Research campaign · Iteration 1 · unassessed

RecessionAlert Bull Market Safari — inferred public-proxy campaign

Can transparent public proxies test the eight Bull Market Safari observations without claiming native source fidelity?

Expressions: 7
Logged trials: Not recorded
Independent events: Not assessed
Evidence: unassessed

What the research found

Seven inferred/proxy variants execute over the longest checked-in SPY/IEF overlap, while the original source-faithful finding remains insufficient_evidence.

Mechanism and falsifiers

Not recorded in this iteration.

Confidence and limitations

Source has no machine-readable BMS state or complete component series.
Proxy components are inferred and not native RecessionAlert inputs.
Underlying release timestamps, vintages, revision policy, and publication clock are unknown.
SPY/IEF overlay, count thresholds, lookbacks, and holding semantics are research assumptions.
No BUY/SELL, exit, re-entry, target allocation, or official indicator admission is supported.
Lag-zero is a non-causal control only.
Results are descriptive exploratory evidence and not investment advice.

Not recorded in this iteration.

Compare expressions

Download evidence

Exploratory results. Check each period, proxy and cost assumption before comparing. — means not recorded.

Expression / family	CAGR	Sharpe	Max drawdown	Test period	Assessment
bms-equal-weight-lb12-lag1-t0equal weight	—	—	-26.7%	2002-09-30 2026-07-31	exploratory inferred public proxy; not native BMS
bms-equal-weight-lb6-lag1-t0equal weight	—	—	-23.4%	2002-08-31 2026-07-31	exploratory inferred public proxy; not native BMS
bms-equal-weight-lb21-lag1-tneg05equal weight	—	—	-25.4%	2003-06-30 2026-07-31	exploratory inferred public proxy; not native BMS
bms-equal-weight-lb12-lag2-t0equal weight	—	—	-25.6%	2002-10-31 2026-07-31	exploratory inferred public proxy; not native BMS
bms-equal-weight-lb12-lag1-tneg05equal weight	—	—	-25.8%	2002-09-30 2026-07-31	exploratory inferred public proxy; not native BMS
bms-technical-lb12-lag1-t0technical	—	—	-19.7%	2002-08-31 2026-07-31	exploratory inferred public proxy; not native BMS
bms-equal-weight-lb12-lag0-controlequal weight	—	—	-19.3%	2002-08-31 2026-07-31	exploratory inferred public proxy; not native BMS

Open an expression to inspect its rules and request confirmation. The request must be submitted by a trusted repository collaborator.

bms-equal-weight-lb12-lag1-t0

claim kind: inferred_best_guess
classification: exploratory inferred public proxy; not native BMS
count threshold: 4
family: equal_weight
id: bms-equal-weight-lb12-lag1-t0
lag bars: 1
lookback months: 12
source fidelity: False
threshold: 0.0

Confirm as strategy Confirm as indicator

bms-equal-weight-lb6-lag1-t0

claim kind: inferred_nearby_variant
classification: exploratory inferred public proxy; not native BMS
count threshold: 4
family: equal_weight
id: bms-equal-weight-lb6-lag1-t0
lag bars: 1
lookback months: 6
source fidelity: False
threshold: 0.0

Confirm as strategy Confirm as indicator

bms-equal-weight-lb21-lag1-tneg05

claim kind: inferred_nearby_variant
classification: exploratory inferred public proxy; not native BMS
count threshold: 4
family: equal_weight
id: bms-equal-weight-lb21-lag1-tneg05
lag bars: 1
lookback months: 21
source fidelity: False
threshold: -0.05

Confirm as strategy Confirm as indicator

bms-equal-weight-lb12-lag2-t0

claim kind: inferred_nearby_variant
classification: exploratory inferred public proxy; not native BMS
count threshold: 4
family: equal_weight
id: bms-equal-weight-lb12-lag2-t0
lag bars: 2
lookback months: 12
source fidelity: False
threshold: 0.0

Confirm as strategy Confirm as indicator

bms-equal-weight-lb12-lag1-tneg05

claim kind: inferred_nearby_variant
classification: exploratory inferred public proxy; not native BMS
count threshold: 4
family: equal_weight
id: bms-equal-weight-lb12-lag1-tneg05
lag bars: 1
lookback months: 12
source fidelity: False
threshold: -0.05

Confirm as strategy Confirm as indicator

bms-technical-lb12-lag1-t0

claim kind: inferred_family_variant
classification: exploratory inferred public proxy; not native BMS
count threshold: 2
family: technical
id: bms-technical-lb12-lag1-t0
lag bars: 1
lookback months: 12
source fidelity: False
threshold: 0.0

Confirm as strategy Confirm as indicator

bms-equal-weight-lb12-lag0-control

claim kind: control_only
classification: exploratory inferred public proxy; not native BMS
count threshold: 4
family: equal_weight
id: bms-equal-weight-lb12-lag0-control
lag bars: 0
lookback months: 12
source fidelity: False
threshold: 0.0

Confirm as strategy Confirm as indicator

Interactive lab

Interactive history unavailable

Loading available evidence…

Research record

Source claims, inferred rules, experiments and the evidence behind the assessment.

Source

fetched at: 2026-08-22T23:48:19Z
native span: Narrative observations only; no continuous BMS series.
published at: 2023-05-22T17:53:36-0400
sha256: sha256:10b2985e4d23ae06f1011ff5fb80bcef8e25adcd227910ac71e82cb8079da60a
title: Bull Market Safari | RecessionAlert
url: https://recessionalert.com/bull-tracker/

What the source claims

The article names eight bullish-context observations across labor, earnings, global leading data, 200-day technicals, breadth, WLEI, a probability model, and CMHI; it says bearish data outweighs bullish data 2:1.
No common score, state machine, BUY/SELL, re-entry, target, or release ledger is disclosed.
This campaign tests equal-weight, technical, lookback, threshold, and lag neighbors using checked-in SPY/EFA/^VIX/INDPRO/UNRATE proxies.
One available monthly bar of lag is the causal default; lag-zero is retained only as a research control.

Rules actually disclosed

The article names eight bullish-context observations across labor, earnings, global leading data, 200-day technicals, breadth, WLEI, a probability model, and CMHI; it says bearish data outweighs bullish data 2:1.
No common score, state machine, BUY/SELL, re-entry, target, or release ledger is disclosed.

What had to be inferred

This campaign tests equal-weight, technical, lookback, threshold, and lag neighbors using checked-in SPY/EFA/^VIX/INDPRO/UNRATE proxies.
One available monthly bar of lag is the causal default; lag-zero is retained only as a research control.

Research questions

Can transparent public proxies test the eight Bull Market Safari observations without claiming native source fidelity?

Data

database: data/market.sqlite
database access: read-only query_only
native span: No continuous BMS signal span; proxy overlap is reported in lab-series.
private data: False
proxy chain: SPY
IEF
BIL
EFA
^VIX
INDPRO
UNRATE
publication lag: Unknown native release lag; one available month is declared for proxy candidates.
source fetched at: 2026-08-22T23:48:19Z
source sha256: sha256:10b2985e4d23ae06f1011ff5fb80bcef8e25adcd227910ac71e82cb8079da60a
source url: https://recessionalert.com/bull-tracker/

Baseline implementation

claim kind

inferred_best_guess

classification

exploratory inferred public proxy; not native BMS

count threshold

family

equal_weight

bms-equal-weight-lb12-lag1-t0

lag bars

lookback months

metrics

annualized return: 0.09224547648032244
benchmark total return: 11.474840284216487
end: 2026-07-31
max drawdown: -0.26728330712768733
observations: 287
start: 2002-09-30
total return: 7.250737302110341

source fidelity

False

threshold

0.0

What to try interactively

name
lookback_months
values
- 6
- 12
- 21
name
threshold
values
- -0.05
- 0.0
name
lag_bars
values
- 0
- 1
- 2

Suggested next research

Obtain source-linked component series and point-in-time vintages under separate authorization.
Identify BMS aggregation, release, re-entry, and portfolio execution semantics.
Replace synthetic breadth stand-ins with reproducible constituent breadth only when authorized.

Trial ledger

decision
insufficient_evidence
iteration
1
status
published
variants
- bms-equal-weight-lb12-lag1-t0
- bms-equal-weight-lb6-lag1-t0
- bms-equal-weight-lb21-lag1-tneg05
- bms-equal-weight-lb12-lag2-t0
- bms-equal-weight-lb12-lag1-tneg05
- bms-technical-lb12-lag1-t0
- bms-equal-weight-lb12-lag0-control

Next questions

Obtain source-linked component series and point-in-time vintages under separate authorization.
Identify BMS aggregation, release, re-entry, and portfolio execution semantics.
Replace synthetic breadth stand-ins with reproducible constituent breadth only when authorized.

Review status

published. Research publication does not imply official admission.

Return to pending research