# EDL Heritage and Landing-Success Hazard: Does Reuse of Flight-Proven Entry-Descent-Landing Architecture Reduce Landing-Failure Risk?

**Candidate:** JPL_AUTONOMY_EDL_05
**Program:** COLLEGIUM 1st Battalion
**NORTH STAR / JPL category:** Entry Descent & Landing Systems
**Methodological anchors:** Robert W. Fogel (cliometric counterfactual analysis); Joel Mokyr (economic history of useful knowledge and cumulative innovation)
**Date:** 2026-06-15

---

## Abstract

Planetary landing remains the highest-risk phase of robotic and crewed missions to bodies with or without atmospheres. Practitioners commonly assert that reusing a flight-proven entry-descent-landing (EDL) architecture lowers landing-failure risk, but the claim is rarely tested as a falsifiable quantitative hypothesis against the full historical record of landing attempts. This dissertation specifies and operationalizes such a test. The contribution is a single falsifiable proposition: planetary landing attempts that reuse a flight-proven EDL architecture lineage exhibit a measurably lower landing-failure hazard than attempts that introduce novel EDL elements, after controlling for target body, entry mass, and landed mass. The null is that EDL architectural novelty has no effect on landing-failure probability. The unit of analysis is the individual landing attempt at the Moon, Mars, or Titan. The primary outcome is a discrete landing success or failure indicator. The principal regressor is an EDL-heritage-reuse index constructed from documented architectural lineage in NASA Technical Reports Server (NTRS) EDL reconstruction reports, TechPort technology readiness records, and the global record of landing attempts, with program-history context from U.S. Government Accountability Office (GAO) reports. The estimator is a discrete-outcome logistic hazard model. Following Fogel, the heritage effect is framed as a counterfactual: the landing-failure probability a given attempt would have faced had it used the next-best heritage alternative. Following Mokyr, novelty is decomposed into propositional (analytically grounded) versus prescriptive (trial-based) elements to test whether the heritage effect operates through codified, verifiable knowledge rather than reputation. This is presented as a design-stage analysis plan. Expected and illustrative results are labeled as not yet executed on the full dataset. The design states the conditions under which the contribution would be falsified, including a near-zero or wrong-signed heritage coefficient, and confounding by mission cost, organizational maturity, or target-body difficulty. The work supports NASA and JPL EDL portfolio decisions by clarifying when heritage reuse is genuinely protective and when it is a proxy for unmeasured program strength.

---

## 1. Introduction and Contribution

### 1.1 The problem

Entry, descent, and landing concentrates a disproportionate share of total mission risk into a few minutes of irreversible, largely autonomous operation. The Mars Science Laboratory (MSL) EDL sequence, the Mars 2020 sequence, and the various lunar soft-landing attempts of the past decade all demonstrate that a single architectural element behaving outside its qualified envelope can end a multi-billion-dollar mission [1][6][7]. The record of recent lunar attempts, including the loss of the Beresheet and Vikram landers and the partial anomalies of several commercial landers, has renewed a practitioner debate that is old in spaceflight: does sticking with a flight-proven architecture meaningfully reduce the chance of losing the vehicle, or does heritage merely correlate with other strengths such as funding, schedule, and institutional experience [12][13]?

The intuition behind heritage reuse is straightforward. A flight-proven EDL architecture has survived the full chain of qualification, integration, and at least one real atmospheric or powered descent. Its failure modes have been observed, reconstructed, and in many cases retired. A novel architecture, by contrast, carries irreducible uncertainty: ground testing cannot fully reproduce the coupled aerothermal, aerodynamic, and control environment of a real descent, and some failure modes are only discoverable in flight [11][14][15]. Yet the intuition is not self-evidently correct. Heritage hardware can be flown outside its original envelope, where its proven status is illusory. Novel elements are sometimes introduced precisely because the heritage element was known to be inadequate for a new target or mass class [3][4]. And the missions that can afford extensive heritage reuse may also be the ones with the deepest engineering reserves, so that any observed heritage advantage could be confounded by program strength rather than caused by the architecture itself.

The recent record sharpens the question rather than settling it. The lunar surface has seen a sequence of attempts whose architectures span the full heritage spectrum, from government landers that inherit decades of descent-engine and guidance lineage to first-flight commercial vehicles whose EDL software and propulsion had never executed a real powered descent. Several of these attempts failed or landed anomalously, and at least one has been the subject of a formal interpreted investigation report [12]. On Mars, by contrast, the United States has assembled an unusually clean chain of near-replications: Phoenix inherited Mars Polar Lander and Mars Surveyor heritage, InSight reused the Phoenix architecture almost without change, and Mars 2020 reused the Mars Science Laboratory sky-crane lineage with bounded additions [5][6][16]. Independent lineages, such as Tianwen-1, reached the same successful outcome class through a largely separate architectural path [20]. A naive reading of these cases could support either side of the debate. The Mars chain looks like evidence that heritage protects; the lunar record looks like evidence that novelty is dangerous; but neither reading controls for the physical difficulty of the target or for the depth of the program behind each attempt. That is precisely why a coefficient estimated against the whole population, with controls, is needed rather than a curated set of anecdotes.

### 1.2 The gap in the literature

The EDL engineering literature is rich in single-mission reconstruction and in forward-looking architecture studies, but it does not contain a systematic, multi-mission statistical test of the heritage hypothesis stated as a falsifiable proposition. Reconstruction reports document, in fine detail, how one EDL sequence performed against its predictions [1][6][7][16][17]. Architecture studies quantify the technology gaps and decelerator trade spaces for future high-mass or human-class Mars EDL, but they reason about a hypothetical future fleet rather than estimating an effect from the historical record [3][4][8][9][10]. Reliability studies of spacecraft populations exist, but they typically address on-orbit longevity of operating satellites rather than the discrete success or failure of a landing event [18][19]. The reliability literature on deployable and high-risk subsystems is closer in spirit but does not isolate EDL architectural novelty as a regressor [18]. No published study, to this candidate's knowledge, estimates a landing-failure hazard as a function of a constructed EDL-heritage-reuse index across the Moon, Mars, and Titan record while controlling for the obvious physical confounders of target body and mass.

This is the gap the dissertation fills. The contribution is not a new EDL technology, a new reconstruction, or a new architecture concept. It is a quantitative, hypothesis-testing treatment of an assertion that the engineering community makes constantly but rarely tests against the full population of attempts.

### 1.3 The single falsifiable contribution

The contribution is stated as one pair of hypotheses about one coefficient.

- **H1 (contribution):** Conditional on target body, entry mass, and landed mass, the coefficient on the EDL-heritage-reuse index in a logistic model of landing failure is negative and statistically distinguishable from zero. Higher documented reuse of a flight-proven EDL lineage is associated with a lower landing-failure hazard.
- **H0 (null):** Conditional on the same controls, the coefficient on the EDL-heritage-reuse index is zero. EDL architectural novelty has no effect on landing-failure probability.

The proposition is falsifiable in the strict sense: a fitted heritage coefficient that is zero, positive, or negative-but-statistically-indistinguishable-from-zero would fail to reject H0 and would falsify the contribution. The design specifies in advance the estimator, the controls, the dataset, and the falsification conditions, so that the result is not a function of post hoc specification choices.

### 1.4 Why it matters for NASA and JPL

NASA and JPL make recurring portfolio decisions in which the heritage-versus-novelty trade is explicit: whether to fly an established sky-crane lineage again, whether to introduce supersonic retropropulsion or inflatable decelerators for higher-mass Mars payloads, and how to weigh a commercial lander's novel architecture against the firm's lack of flight-proven EDL [3][4][9][20]. If the heritage effect is real and large, the portfolio implication is to weight reuse heavily and to demand more flight-like qualification for novel elements. If the heritage effect is small once program strength is controlled, the implication is that resources are better spent on the underlying engineering reserves than on heritage per se. Either finding is decision-relevant, which is the mark of a contribution worth defending. The cliometric tradition this dissertation draws on, beginning with Fogel's insistence that an innovation's value is defined only relative to the next-best counterfactual, is precisely the discipline NASA needs to avoid treating heritage as an absolute good [Fogel framework].

---

## 2. Background and Literature

### 2.1 The EDL record as a population of natural experiments

The historical record of landing attempts at the Moon, Mars, and Titan is, from a statistical standpoint, a population of discrete, well-documented trials with a binary outcome. Mars EDL is the most thoroughly reconstructed segment. The MSL system overview and its in-flight performance and reconstruction reports describe a guided-entry, supersonic-parachute, powered-descent, sky-crane architecture whose elements were deliberately traced to, and deliberately departed from, Mars Pathfinder and Mars Exploration Rover heritage [1][6][7][16]. The Phoenix lander reused Mars Polar Lander and Mars Surveyor 2001 heritage with documented architectural lineage, and its EDL performance was reconstructed in detail [5]. Mars 2020 reused the MSL sky-crane lineage with incremental additions such as terrain-relative navigation and the MEDLI2 instrumentation suite [6][17]. InSight reused the Phoenix EDL architecture almost directly, providing one of the cleanest near-replication cases in the record [16]. Tianwen-1 represents a largely independent architectural lineage reaching the same outcome class [20]. Each of these is a documented data point with a known architecture, a known mass class, and a known outcome.

The lunar record adds attempts with shorter or absent atmospheres, where the EDL problem is dominated by powered descent, guidance, navigation, and terminal hazard avoidance rather than aerothermal entry. Recent lunar attempts span a wide range of architectural novelty, from heritage-derived government landers to first-flight commercial architectures, and several recent losses have been formally investigated, including the Vikram lander loss during the lunar landing phase [12][13]. The SLIM small lunar lander demonstrated a pinpoint landing with a novel architecture and a partially anomalous touchdown attitude, illustrating that novelty and outcome do not map one-to-one [21]. Titan contributes a single but important data point in Huygens, a probe descent whose architecture was effectively novel for its target and whose outcome is part of the record of in-situ giant-planet and moon entry [22]. The Titan case matters less for statistical power than for external validity: it tests whether any heritage effect estimated mostly on Mars and the Moon generalizes to a chemically and dynamically different target.

### 2.2 Architecture studies and the technology-gap framing

A second body of literature reasons forward rather than backward. The NASA EDL Systems Analysis studies and the high-mass Mars concept work quantify the decelerator and propulsion trade space for payloads beyond the proven MSL class, and they identify supersonic retropropulsion and inflatable aerodynamic decelerators as the principal novel technologies required [3][4][8]. The human Mars EDL architecture studies extend this to crewed-class masses and enumerate technology development gaps and their mitigations [9][10]. Supersonic retropropulsion in particular has a documented maturation history through wind-tunnel and analysis campaigns, which makes it a useful test case for the propositional-versus-prescriptive decomposition described below [11][14]. These studies are the source of the novelty side of the heritage index: they identify, with citations and technology readiness assessments, exactly which EDL elements would be new relative to flown heritage.

### 2.3 The reliability and failure literature

A third body of work addresses spacecraft reliability statistically. Population-level analyses of deep-space and deployable spacecraft estimate failure distributions and infant-mortality effects across launch cohorts [18]. Studies of spacecraft longevity and of failure prediction from telemetry establish that discrete-outcome and time-to-failure modeling are accepted tools in the domain [19]. Work on fatal software failures in spaceflight catalogs a class of failure modes that are disproportionately associated with novel or modified flight software, which is directly relevant because EDL novelty often resides in guidance and control software rather than in structures [15]. This literature supplies the statistical precedent for a logistic or hazard treatment but has not been applied to the specific question of EDL architectural heritage.

### 2.4 The Fogel framework applied in plain language

Robert Fogel's cliometric program rests on one rule: to measure the value of an innovation, you must specify and quantify the world without it, the counterfactual, not merely observe the world with it [Fogel dossier; 23]. An economy never depends on a technology in the absolute; it depends on it only relative to the next-best substitute. Fogel's railroad study constructed an explicit counterfactual transport system to compute the social saving of rail, and the central methodological lesson is that an apparently indispensable technology can have a small marginal effect once the substitute is correctly specified [23].

Applied here, the rule says that the question is not whether heritage-reuse missions succeed, which they mostly do, but whether they succeed more than they would have under the next-best counterfactual, which is the same mission carried out with a novel architecture at the same target and mass. The heritage-reuse index and the logistic model are the apparatus for estimating that counterfactual difference. Fogel's later warning about partial counterfactuals also applies: a single-coefficient estimate omits induced effects, such as the possibility that the availability of heritage changes which missions are attempted at all. The design addresses this through controls and through an explicit discussion of selection.

### 2.5 The Mokyr framework applied in plain language

Joel Mokyr's economic history of useful knowledge distinguishes propositional knowledge, the understanding of why a technique works, from prescriptive knowledge, the recipe for doing it [Mokyr dossier; 24]. Techniques that rest on deep propositional understanding are extensible and self-correcting; techniques discovered by trial without underlying theory tend to stagnate and to fail unpredictably outside their tested range [24]. Mokyr also stresses that progress is cumulative and that the conversion of isolated invention into reliable innovation depends on codification and public verification of knowledge.

Applied here, Mokyr's distinction generates a sharper version of the heritage hypothesis. Heritage may protect against failure not because the hardware is old but because flight has converted prescriptive recipes into propositional understanding: the failure modes are now understood, not merely avoided. This predicts that the protective effect of heritage should be concentrated in EDL elements whose flight history produced codified, reconstructed knowledge, of which the NTRS reconstruction reports are the literal artifact. It also predicts that novelty grounded in strong propositional analysis, such as supersonic retropropulsion developed through extensive wind-tunnel and computational campaigns, should carry less excess risk than novelty introduced without such grounding [11][14]. The design operationalizes this as a decomposition of the novelty term, described in Section 4.

Two further Mokyr propositions bear on the design. First, Mokyr emphasizes that technological progress is fragile and reversible, and that incumbents and risk-averse institutions can resist novelty even when it is warranted. In the EDL context the resistance runs the other way as well: an institution that over-weights heritage can fly a proven element outside its qualified regime, converting an apparent strength into a hidden hazard. The heritage index is therefore coded against the regime in which an element was proven, not merely against whether it flew before, so that an element flown well outside its envelope receives a low rather than a high heritage score. Second, Mokyr's account of cumulative innovation implies that the reliability dividend of heritage is not a one-time step but a function of how thoroughly the prior flight was reconstructed and codified. A mission that flew but was poorly reconstructed yields weak propositional knowledge and should confer a smaller protective effect than a mission whose EDL was instrumented and reconstructed in fine detail, such as those carrying the MEDLI and MEDLI2 suites [17]. This is the conceptual justification for weighting the heritage index by the depth of reconstruction documented in NTRS, rather than treating all prior flights as equivalent. Perez's complementary account of how isolated inventions consolidate into reliable techno-economic paradigms reinforces the same point in plain terms: reliability is earned through accumulation and codification, not conferred by a single demonstration [24].

---

## 3. Data

### 3.1 Named datasets and sources

The analysis draws on four named, real data sources.

1. **NTRS EDL reconstruction reports.** The NASA Technical Reports Server hosts the authoritative post-flight reconstruction reports for U.S. Mars EDL events, including Mars Pathfinder atmospheric entry reconstruction, the Mars Exploration Rovers EDL trajectory analysis, the Phoenix EDL performance report, the MSL trajectory and atmosphere reconstruction, the InSight reconstruction, and the MEDLI and MEDLI2 instrumentation reconstructions [16][17, and NTRS IDs 20210005354, 20040095912, 20080034645, 20130010087, 20200002910/20200003204]. These documents define each architecture's elements and their measured in-flight behavior. Access path: the NTRS public API at `https://ntrs.nasa.gov/api/citations/search` and the NTRS web interface. They are the primary source for coding architectural lineage and for the propositional-knowledge weighting of the heritage index.

2. **The global record of Mars, Moon, and Titan landing attempts.** This is the population frame: every documented attempt to land a vehicle on these three bodies, with its outcome. It is assembled from mission overviews and catalogs in the peer-reviewed and agency literature [1][5][6][7][20][21][22] and from consolidated mission catalogs [25]. Each attempt contributes one row.

3. **TechPort EDL-technology TRL records.** NASA TechPort records the technology readiness level history of EDL technologies, including decelerators, retropropulsion, terrain-relative navigation, and landing radar [11][14, and the NASA Technology Taxonomy 20190029323]. Access path: the TechPort public portal and API. These records date when a given EDL element first reached flight-proven status, which is the basis for classifying an element as heritage or novel for a given attempt.

4. **GAO program-history reports.** GAO reports on NASA major projects and on specific Mars and lunar programs supply independent program-level context: cost, schedule, and organizational assessments used to construct the program-strength control that guards against confounding. Access path: the GAO public report archive.

### 3.2 Unit of analysis

The unit of analysis is the individual landing attempt at the Moon, Mars, or Titan, defined as a vehicle committing to an EDL or powered-descent sequence intended to place a payload on the surface. Orbital insertions, flybys, and sample-return Earth entries are excluded. The Mars Sample Return Earth Entry Vehicle and similar Earth-return systems are excluded from the primary population because Earth entry is a different target body, though they are noted as a possible robustness extension.

### 3.3 Outcome variable

The dependent variable is a binary landing outcome: failure equals one if the vehicle did not achieve a survivable surface placement enabling nominal post-landing operations, and zero otherwise. Partial successes, such as a hard landing with loss of function or a successful touchdown with anomalous attitude, are coded against a pre-registered rule: loss of the mission's primary surface function is coded as failure; a survivable landing with degraded but operable function is coded as success, with a sensitivity analysis that recodes the boundary cases. The SLIM anomalous-attitude landing and similar cases are the explicit boundary cases for this rule [21].

### 3.4 Variable construction: the EDL-heritage-reuse index

The principal regressor is a continuous EDL-heritage-reuse index in the interval from zero to one. The architecture of each attempt is decomposed into a fixed set of EDL functional elements: aeroshell and thermal protection, entry guidance, supersonic deceleration (parachute or retropropulsion), terminal descent and propulsion, terminal guidance and hazard avoidance, and touchdown mechanism. For each element, a heritage score is assigned from the TechPort TRL history and the NTRS lineage documentation: an element scores high if it had been flown successfully in a comparable regime on a prior mission and low if it is introduced for the first time or flown well outside its proven envelope. The index is the mass-or-criticality-weighted mean of the element scores, with weights fixed in advance. The index is deliberately constructed from documented sources rather than from expert opinion, so that it is reproducible from the named datasets.

### 3.5 Controls

Three physical controls are specified in the contribution: target body (a categorical control for Moon, Mars, Titan), entry mass, and landed mass. Target body absorbs the gross difficulty differences among an airless body, a thin-atmosphere body, and a thick-atmosphere body. Entry and landed mass absorb the well-documented scaling of EDL difficulty with mass, which is the entire motivation for the high-mass architecture studies [3][4][9]. A fourth control, a program-strength index built from GAO cost-and-schedule data and from organizational flight experience, is included in the main specification to address the central confounding threat that heritage proxies for program strength.

### 3.6 Coverage and limitations

Coverage spans the documented Mars, Moon, and Titan landing attempts from the earliest reconstructed events through 2026. The frame is intrinsically small by statistical standards: the total number of attempts is on the order of several dozen, with Mars and the Moon contributing most rows and Titan contributing one. This limits statistical power and motivates the careful, pre-registered, low-parameter specification described next. A second limitation is documentation asymmetry: U.S. and European attempts are far better reconstructed than some others, which can bias the heritage index where lineage is poorly documented; the design flags every low-documentation row and tests sensitivity to their exclusion. A third limitation is that outcome and heritage are both coded by the analyst from documents, introducing construct risk that Section 4 addresses through inter-coder checks and pre-registration.

---

## 4. Research Design and Identification

### 4.1 Estimator

The primary estimator is a logistic regression of the binary landing-failure outcome on the EDL-heritage-reuse index and controls. Because each attempt is a single discrete trial, the logistic model is the discrete-outcome hazard for a one-shot event: it estimates the probability that the vehicle is lost during the EDL event conditional on the covariates. The model is

logit Pr(failure_i = 1) = beta_0 + beta_1 (heritage_index_i) + gamma' (controls_i) + epsilon_i,

where the controls vector contains target-body indicators, entry mass, landed mass, and the program-strength index. The contribution lives entirely in beta_1. H1 predicts beta_1 < 0 and statistically distinguishable from zero; H0 predicts beta_1 = 0.

Given the small sample, the primary fit uses penalized likelihood (Firth's bias-reduced logistic regression) to handle small-sample bias and any quasi-separation, with exact or permutation-based inference for the heritage coefficient. A complementary specification uses a discrete-time complementary-log-log hazard to confirm that the conclusion is not an artifact of the logistic link.

### 4.2 Identification strategy

Identification rests on conditional comparison: among attempts at the same target body and similar mass, do those with higher documented heritage reuse fail less often? The physical controls remove the most obvious sources of spurious correlation, and the program-strength control removes the most dangerous confounder. The Fogelian counterfactual makes the identifying claim explicit: beta_1 estimates the change in failure probability that would occur if a given attempt's heritage index were lowered toward the novel end while holding target, mass, and program strength fixed. This is a partial counterfactual in Fogel's sense, and the design is candid that it does not capture general-equilibrium effects such as heritage availability changing which missions are attempted [23].

The Mokyr decomposition sharpens identification by splitting the novelty captured in the heritage index into propositionally grounded novelty (elements with extensive analytical and ground-test maturation, such as supersonic retropropulsion) and ungrounded novelty (elements flown first with thin analytical backing) [24][11][14]. If the heritage effect operates through codified knowledge rather than mere age, the excess risk should load on ungrounded novelty. This is a testable sub-implication that strengthens or weakens the causal interpretation without changing the headline hypothesis.

### 4.3 Threats to validity

**Internal validity.** The dominant threat is confounding by program strength: well-funded, experienced programs both reuse heritage and execute better. The program-strength control directly targets this, and a sensitivity analysis reports beta_1 with and without the control to bound the confounding. A second threat is reverse causation in element selection: novelty is sometimes introduced because heritage was inadequate for the target, which would attenuate beta_1 toward zero and bias against H1, making any rejection of H0 conservative. A third threat is coding endogeneity, where knowledge of the outcome contaminates the heritage coding; pre-registration of the index rules and blind coding of architecture before outcome address this.

**External validity.** Estimates dominated by Mars and the Moon may not generalize to Titan or to outer-planet entries. The single Titan data point is retained precisely to probe this, and the design states that the contribution's external claim is limited to the three named bodies. Generalization to crewed-class masses is explicitly out of scope, since no crewed landing attempts exist in the frame.

**Construct validity.** The heritage index is a constructed proxy for an unobservable, the true architectural similarity to flown systems. The element-wise, document-based construction and an inter-coder reliability check on a subsample defend the construct. The outcome construct is defended by the pre-registered partial-success coding rule and its sensitivity analysis.

**Statistical-conclusion validity.** The small sample is the central threat. Penalized likelihood, exact or permutation inference, and a pre-registered single primary specification guard against both small-sample bias and specification search. The design reports the minimum detectable effect size given the realized sample, so that a failure to reject H0 can be correctly interpreted as either a true null or as insufficient power, not conflated.

### 4.4 Specification discipline and pre-registration

The small frame makes specification discipline non-negotiable. The design fixes, before any model is fitted, a single primary specification: Firth-penalized logistic regression of the binary failure indicator on the heritage index, the three physical controls, and the program-strength control, with exact or permutation inference on the heritage coefficient. The number of estimated parameters is kept deliberately low relative to the number of attempts to avoid overfitting a few dozen rows. The heritage-index weights, the partial-success coding rule, the program-strength index construction, and the robustness set are all written down in advance, so that the headline result cannot be a product of searching over specifications until the coefficient turns significant. This is the statistical analogue of Fogel's insistence that the counterfactual be specified explicitly rather than chosen to fit the conclusion: the comparison is fixed before the data speak [23].

The robustness set is itself pre-registered and is reported in full regardless of whether it strengthens or weakens H1. It comprises the complementary-log-log link as an alternative to the logit; exclusion of low-documentation rows whose heritage coding is uncertain; recoding of the partial-success boundary cases such as SLIM in both directions [21]; estimation with and without the program-strength control to bound confounding; and the Mokyr decomposition that splits novelty into propositionally grounded and ungrounded components [11][14][24]. A contribution that survives only the primary specification but collapses across the robustness set is reported as fragile, not as confirmed.

---

## 5. Analysis Plan and Findings

**This section is a design-stage analysis plan. No results below are executed on the full dataset. All numbers are illustrative placeholders that show the intended form of the output, not empirical findings.**

### 5.1 Estimation procedure

The analysis proceeds in pre-registered steps. First, assemble the population frame from the named datasets and freeze it. Second, code each attempt's six EDL elements for heritage and assemble the index using the fixed weights, with a second coder independently coding a random subsample for reliability. Third, code outcomes under the pre-registered rule, blind to the heritage coding. Fourth, fit the primary Firth-penalized logistic model with the three physical controls and the program-strength control. Fifth, report beta_1, its exact or permutation-based confidence interval, and the implied change in failure probability across the interquartile range of the heritage index. Sixth, run the pre-specified robustness set: complementary-log-log link, dropping low-documentation rows, recoding boundary partial-successes, removing the program-strength control to bound confounding, and the Mokyr propositional-versus-ungrounded novelty decomposition. Seventh, report the minimum detectable effect size and a power assessment.

### 5.2 Expected (illustrative) findings

Under H1, the expected output is a negative beta_1 whose exact confidence interval excludes zero, corresponding to a meaningfully lower modeled failure probability for high-heritage attempts than for low-heritage attempts at the same target and mass. For example, an illustrative result of the form "a move from the lowest to the highest quartile of the heritage index is associated with a reduction in modeled landing-failure probability from roughly forty percent to roughly fifteen percent at fixed Mars mass" would support H1. These figures are placeholders chosen to show the reporting format; they are not estimates from data.

Under the Mokyr decomposition, the expected pattern if the heritage effect is knowledge-based rather than age-based is that the excess-risk loading concentrates on ungrounded novelty, with propositionally grounded novelty such as supersonic retropropulsion carrying a smaller risk premium. A finding that all novelty carries equal excess risk regardless of analytical grounding would still be consistent with H1 but would weaken the Mokyr interpretation.

### 5.3 What a null or contrary result would look like

If the fitted beta_1 is indistinguishable from zero after controlling for program strength, H0 is not rejected and the contribution is falsified: heritage would be shown to carry no independent protective effect once program strength and physics are accounted for. If beta_1 is large and negative without the program-strength control but collapses toward zero when it is added, the honest conclusion is that the apparent heritage effect is largely confounded, which is itself a decision-relevant finding for NASA portfolio reasoning. The analysis plan commits in advance to reporting all three of these outcomes with equal prominence.

---

## 6. Discussion

### 6.1 Implications

If H1 survives the full robustness set, the practical implication is that EDL heritage reuse has independent protective value beyond the program strength it correlates with, and that the value is concentrated in elements whose flight history produced codified, reconstructed knowledge. This would justify, in quantitative terms, the conservative EDL-lineage strategy that programs such as InSight followed in reusing the Phoenix architecture, and it would set a measurable bar for how much extra qualification a novel element should receive to offset its excess risk [16][5]. If instead the heritage effect is largely confounded by program strength, the implication is that NASA and JPL should invest in the underlying engineering reserves and verification rigor rather than treating heritage as a goal in itself, a conclusion squarely in the Fogelian spirit that a technology's value is only ever relative to its substitute [23].

### 6.2 Rival explanations

The principal rival explanation is the confounding-by-program-strength account, addressed directly by the program-strength control and the with-and-without sensitivity analysis. A second rival is target-difficulty selection: heritage may be reused mainly on easier missions, so that the heritage rows are systematically less demanding. The target-body and mass controls absorb the measurable part of this, and the design flags the residual as a limitation. A third rival is documentation bias, where well-documented heritage missions are coded more favorably; the inter-coder check and low-documentation sensitivity analysis bound this. A fourth rival, consistent with Mokyr, is that the real driver is codified knowledge rather than heritage hardware, in which case the heritage index is a proxy for knowledge maturity; the propositional-versus-ungrounded decomposition is designed to surface exactly this and to reframe rather than refute the contribution if it holds [24].

### 6.3 External validity

The contribution's external claim is bounded to landing attempts at the Moon, Mars, and Titan within the documented historical mass range. It does not claim to predict crewed-class Mars EDL, for which no in-frame data exist and for which the architecture studies explicitly warn that novel decelerators are unavoidable [9][10]. The single Titan data point tests, but cannot establish, generalization beyond Mars and the Moon. Extension to outer-planet and ocean-world entries is identified as future work [22].

### 6.4 What would falsify the contribution

The contribution is falsified by any of the following pre-registered conditions: a heritage coefficient that is zero or positive in the main specification; a heritage coefficient that is negative without the program-strength control but indistinguishable from zero with it; or a demonstration that the heritage index is collinear with the program-strength index to the point that the two cannot be separately identified, in which case the claim that heritage has independent value cannot be sustained. Stating these conditions in advance is what makes the proposition a genuine hypothesis rather than a narrative.

---

## 7. Contribution and Conclusion

This dissertation does not introduce a new EDL technology or reconstruction. Its contribution is methodological and evidentiary: it converts a constant practitioner assertion, that flight-proven EDL heritage lowers landing-failure risk, into a single falsifiable proposition and specifies, in advance and in full, the dataset, estimator, controls, and falsification conditions required to test it. The design fuses two methodological anchors. From Fogel it takes the discipline of the counterfactual: heritage is valued not by the success of heritage missions but by the difference between their outcomes and those of the next-best novel-architecture counterfactual at the same target and mass [23]. From Mokyr it takes the distinction between propositional and prescriptive knowledge, which reframes any heritage effect as plausibly a knowledge-codification effect and provides a testable decomposition that flight history, captured in NTRS reconstruction reports, converts recipes into understanding [24].

The work is presented honestly as a design-stage analysis plan. The illustrative figures in Section 5 are placeholders, and the central deliverable at this stage is the pre-registered, falsifiable design itself. For NASA and JPL, the value is that whichever way the coefficient falls, the result is decision-relevant: a real and unconfounded heritage effect justifies a conservative lineage strategy and sets a measurable qualification bar for novelty, while a confounded or null effect redirects investment from heritage as a goal to the engineering reserves and verification rigor that heritage merely proxies. The next step is execution of the frozen analysis plan on the assembled population, followed by the full pre-registered robustness set.

---

## References

1. Steltzner, A. D., et al. "Mars Science Laboratory Entry, Descent, and Landing System Overview." IEEE Aerospace Conference, 2008. doi: [10.1109/aero.2008.4526283](https://doi.org/10.1109/aero.2008.4526283)
2. Prakash, R., et al. "Mars Science Laboratory Entry, Descent, and Landing System." IEEE Aerospace Conference, 2006. doi: [10.1109/aero.2006.1655796](https://doi.org/10.1109/aero.2006.1655796)
3. Dwyer Cianciolo, A., et al. "Entry, Descent and Landing Systems Analysis Study: Phase 1 Report." NASA Technical Report, 2010. NTRS: [https://ntrs.nasa.gov/citations/20100033754](https://ntrs.nasa.gov/citations/20100033754)
4. Korzun, A. M., Dubos, G. F., Iwata, C. K., et al. "A concept for the entry, descent, and landing of high-mass payloads at Mars." Acta Astronautica, 2010. doi: [10.1016/j.actaastro.2009.10.003](https://doi.org/10.1016/j.actaastro.2009.10.003)
5. Desai, P. N., et al. "Entry, Descent, and Landing Performance of the Mars Phoenix Lander." Journal of Spacecraft and Rockets, 2011. doi: [10.2514/1.48239](https://doi.org/10.2514/1.48239)
6. Way, D. W., et al. "Mars 2020 Entry, Descent, and Landing System Overview." IEEE Aerospace Conference, 2019. doi: [10.1109/aero.2019.8742167](https://doi.org/10.1109/aero.2019.8742167)
7. Steltzner, A. D., et al. "Mars Science Laboratory: Entry, Descent, and Landing System Performance." IEEE Aerospace Conference, 2007. doi: [10.1109/aero.2007.352821](https://doi.org/10.1109/aero.2007.352821)
8. Dwyer Cianciolo, A., et al. "Overview of the NASA Entry, Descent and Landing Systems Analysis Study." AIAA, 2010. doi: [10.2514/6.2010-8649](https://doi.org/10.2514/6.2010-8649)
9. Polsgrove, T., et al. "Human Mars Entry, Descent and Landing Architecture Study Overview." AIAA SPACE, 2016. doi: [10.2514/6.2016-5494](https://doi.org/10.2514/6.2016-5494)
10. Cianciolo, A. D., et al. "Human Mars Entry, Descent, and Landing Architecture Study: Rigid Decelerators." AIAA SPACE, 2018. doi: [10.2514/6.2018-5192](https://doi.org/10.2514/6.2018-5192)
11. Edquist, K. T., Korzun, A. M., Dyakonov, A. A., et al. "Development of Supersonic Retropropulsion for Future Mars Entry, Descent, and Landing Systems." Journal of Spacecraft and Rockets, 2014. doi: [10.2514/1.a32715](https://doi.org/10.2514/1.a32715)
12. "Interpreted Investigation Report: Loss of Vikram Lander During Lunar Landing Phase." 2023. doi: [10.61359/11.2106-2309](https://doi.org/10.61359/11.2106-2309); preprint doi: [10.48550/arxiv.2309.14384](https://doi.org/10.48550/arxiv.2309.14384)
13. Wander, A. "Fatal Software Failures in Spaceflight." Encyclopedia, 2024. doi: [10.3390/encyclopedia4020061](https://doi.org/10.3390/encyclopedia4020061)
14. Berry, S. A., et al. "Supersonic Retropropulsion Experimental Results from NASA Ames 9x7 Foot Supersonic Wind Tunnel." Journal of Spacecraft and Rockets, 2014. doi: [10.2514/1.a32650](https://doi.org/10.2514/1.a32650)
15. (See ref. 13; software-failure taxonomy.) Wander, A. "Fatal Software Failures in Spaceflight." Encyclopedia, 2024. doi: [10.3390/encyclopedia4020061](https://doi.org/10.3390/encyclopedia4020061)
16. "Mars InSight Entry, Descent, and Landing Trajectory and Atmosphere Reconstruction." NASA Technical Report, 2020. NTRS: [https://ntrs.nasa.gov/citations/20200002910](https://ntrs.nasa.gov/citations/20200002910)
17. "Mars 2020 Entry, Descent and Landing Instrumentation 2 (MEDLI2)." NASA Technical Report, 2016. NTRS: [https://ntrs.nasa.gov/citations/20160008391](https://ntrs.nasa.gov/citations/20160008391)
18. Grile, R., and Bettinger, R. "Reliability analysis of deep space satellites launched 1991-2020: Bulk population and deployable satellite performance analysis." Quality and Reliability Engineering International, 2024. doi: [10.1002/qre.3600](https://doi.org/10.1002/qre.3600)
19. "Predicting Failures and Estimating Duration of Service Life from Satellite Telemetry." 2025. doi: [10.26077/md4y-zx37](https://doi.org/10.26077/md4y-zx37)
20. Yu, Z., et al. "The Tianwen-1 Guidance, Navigation, and Control for Mars Entry, Descent, and Landing." Space: Science & Technology, 2021. doi: [10.34133/2021/9846185](https://doi.org/10.34133/2021/9846185)
21. "Overview of the Small Lunar Lander SLIM and Its Lunar Landing Results." IEEJ Transactions on Industry Applications, 2026. doi: [10.1541/ieejjia.20250916](https://doi.org/10.1541/ieejjia.20250916)
22. Mousis, O., et al. "Scientific rationale for Saturn's in situ exploration." Planetary and Space Science, 2014. doi: [10.1016/j.pss.2014.09.014](https://doi.org/10.1016/j.pss.2014.09.014)
23. Fogel, R. W. "Railroads and American Economic Growth: Essays in Econometric History." Johns Hopkins Press, 1964. (Reviewed/abstracted) doi: [10.2307/2552284](https://doi.org/10.2307/2552284)
24. Perez, C. "Technological revolutions and techno-economic paradigms." Cambridge Journal of Economics, 2010. doi: [10.1093/cje/bep051](https://doi.org/10.1093/cje/bep051) (applied alongside Mokyr, "The Gifts of Athena," Princeton, 2002, and "A Culture of Growth," Princeton, 2017)
25. "Master Catalogue of Lunar and Mars Exploration Missions." OSF, 2021. doi: [10.31219/osf.io/9s6jn](https://doi.org/10.31219/osf.io/9s6jn)
26. "2020 NASA Technology Taxonomy." NASA, 2019. NTRS: [https://ntrs.nasa.gov/citations/20190029323](https://ntrs.nasa.gov/citations/20190029323)