not native DeMark RA series |
Open an expression to inspect its rules and request confirmation. The request must be submitted by a trusted repository collaborator.
demark-canonical-buy9-lag1-gspc
- claim kind
documented_canonical
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
canonical-td-setup-9
- id
demark-canonical-buy9-lag1-gspc
- label
causal canonical 9-count buy setup on ^GSPC long history
- params
- holding bars
3
- lag bars
1
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-ra-buy8-lag1-gspc
- claim kind
inferred_documented_ra_threshold
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
ra-threshold-buy8-sell9
- id
demark-ra-buy8-lag1-gspc
- label
causal RA buy>=8 setup flip on ^GSPC
- params
- buy threshold
8
- holding bars
3
- lag bars
1
- sell threshold
9
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-class-b-summation-lag1-gspc
- claim kind
inferred_documented_summation
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
ra-class-b-summation-cross
- id
demark-class-b-summation-lag1-gspc
- label
causal Class-B summation cross-below 8 on ^GSPC
- params
- holding bars
5
- lag bars
1
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-class-a-sell-lag2-spy
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
SPY
- family
ra-class-a-setup-flip
- id
demark-class-a-sell-lag2-spy
- label
lag-2 Class-A sell flip on SPY ETF window
- params
- holding bars
3
- lag bars
2
- side
sell
- source fidelity
False
- span
- end
2026-07-24
- n bars
8428
- start
1993-01-29
demark-canonical-buy9-lag0-gspc
- claim kind
control_only
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
canonical-td-setup-9
- id
demark-canonical-buy9-lag0-gspc
- label
same-close lookahead control; never causal primary
- params
- holding bars
3
- lag bars
0
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-canonical-buy9-lag0-gspc-h3
- claim kind
control_only
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
canonical-td-setup-9
- id
demark-canonical-buy9-lag0-gspc-h3
- label
canonical 9-count buy lag 0 on ^GSPC
- params
- holding bars
3
- lag bars
0
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-canonical-buy9-lag1-gspc-h3
- claim kind
documented_canonical
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
canonical-td-setup-9
- id
demark-canonical-buy9-lag1-gspc-h3
- label
canonical 9-count buy lag 1 on ^GSPC
- params
- holding bars
3
- lag bars
1
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-canonical-buy9-lag2-gspc-h3
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
canonical-td-setup-9
- id
demark-canonical-buy9-lag2-gspc-h3
- label
canonical 9-count buy lag 2 on ^GSPC
- params
- holding bars
3
- lag bars
2
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-canonical-buy9-lag1-gspc-h1
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
canonical-td-setup-9
- id
demark-canonical-buy9-lag1-gspc-h1
- label
canonical buy lag1 holding 1 bars on ^GSPC
- params
- holding bars
1
- lag bars
1
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-canonical-buy9-lag1-gspc-h5
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
canonical-td-setup-9
- id
demark-canonical-buy9-lag1-gspc-h5
- label
canonical buy lag1 holding 5 bars on ^GSPC
- params
- holding bars
5
- lag bars
1
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-ra-buy8-lag2-gspc
- claim kind
inferred_documented_ra_threshold
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
ra-threshold-buy8-sell9
- id
demark-ra-buy8-lag2-gspc
- label
RA buy>=8 lag 2 on ^GSPC
- params
- buy threshold
8
- holding bars
3
- lag bars
2
- sell threshold
9
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-class-b-summation-lag2-gspc
- claim kind
inferred_documented_summation
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
ra-class-b-summation-cross
- id
demark-class-b-summation-lag2-gspc
- label
Class-B summation cross lag 2 on ^GSPC
- params
- holding bars
5
- lag bars
2
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-reaction-hold1-classb-lag1-spy
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
SPY
- family
ra-class-b-summation-cross
- id
demark-reaction-hold1-classb-lag1-spy
- label
reaction-window hold 1 bars Class-B on SPY
- params
- holding bars
1
- lag bars
1
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
8428
- start
1993-01-29
demark-reaction-hold3-classb-lag1-spy
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
SPY
- family
ra-class-b-summation-cross
- id
demark-reaction-hold3-classb-lag1-spy
- label
reaction-window hold 3 bars Class-B on SPY
- params
- holding bars
3
- lag bars
1
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
8428
- start
1993-01-29
demark-reaction-hold5-classb-lag1-spy
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
SPY
- family
ra-class-b-summation-cross
- id
demark-reaction-hold5-classb-lag1-spy
- label
reaction-window hold 5 bars Class-B on SPY
- params
- holding bars
5
- lag bars
1
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
8428
- start
1993-01-29
demark-class-a-sell-lag0-spy
- claim kind
control_only
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
SPY
- family
ra-class-a-setup-flip
- id
demark-class-a-sell-lag0-spy
- label
Class-A sell flip lag 0 on SPY
- params
- holding bars
3
- lag bars
0
- side
sell
- source fidelity
False
- span
- end
2026-07-24
- n bars
8428
- start
1993-01-29
demark-class-a-sell-lag1-spy
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
SPY
- family
ra-class-a-setup-flip
- id
demark-class-a-sell-lag1-spy
- label
Class-A sell flip lag 1 on SPY
- params
- holding bars
3
- lag bars
1
- side
sell
- source fidelity
False
- span
- end
2026-07-24
- n bars
8428
- start
1993-01-29
demark-canonical-sell9-lag1-gspc
- claim kind
documented_canonical
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
canonical-td-setup-9
- id
demark-canonical-sell9-lag1-gspc
- label
canonical 9-count sell setup lag1 on ^GSPC
- params
- holding bars
3
- lag bars
1
- side
sell
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-ra-sell9-lag1-gspc
- claim kind
inferred_documented_ra_threshold
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
ra-threshold-buy8-sell9
- id
demark-ra-sell9-lag1-gspc
- label
RA sell>=9 threshold lag1 on ^GSPC
- params
- buy threshold
8
- holding bars
3
- lag bars
1
- sell threshold
9
- side
sell
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-class-c-summation-lag1-gspc
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
ra-class-c-summation-flip
- id
demark-class-c-summation-lag1-gspc
- label
Class-C summation flip lag1 on ^GSPC
- params
- holding bars
3
- lag bars
1
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
demark-canonical-buy9-lag1-spy
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
SPY
- family
canonical-td-setup-9
- id
demark-canonical-buy9-lag1-spy
- label
canonical buy lag1 on SPY ETF window
- params
- holding bars
3
- lag bars
1
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
8428
- start
1993-01-29
demark-overlay-classb-lag1-spy-ief
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
SPY
- family
spy-ief-risk-off-overlay
- id
demark-overlay-classb-lag1-spy-ief
- label
SPY/IEF overlay on Class-B summation buy lag1
- params
- holding bars
3
- lag bars
1
- risk off
IEF
- signal family
ra-class-b-summation-cross
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
6035
- start
2002-07-30
demark-overlay-canonical-lag1-spy-ief
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
SPY
- family
spy-ief-risk-off-overlay
- id
demark-overlay-canonical-lag1-spy-ief
- label
SPY/IEF overlay on canonical buy lag1
- params
- holding bars
3
- lag bars
1
- risk off
IEF
- signal family
canonical-td-setup-9
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
6035
- start
2002-07-30
demark-overlay-ra-buy8-lag2-spy-ief
- claim kind
inferred_nearby_variant
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
SPY
- family
spy-ief-risk-off-overlay
- id
demark-overlay-ra-buy8-lag2-spy-ief
- label
SPY/IEF overlay on RA buy>=8 lag2
- params
- holding bars
3
- lag bars
2
- risk off
IEF
- signal family
ra-threshold-buy8-sell9
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
6035
- start
2002-07-30
Interactive lab
Interactive history unavailableLoading available evidence…
Research record
Source claims, inferred rules, experiments and the evidence behind the assessment.
Source
- sha256
sha256:2182cdc8962b8bcadc9008acd3a880fbe5486dfc804f728e661902885680cc3b
- fetched at
2026-08-22T23:44:53Z
What the source claims
Modified TD-Setup Class-A/B/C buy and sell rules with RA buy>=8/sell>=9 thresholds.
No machine-readable setup-count or summation history in checked-in snapshot.
Twenty-four labeled proxy variants across lag/holding/driver grids.
Daily close with lag>=1 causal default; lag0 control_only.
Rules actually disclosed
Modified TD-Setup Class-A/B/C buy and sell rules with RA buy>=8/sell>=9 thresholds.
No machine-readable setup-count or summation history in checked-in snapshot.
What had to be inferred
Twenty-four labeled proxy variants across lag/holding/driver grids.
Daily close with lag>=1 causal default; lag0 control_only.
Research questions
Can public ^GSPC/SPY/IEF proxies test disclosed TD-Setup Class-A/B/C rules without native RecessionAlert series parity?
Data
- source sha256
sha256:2182cdc8962b8bcadc9008acd3a880fbe5486dfc804f728e661902885680cc3b
- source fetched at
2026-08-22T23:44:53Z
- database
data/market.sqlite
- database access
read-only query_only
- native span
No native series; ^GSPC proxy to 1927
- publication lag
Daily close assumed; exact RA release clock unknown
- proxy chain
^GSPC
SPY
IEF
- private data
False
Baseline implementation
- claim kind
documented_canonical
- classification
exploratory inferred public proxy; not native DeMark RA series
- driver
^GSPC
- family
canonical-td-setup-9
- id
demark-canonical-buy9-lag1-gspc
- label
causal canonical 9-count buy setup on ^GSPC long history
- metrics
- benchmark cagr
0.06319803645677458
- benchmark total return
418.70441928144635
- cagr
-0.0043255452476459455
- cagr delta vs buy hold
-0.06752358170442052
- event rate
0.009613442662681262
- events
238
- n bars
24757
- sharpe
-0.08745920481651166
- vol
0.04945786160211365
- params
- holding bars
3
- lag bars
1
- side
buy
- source fidelity
False
- span
- end
2026-07-24
- n bars
24757
- start
1927-12-30
What to try interactively
- name
lag_bars
- values
0
1
2
- name
holding_bars
- values
1
3
5
Suggested next research
Obtain native RA setup/summation series or workbook bytes under separate authorization.
Resolve sell nine-count vs eleven-count cancellation conflict from the page text.
Document exact publication lag and Trendex successor semantics for post-2020 charts.
Test proprietary RA modifications beyond public threshold hints if disclosed.
Validate crisis-window behavior on authorized DeMark vendor or constituent-level data.
Trial ledger
- number
1
- status
published
- summary
Initial exploratory proxy campaign with 24 variants.
Next questions
Obtain native RA setup/summation series or workbook bytes under separate authorization.
Resolve sell nine-count vs eleven-count cancellation conflict from the page text.
Document exact publication lag and Trendex successor semantics for post-2020 charts.
Test proprietary RA modifications beyond public threshold hints if disclosed.
Validate crisis-window behavior on authorized DeMark vendor or constituent-level data.
Review status
published. Research publication does not imply official admission.
Return to pending research