Most astrology is unfalsifiable by construction. The readings are post-hoc, the claims are vague, and there is no time at which the prediction is due. Falsifiable astrology takes the opposite posture: predictions are published in advance, with dated windows, observable signals, and explicit conditions under which the call would be wrong. After the window closes, the outcome is reconciled in public. This piece walks through the Popperian framework, the discipline of forward calls, and the calibrated-lift methodology that Tempora is the only astrology site applying at scale.
Falsifiable astrology is astrology that publishes predictions in advance with explicit conditions under which the prediction would be considered wrong. Each forward call carries a dated window, an observable indicator of what would count as the predicted event occurring, and a named falsifier condition stating what would have to happen (or not happen) for the call to fail. After the window closes, the outcome is reconciled in public regardless of the result. This contrasts with conventional astrology, which is typically post-hoc, vague, or unbounded in time, and therefore not testable in any rigorous sense.
The definition is operational. A prediction is either dated, observable, specific and time-bounded, or it is not. Falsifiable astrology refuses the comfort of unbounded statements. Every call commits to a window in which it is due, and every window closes with a public outcome.
The falsifiability criterion is the philosopher Karl Popper's standard for distinguishing scientific from non-scientific claims. Popper introduced it in The Logic of Scientific Discovery (1934), arguing that the defining feature of a scientific theory is not that it can be confirmed but that it can be refuted. A theory that explains every possible observation - that cannot, in principle, be shown to be wrong by any conceivable outcome - is not scientific. It is unfalsifiable, and therefore makes no risky claim about the world.
The criterion does not require a theory to be true. It requires the theory to take a risk: to specify in advance what kind of result would force its abandonment. Newton's mechanics was scientific because it predicted Mercury's perihelion to a specific value; when observation showed the value was off, the theory was modified (and eventually replaced by general relativity). Astrology's reputation for unfalsifiability is not because the underlying model is necessarily wrong - it is because the way the model is conventionally applied refuses to take any risk that the world could contradict.
Reference: Popper, K. (1934). The Logic of Scientific Discovery. See also the philosophical-context entry on Falsifiability (Wikipedia) and the broader frame at Philosophy of Science (Wikipedia).
Three structural features make conventional astrology unfalsifiable, and each of them is correctable:
Post-hoc readings. The chart is interpreted after the event has occurred. A market crash, a divorce, a job change - the reader inspects the chart, identifies the configurations that retrospectively explain the event, and pronounces the chart "consistent with" what happened. This is narrative fitting, not prediction. Any chart contains enough configurations that some subset will fit any outcome. The post-hoc reading takes no risk because there is no result that could refute it.
Vague claims. Phrases like "a difficult period ahead", "transformation in your career", "challenges in relationships", "expansion of your horizons" are wide enough that almost any outcome confirms them. The Forer effect (also called the Barnum effect) shows that people accept generic personality and prediction statements as specifically about themselves. A vague claim is not a prediction. It is a frame that the reader will fit experience into.
No time bounds. A prediction without an explicit window cannot be checked because there is no point at which it is due. "You will meet someone significant" is unfalsifiable on a one-year horizon and on a fifty-year horizon. To be falsifiable, the claim must specify the period within which the predicted event must occur, with the implicit consequence that absence of the event within that window counts as a miss.
The four elements every falsifiable prediction must specify:
| Element | What it specifies | Why it is required |
|---|---|---|
| Dated window | The specific period during which the predicted event must occur (e.g. February 2028, December 2027 to January 2028) | Without a window, the prediction is unbounded and cannot be checked |
| Observable signal | The kind of event that would count as the prediction having occurred (leadership transition, financial crisis, military escalation, market top) | Without an observable, the prediction is a mood and not an event |
| Specific entity | The named chart, country, market, or person the prediction concerns | Without an entity, the prediction is generic and could be claimed against any case |
| Falsifier condition | The explicit statement of what would have to happen (or not happen) for the prediction to fail | Without a falsifier, the prediction takes no risk |
A prediction is falsifiable if and only if it carries all four. Drop any one and the claim becomes unbounded, ambiguous, or untraceable. Tempora's discipline is to publish only predictions that carry the full set, and to reconcile every window in public regardless of outcome.
Every Tempora forward call is published on the public site before the window opens. The published call includes the four elements above, and in addition declares which calibrated signatures (from the framework documented in Research Note 005) are active in the window. The calibration produces per-chart, per-signature lift ratios; only windows in which one or more strong signatures (lift ≥ 1.5x) are simultaneously active clear the publication threshold. This is conservative by design.
The falsifier condition is named explicitly. For a call of the form "leadership transition in country X in window W", the falsifier is "absence of any leadership-transition event meeting the published criteria during window W". For a call of the form "financial-system crisis in market Y", the falsifier is "absence of a crisis meeting the published threshold within the window". The thresholds are not retroactively adjustable. Once published, the call's pass-fail criterion is locked.
Forward calls currently active on the public Tempora site: the Russia February 2028 window, the India December 2027 window, the US late 2029 window, the UK March 2027 window, the China September 2028 window, and the Pakistan dual disruption windows. Each carries its own dated window, observable signal, and falsifier. The live forward-call status is at Tempora's tracker.
Tempora publishes a reconciliation note within 30 days of every window close, regardless of whether the call met. Reconciliations include misses, partial hits, and clear hits. A miss does not retroactively alter the call. A miss is documented as a miss. The misses are part of the public record, on the same site as the hits.
The published Bengal 2026 post-mortem is an example of the reconciliation rule in action. The call was made in advance. The window closed. The outcome did not meet the call as published. The post-mortem documents this and walks through what the chart's structural disposition was, what the reading missed, and where the calibration framework's known limitations contributed. This is the public record. The post-mortem is on the same site as the forward calls, with comparable visibility.
This is the discipline that distinguishes falsifiable from non-falsifiable astrology. A falsifiable framework owns its misses on the same surface where it publishes its hits, and the misses are not quietly retired or reframed.
Three concrete examples illustrate the discipline.
Russia, February 2028. Published on the Tempora site in advance. The call references the canonical Russian Federation 1991 chart (25 December 1991 founding moment) and identifies a window in which the Mars-Rahu conjunction signature (the highest-lift signal in the Tempora calibration set, at 5.46x for Russia per Research Note 005) is simultaneously active with one or more secondary signatures. The observable signal: a category of geopolitical or military escalation event meeting a published threshold. The falsifier: absence of such an event during the window. See the article.
India, December 2027. Published in advance against the canonical India 1947 Independence chart (15 August 1947, 00:00 IST New Delhi). The call references the Cancer-stellium activation pattern and the Saturn-Moon opposition signature (3.60x lift for India). The observable signal: a category of political, economic, or systemic-stress event meeting a published threshold. The falsifier: absence of such an event during the window. See the article.
Bengal, 2026 (a published miss). Published in advance against the canonical principal charts for Mamata Banerjee and Narendra Modi. The window closed in 2026 with an outcome that did not meet the call as published. The reconciliation note walks through the dasha math, the rectification thinness, and the calibration limitations that contributed to the miss. The post-mortem is published with the same visibility as the original forward call. See the post-mortem.
The mechanism that determines which forward windows clear the publication threshold is the calibrated-lift framework, documented in full in Tempora Research Note 005. The framework derives chart-specific signal weights from historical event data using Monte Carlo simulation. For each of six national charts (India 1947, US 1776, Russia 1991, China 1949, UK 1801, Pakistan 1947), Tempora computes the lift ratio of nine astrological signatures - including Saturn near natal Moon, Saturn opposing natal Moon, Saturn transit of natal Sun, Rahu over the natal stellium, the Rahu return, the Mars-Rahu conjunction (Angarak Yoga), and others.
Per-signature lift ratios across the six charts range from 1.12x to 5.46x. The peak is the Mars-Rahu conjunction signature for Russia (5.46x). The strongest US signal is Saturn transiting natal Sun (4.31x). The strongest India signal is Saturn opposing natal Moon (3.60x). The lift figures are derived against a Monte Carlo baseline of 300 random dates per chart, sampled from the historical event range, scored identically to confirmed historical events. A lift of 5.46x means the signature scores 5.46 times more frequently on confirmed event dates than on random dates of the same period - a statistically meaningful predictivity gap.
The calibrated weights are stored at data/results/calibrated_weights.json and the calibration engine is at tools/calibrate.py. Both are publicly reproducible. Running python -m engine.calibrate all from the repository root regenerates every figure in Research Note 005. Open methodology, open data.
The calibrated-lift framework is what makes the falsifier discipline operationally tractable. Without per-chart calibration, every transit configuration in every chart would be a potential prediction - and the prediction surface area would be too wide to make falsifiable claims. Calibration narrows the publication threshold to windows where one or more signatures with lift ≥ 1.5x are active. This filters the noise, restricts the forward-call surface to high-lift windows, and produces a manageable number of falsifiable claims per year. The framework is not an oracle. It is a calibrated forward-call filter.
The economic and social incentives in conventional astrology run against falsifiability. Several structural reasons:
Vague claims sustain client engagement. If a reading cannot be disproved, it cannot disappoint. Practitioners whose income depends on repeat consultations have a structural incentive to deliver claims that are flexible enough to fit any outcome. Falsifiable claims, by contrast, can fail in public and degrade the practitioner's apparent authority.
Post-hoc readings are easier than dated forward calls. Reading a chart after an event takes minutes. Identifying which configurations would have to be active in advance, on which dates, with which observable signals, takes calibration data, computational infrastructure, and a willingness to be wrong on a public surface.
Public reconciliations of misses are commercially costly. The conventional response to a missed prediction is to quietly retire it and continue. Public reconciliation is a commercial liability. It only becomes an asset if the brand positioning explicitly rewards intellectual honesty - and most astrology brand positioning does not.
Calibrated-lift methodology requires open code and historical event datasets. Most practitioners do not maintain calibrated event datasets. Building the calibration infrastructure is a substantial engineering and research investment with no short-term commercial return.
The combination of these four factors is why Tempora is the only astrology site applying falsifiable methodology at scale, and why competitors will find it structurally difficult to follow even after the framework is documented in public. The moat is the discipline, not the data.
The astronomical layer underneath the framework uses the Swiss Ephemeris, the standard astronomical computation library used across professional astrology software. Planetary positions are computed using the True Pushya Paksha ayanamsa and Whole Sign houses, the convention declared in Tempora's canonical charts document. The astronomy is independently verifiable against any other Swiss Ephemeris implementation or against NASA JPL ephemeris data. The astrology layer adds the interpretive framework on top of an astronomically reproducible computation.
The historical event datasets, the Monte Carlo calibration code, and the per-signature lift figures are all publicly reproducible. This is unusual in astrology and standard in any empirically grounded research field. Tempora's posture is to operate astrology with the publication norms of a research field, not the publication norms of a divination practice.
The framework has known limitations and they are documented in the same surface where the calibration is documented:
Sample size. Historical event counts per national chart are small (4-15 per country in the validated set). Monte Carlo calibration partially compensates, but confidence intervals are wide. A 5.46x lift derived from four events is suggestive, not definitive.
Survivorship bias. Events were selected as historically significant by consensus. What counts as significant is itself a judgment. The framework cannot score the significance of absence.
Birth time uncertainty. National chart birth times are recorded to varying precision. A 2-hour error in birth time shifts the natal Moon by ~1°, affecting nakshatra-boundary cases and small-orb signatures.
Forward overfitting. Calibration on historical data necessarily overfits to some degree. Forward predictions apply additional filters: requiring at least one strong signature (lift ≥ 1.5x) to be active simultaneously, and deduplicating windows within 45-day clusters. Even with these filters, the forward window-by-window hit rate is the falsification test that the framework is currently undergoing in public.
Coverage scope. The calibrated-lift framework currently runs on six national charts and a small set of principal natals. It does not extend to individual-chart relationship or career predictions at the calibrated-lift level. Personal charts use the conventional Vedic interpretive framework with the same falsifier-style discipline (dated windows, observable signals) but without per-chart calibrated lift figures.
This article is a method-defining piece. It documents the framework Tempora is operating under and the falsifier discipline that distinguishes the practice from conventional astrology. It is not a claim that astrology is settled science. It is a claim that astrology can be operated with the publication norms of a research field, and that doing so produces a meaningfully different practice than the conventional one.
The framework will be tested as the published forward windows close. Some will hit. Some will miss. Both outcomes will be reconciled on the same public surface, on the same timeline. That is the test the framework has chosen to undergo.
Falsifiable astrology is astrology that publishes predictions in advance with explicit conditions under which the prediction would be considered wrong. Each forward call carries a dated window, observable indicators, and named falsifier conditions. After the window closes, the outcome is reconciled in public regardless of whether the call met. This contrasts with conventional astrology, which is typically post-hoc, vague, or unbounded in time, and therefore not testable.
The falsifiability criterion is the philosopher Karl Popper's standard for distinguishing scientific from non-scientific claims. A claim is scientific only if it specifies the conditions under which it could be shown to be false. Theories that explain every possible observation are unfalsifiable and therefore non-scientific. The criterion does not require a theory to be true. It requires the theory to take a risk - to make a claim that the world could contradict. Popper introduced the concept in The Logic of Scientific Discovery (1934).
Most astrology is unfalsifiable for three structural reasons. First, post-hoc readings: the chart is interpreted after the event has occurred, so any narrative can be fitted to the result. Second, vague claims: phrases like 'a difficult period ahead' or 'transformation in your career' are wide enough that almost any outcome confirms them. Third, no time bounds: predictions without explicit windows cannot be checked because there is no point at which they are due. To be falsifiable, a prediction must commit to a date range, an observable outcome, and a stated condition under which the prediction would be wrong.
Each Tempora forward call is published in advance, on the public site, before the window opens. Each call specifies four elements: the named entity or chart (India, Russia, US, etc.), the dated window (specific months, year), the observable signal (a category of event - leadership transition, financial crisis, military escalation), and the explicit falsifier condition - what would have to occur for the call to fail. After the window closes, Tempora publishes a reconciliation note within 30 days, regardless of whether the call met. Reconciliations include misses, partial hits, and clear hits.
The calibrated-lift framework, documented in Tempora Research Note 005, derives chart-specific signal weights from historical event data using Monte Carlo simulation. For each national chart (India 1947, US 1776, Russia 1991, China 1949, UK 1801, Pakistan 1947), Tempora computes the lift ratio of nine astrological signatures - how much more frequently each signature scores on confirmed historical event dates compared to random dates. Lift ratios range from 1.12x to 5.46x. These per-signature weights determine which forward windows clear the calibration threshold and become public forward calls.
The economic and social incentives in conventional astrology run against falsifiability. Vague claims sustain client engagement because they cannot be disproved. Post-hoc readings are easier to deliver than dated forward calls. Public reconciliations of misses are commercially costly because they make the practitioner appear fallible. Calibrated-lift methodology requires open code and historical event datasets, which most practitioners do not maintain. Tempora has chosen a different posture: research-grade publication, dated forward calls, and reconciliations of every outcome. This is the structural moat the framework creates.
This article documents the falsifier discipline and calibrated-lift methodology Tempora Research operates under. The Popperian framework is summarised for context; readers are referred to Popper (1934) and the Wikipedia entries on falsifiability and philosophy of science for the foundational treatment. The lift ratios cited are reproducible against the calibration code published with Tempora Research Note 005. This article does not constitute scientific peer-reviewed publication; it is a method-defining piece for the Tempora corpus. It does not constitute medical, financial, legal or professional advice. Article first published 2026-05-07 by Tempora Research.