Occam’s Revenge

Occam’s Revenge is a project to provide an improved logical problem-solving method to evaluate competing hypotheses for paranormal claims. 

Named for William of Occam’s heuristic method, Occam’s Razor this method follows similar principals inside a developed framework. This framework reduces subjectivity, helps foster adherence to logic and provides transparency.


Step #0 Choose an Appropriate Case

Since this method is an involved process and relies on a large body of evidence for accuracy, investigators should exercise care when choosing cases for evaluation. Cases should have the opportunity to collect physical evidence, supporting media (photos, video, audio) and eyewitnesses available for interview. 


Step #1 Evidence Collection

Investigators will collect evidence based on best practices for the organization. Recommended practices will be developed at a later date. 


Step #2 Evidence Evaluation

Evidence that is determined to be a product of a hoax, misidentification or an untrustworthy source, with a high degree of confidence will be thrown out. The remaining evidence will be analyzed and rated based on type of evidence. 

Eyewitness testimony will be included but will be adjusted. For example an observation of size in low light conditions will be changed to an appropriate size range. These will form adjustment rules and will be developed based on appropriate scholarly research and independent testing. 

BAC (Base Accuracy Coefficient) – According to historical testing, how accurate is certain evidence likely to be? For example eyewitness testimony, even after adjustment, will be less accurate than a photo or a video. 

TC (Truthfulness Coefficient) – Is applied to downgrade the rating of evidence that exhibits characteristics of being untruthful. Is also used when error or bias may have influenced the evidence. 

e = Evidence identification code. 

The BAC and TC will be applied to each piece of evidence. Each piece of evidence will also receive an identification code, expressed as e


Step #3 Evidence Analysis

Each piece of evidence will be further analyzed and concluding/analysis statements will be generated. These statements will describe findings and inferences based on the evidence. 

ASLC (Analysis Statement Likelihood Coefficient) – Used to downgrade/modify statements that are not 100% conclusive or in which there is some disagreement over the interpretation. 


Step #4 Hypothesis Development

Investigators should seek to create as many possible explanations that fit the evidence. Each hypothesis should be specific and seek to account accurately for every piece of evidence. This hypothesis is NOT the same as a scientific hypothesis as it may include multiple assumptions. 


Step #5 Breakout Hypothesis Components

Each hypothesis will be broken down into individual statements or assumptions. The breakdown should continue so every individual assumption is a separate statement. 

c = Identification for an individual hypothesis component


Step #6 Apply Evidence to Hypotheses

Apply the evidence to each statement of each of the competing hypotheses. Rate each piece of evidence on how well it supports or refutes each statement. Evidence can also be neutral where it has no relevance to a specific statement. 


Step #7 Measure Likely-hood and Confidence Level

Likely-hood for each hypothesis will be calculated based on how consistent the evidence is with the hypothesis. Confidence level will be calculated based on the quality, amount and distribution of the evidence. 

SRM (Support Refute Multiplier) – A multiplier used to express how well a piece of evidence supports or refutes a specific hypothesis component.


TEC (Total Evidence Contribution) – The point value contribution of a specific piece of evidence (e) to a specific hypothesis component (c)

TECc:e = SRMe * ASLCe * TCe * BACe


OLHC (Overall Likelihood of Hypothesis Component) 

Sum of TECc values for all relevant e values. 


OLH (Overall Likelihood of Hypothesis)

 Average of all TEC values for all c values. 


 Confidence Level Determined by Standard Deviation. 


Step #8 Identify Gaps and Opportunities

The hypotheses and evidence will be analyzed to look for gaps and opportunities. At this point critical faults will also be identified. A critical fault is when a component of a hypothesis is unsupported or unlikely or when a hypothesis is unable to account for a piece of evidence with a moderate support rating. 


Step #9 Refine & Repeat

The gaps and opportunities identified in Step #8 will be used to refine the model. Once refinements are made the process will repeat. 





Leave a Reply