Research campaign · Iteration 1 · unassessed
PRO>Trendex Charts | RecessionAlert
Can labeled close-only public proxies and the related WeeklyData TRENDEX workbook extract any useful SPY/IEF market-timing overlay from the incomplete PRO>Trendex Charts page?
- Expressions
- 29
- Logged trials
- Not recorded
- Independent events
- Not assessed
- Evidence
- unassessed
What the research found
Twenty-nine labeled variants executed with lag>=1 causal_primary (26) and lag-0/same-close control_only (3).
Best causal Sharpe: be_spy_lb12_lag1_t0 sharpe=1.031 (best-effort SMA grid; not a Trendex reconstruction).
Best causal CAGR: be_spy_lb21_lag1_tneg05 cagr=0.116.
Least drawdown: wb_trendex_overlay_lag1 max_drawdown=-0.341 (off-row weekly workbook; not native).
Best source fidelity dem21_causal_lag1 sharpe=0.769 with 100% risk-on share: channel/diffusion/exceedance analogues converge near buy-and-hold on monthly bars.
Regime split on dem21: GFC -1.03, COVID +0.78, 2022 -0.76, recent +1.61; nine families agree risk-on ~92% of months.
Native DEM++/diffusion/probability fidelity remains NO-GO; publication is Other/Research exploratory_tested_proxy.
Mechanism and falsifiers
Not recorded in this iteration.
Not recorded in this iteration.
Confidence and limitations
Not native Trendex DEM++/diffusion/probability parity
On-row assets=0; SPXTDX.gif SHA256 unavailable
market.sqlite ohlcv is close-only; native uses undisclosed support/resistance
WeeklyData TRENDEX PROB MODELS sheet is off-row weekly vs daily page cadence
VTI absent from market cache; SPY/IWM cover broad/small legs
Publication lag unknown; causal default lag>=1
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 |
|---|---|---|---|---|---|
| dem10_causal_lag1dem-short-channel10-close-proxy | — | — | — | Not recorded |
exploratory |
| dem10_causal_lag2dem-short-channel10-close-proxy | — | — | — | Not recorded |
exploratory |
| dem10_sameclose_controldem-short-channel10-close-proxy | — | — | — | Not recorded |
exploratory |
| dem21_causal_lag1dem-medium-channel21-close-proxy | — | — | — | Not recorded |
exploratory |
| dem21_causal_lag2dem-medium-channel21-close-proxy | — | — | — | Not recorded |
exploratory |
| dem34_causal_lag1dem-medium-channel21-close-proxy | — | — | — | Not recorded |
exploratory |
| dem50_causal_lag1dem-long-channel50-close-proxy | — | — | — | Not recorded |
exploratory |
| dem50_causal_lag2dem-long-channel50-close-proxy | — | — | — | Not recorded |
exploratory |
| dem72_causal_lag1dem-long-channel50-close-proxy | — | — | — | Not recorded |
exploratory |
| diff_warn32_onlydiffusion-3way-aboveness-staging | — | — | — | Not recorded |
exploratory |
| diff_act21_lag1diffusion-3way-aboveness-staging | — | — | — | Not recorded |
exploratory |
| diff_act10_cashshort_lag1diffusion-3way-aboveness-staging | — | — | — | Not recorded |
exploratory |
| diff_act21_lag2diffusion-3way-aboveness-staging | — | — | — | Not recorded |
exploratory |
| ptop85_lag1duration-exceedance-top-probability | — | — | — | Not recorded |
exploratory |
| ptop90_lag1duration-exceedance-top-probability | — | — | — | Not recorded |
exploratory |
| ptop85_lag2duration-exceedance-top-probability | — | — | — | Not recorded |
exploratory |
| ptop85_lag0_controlduration-exceedance-top-probability | — | — | — | Not recorded |
exploratory |
| pbot85_lag1duration-exceedance-trough-probability | — | — | — | Not recorded |
exploratory |
| pbot95_lag1duration-exceedance-trough-probability | — | — | — | Not recorded |
exploratory |
| pbot85_lag2duration-exceedance-trough-probability | — | — | — | Not recorded |
exploratory |
| xconv_any85_lag1conviction85-gated-cross-overlay | — | — | — | Not recorded |
exploratory |
| xconv_maj85_lag1conviction85-gated-cross-overlay | — | — | — | Not recorded |
exploratory |
| xconv_any90_lag2conviction85-gated-cross-overlay | — | — | — | Not recorded |
exploratory |
| prob_only_controlduration-exceedance-top-probability | — | — | — | Not recorded |
exploratory |
| be_spy_lb12_lag1_t0best-effort-neighbor-grid | — | — | — | Not recorded |
exploratory |
| be_spy_lb21_lag1_tneg05best-effort-neighbor-grid | — | — | — | Not recorded |
exploratory |
| be_spy_lb6_lag2_t0best-effort-neighbor-grid | — | — | — | Not recorded |
exploratory |
| wb_trendex_overlay_lag1related-workbook-overlay | — | — | — | Not recorded |
exploratory |
| wb_trendex_overlay_lag2related-workbook-overlay | — | — | — | Not recorded |
exploratory |
Open an expression to inspect its rules and request confirmation. The request must be submitted by a trusted repository collaborator.
dem10_causal_lag1
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
dem-short-channel10-close-proxy
- id
dem10_causal_lag1
- parameters
- lag bars
1
- lookback
10
dem10_causal_lag2
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
dem-short-channel10-close-proxy
- id
dem10_causal_lag2
- parameters
- lag bars
2
- lookback
10
dem10_sameclose_control
- claim kind
documented_but_incomplete
- claim role
control_only
- exploratory
True
- family
dem-short-channel10-close-proxy
- id
dem10_sameclose_control
- parameters
- lag bars
0
- lookback
10
dem21_causal_lag1
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
dem-medium-channel21-close-proxy
- id
dem21_causal_lag1
- parameters
- lag bars
1
- lookback
21
dem21_causal_lag2
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
dem-medium-channel21-close-proxy
- id
dem21_causal_lag2
- parameters
- lag bars
2
- lookback
21
dem34_causal_lag1
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
dem-medium-channel21-close-proxy
- id
dem34_causal_lag1
- parameters
- lag bars
1
- lookback
34
dem50_causal_lag1
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
dem-long-channel50-close-proxy
- id
dem50_causal_lag1
- parameters
- lag bars
1
- lookback
50
dem50_causal_lag2
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
dem-long-channel50-close-proxy
- id
dem50_causal_lag2
- parameters
- lag bars
2
- lookback
50
dem72_causal_lag1
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
dem-long-channel50-close-proxy
- id
dem72_causal_lag1
- parameters
- lag bars
1
- lookback
72
diff_warn32_only
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
diffusion-3way-aboveness-staging
- id
diff_warn32_only
- parameters
- lag bars
1
- stage
warn32
diff_act21_lag1
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
diffusion-3way-aboveness-staging
- id
diff_act21_lag1
- parameters
- lag bars
1
- stage
act21
diff_act10_cashshort_lag1
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
diffusion-3way-aboveness-staging
- id
diff_act10_cashshort_lag1
- parameters
- lag bars
1
- stage
act10
diff_act21_lag2
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
diffusion-3way-aboveness-staging
- id
diff_act21_lag2
- parameters
- lag bars
2
- stage
act21
ptop85_lag1
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
duration-exceedance-top-probability
- id
ptop85_lag1
- parameters
- lag bars
1
- threshold
85.0
ptop90_lag1
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
duration-exceedance-top-probability
- id
ptop90_lag1
- parameters
- lag bars
1
- threshold
90.0
ptop85_lag2
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
duration-exceedance-top-probability
- id
ptop85_lag2
- parameters
- lag bars
2
- threshold
85.0
ptop85_lag0_control
- claim kind
inferred
- claim role
control_only
- exploratory
True
- family
duration-exceedance-top-probability
- id
ptop85_lag0_control
- parameters
- lag bars
0
- threshold
85.0
pbot85_lag1
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
duration-exceedance-trough-probability
- id
pbot85_lag1
- parameters
- lag bars
1
- threshold
85.0
pbot95_lag1
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
duration-exceedance-trough-probability
- id
pbot95_lag1
- parameters
- lag bars
1
- threshold
95.0
pbot85_lag2
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
duration-exceedance-trough-probability
- id
pbot85_lag2
- parameters
- lag bars
2
- threshold
85.0
xconv_any85_lag1
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
conviction85-gated-cross-overlay
- id
xconv_any85_lag1
- parameters
- diffusion gate
any
- lag bars
1
- threshold
85.0
xconv_maj85_lag1
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
conviction85-gated-cross-overlay
- id
xconv_maj85_lag1
- parameters
- diffusion gate
majority
- lag bars
1
- threshold
85.0
xconv_any90_lag2
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
conviction85-gated-cross-overlay
- id
xconv_any90_lag2
- parameters
- diffusion gate
any
- lag bars
2
- threshold
90.0
prob_only_control
- claim kind
inferred
- claim role
control_only
- exploratory
True
- family
duration-exceedance-top-probability
- id
prob_only_control
- parameters
- lag bars
0
- threshold
85.0
be_spy_lb12_lag1_t0
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
best-effort-neighbor-grid
- id
be_spy_lb12_lag1_t0
- parameters
- lag bars
1
- lookback
12
- threshold
0.0
be_spy_lb21_lag1_tneg05
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
best-effort-neighbor-grid
- id
be_spy_lb21_lag1_tneg05
- parameters
- lag bars
1
- lookback
21
- threshold
-0.5
be_spy_lb6_lag2_t0
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
best-effort-neighbor-grid
- id
be_spy_lb6_lag2_t0
- parameters
- lag bars
2
- lookback
6
- threshold
0.0
wb_trendex_overlay_lag1
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
related-workbook-overlay
- id
wb_trendex_overlay_lag1
- parameters
- lag bars
1
- stage
act21
wb_trendex_overlay_lag2
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
related-workbook-overlay
- id
wb_trendex_overlay_lag2
- parameters
- lag bars
2
- stage
act21
Interactive lab
Interactive history unavailableLoading available evidence…
Research record
Source claims, inferred rules, experiments and the evidence behind the assessment.
Source
- fetched at
2026-08-22T23:39:05Z
- kind
recessionalert_page
- sha256
sha256:1e8169579e056a0f50ab0fc97df47497897c37c9f169128eef54ab910abd4587
- title
PRO>Trendex Charts | RecessionAlert
What the source claims
Three-horizon DEM++ support/resistance stops (short light-blue, medium black, long red); SP500 close below a stop flips that horizon bearish, close above flips it bullish.
Diffusion 0-3 count of stops price is above with 3->2 warning, 2->1 serious sell, 1->0 cash-or-short narrative staging.
Per-stop 20y duration-exceedance top/trough probabilities; 85%+ readings are conviction only — the stop cross is the BUY/SELL signal.
Stop formula, support/resistance construction, 20y window definition, allocation map, lag clock, and on-row chart/xlsx SHA256 are absent on this page SHA.
Close-only ^GSPC/SPY channel-stop/diffusion/exceedance reconstructions and lagged weekly TRENDEX-workbook overlays stand in for the undisclosed engine.
Rules actually disclosed
Three-horizon DEM++ support/resistance stops (short light-blue, medium black, long red); SP500 close below a stop flips that horizon bearish, close above flips it bullish.
Diffusion 0-3 count of stops price is above with 3->2 warning, 2->1 serious sell, 1->0 cash-or-short narrative staging.
Per-stop 20y duration-exceedance top/trough probabilities; 85%+ readings are conviction only — the stop cross is the BUY/SELL signal.
Stop formula, support/resistance construction, 20y window definition, allocation map, lag clock, and on-row chart/xlsx SHA256 are absent on this page SHA.
What had to be inferred
Close-only ^GSPC/SPY channel-stop/diffusion/exceedance reconstructions and lagged weekly TRENDEX-workbook overlays stand in for the undisclosed engine.
Research questions
Can labeled close-only public proxies and the related WeeklyData TRENDEX workbook extract any useful SPY/IEF market-timing overlay from the incomplete PRO>Trendex Charts page?
Data
- database
data/market.sqlite + data/recessionalert.sqlite read-only
- fetched at
2026-08-22T23:39:05Z
- linked assets
0
- page sha256
sha256:1e8169579e056a0f50ab0fc97df47497897c37c9f169128eef54ab910abd4587
- related trendexguide sha256
sha256:a4f92fe9963bc0f8240505c23b1ad0afafdc7d462bfe3171f94581f6ad59ae16
- related workbook sha256
sha256:fc2be5f1fc2a2e38949ae566f6e8cb4699a95562975a498fb5a53510cc40d67c
- related workbook url
- https://recessionalert.com/download/?file=WeeklyData.xlsx
Baseline implementation
- claim kind
documented_but_incomplete
- claim role
causal_primary
- exploratory
True
- family
dem-short-channel10-close-proxy
- id
dem10_causal_lag1
- parameters
- lag bars
1
- lookback
10
What to try interactively
- id
lag_bars
- values
0
1
2
- id
channel_lookback
- values
10
17
21
34
50
72
- id
diffusion_stage
- values
warn32
act21
act10
- id
conviction_threshold
- values
85.0
90.0
95.0
Suggested next research
Can a separately authorized refresh provide source-linked daily Trendex stop/diffusion/probability observations with point-in-time timestamps?
Would intraday high/low history sharpen the channel-stop analogue without being presented as native DEM++?
Does the weekly TRENDEX workbook TDIFF overlay add value at weekly rebalance instead of monthly?
Trial ledger
- iteration
1
- n trials
29
- note
Initial full-grid exploration: 29 labeled variants, causal lag>=1 primary.
- status
published
Agent assessment
Disposition exploratory_tested_proxy under Other/Research.
Explicit NO-GO for playable Other/Indicators native Trendex promotion.
Combinations
- families tested
best-effort-neighbor-grid
conviction85-gated-cross-overlay
dem-long-channel50-close-proxy
dem-medium-channel21-close-proxy
dem-short-channel10-close-proxy
diffusion-3way-aboveness-staging
duration-exceedance-top-probability
duration-exceedance-trough-probability
related-workbook-overlay
- n causal
26
- n control
3
- n trials
29
Historical events
Regime slices on dem21_causal_lag1: GFC/2022 negative; COVID/recent positive.
Parameter sensitivity
Channel windows 10/21/50/72 behave alike on monthly bars; lag 2 matches lag 1.
Conviction thresholds 85/90/95 rarely bind because exceedance catalogs start empty (risk-on default).
Regime behavior
- covid
- cagr
0.1833161370813985
- max drawdown
-0.19415148071011146
- n months
12
- sharpe
0.7767871774015338
- vol
0.2575470524941025
- gfc
- cagr
-0.23316728009010557
- max drawdown
-0.5078480735726956
- n months
21
- sharpe
-1.0334279822061918
- vol
0.22907402590734385
- inflation 2022
- cagr
-0.1817536253627169
- max drawdown
-0.20247013171234207
- n months
12
- sharpe
-0.7570024070758499
- vol
0.23059580676551641
- recent
- cagr
0.21691034586031166
- max drawdown
-0.08325836002549147
- n months
43
- sharpe
1.6083515598744766
- vol
0.12794815457582867
State-space exploration
29 labeled variants across 9 families; negatives retained in the trial ledger.
Timing and cost sensitivity
Causal lag>=1 only; overlay is month-end SPY/IEF without explicit trading-cost drag in this campaign.
What did not work
Channel/diffusion/exceedance analogues converge near buy-and-hold on monthly bars (dem21 risk-on share 100%).
GFC and 2022 regimes were negative for the dem21 proxy.
Lag-0/same-close controls are non-tradable and must not be claimed.
What the signal looks like
Month-end SPY/IEF overlay weights from lagged channel stops, diffusion staging, exceedance ranks, conviction gating, SMA neighbors, and off-row weekly TRENDEX TDIFF.
DSL SignalSpecs are SMA stand-ins; executable primitives live in research/research_campaign/recessionalert_1598.py.
What worked
be_spy_lb12_lag1_t0 (SMA12) and be_spy_lb21_lag1_tneg05 were the strongest causal overlays in-sample.
wb_trendex_overlay_lag1 (weekly TRENDEX TDIFF>=2) cut max drawdown to -0.34 from 2005.
Why the failures appear to happen
Close-only monthly bars cannot reproduce support/resistance stop flips; inferred stops rarely trigger.
Stop formula and 20y window definition are undisclosed, so reconstructions are coarse.
Next questions
Can a separately authorized refresh provide source-linked daily Trendex stop/diffusion/probability observations with point-in-time timestamps?
Would intraday high/low history sharpen the channel-stop analogue without being presented as native DEM++?
Does the weekly TRENDEX workbook TDIFF overlay add value at weekly rebalance instead of monthly?
Review status
published. Research publication does not imply official admission.
Return to pending research