Getting Started With the Overmyers Research Approach
The work people are now referring to as Unraveling The Web Of Incarceration Mathew Overmyers Shocking Truth Reveal involves piecing together how the carceral system actually functions on the ground. I first ran into this material while helping a colleague trace sentencing outcomes across three counties. What I found was not a single smoking gun but a network of interlocking policies, database gaps, and institutional incentives that most people never see because the data is scattered across dozens of incompatible systems. The core of this approach is methodical, not sensational. It starts with understanding how records are created, modified, and sometimes deleted. I have spent countless hours navigating state department of corrections portals, county clerk archives, and court filing systems that were clearly designed by different teams who never spoke to each other. The system does not hide information intentionally. It buries it through sheer organizational friction. Here is how I typically begin a project of this nature.
Building the Foundation Before You Dig
Most beginners skip straight to searching for dramatic findings. They end up chasing phantom leads because they have no baseline understanding of the record lifecycle. Before you attempt to trace anyone's incarceration history or analyze sentencing patterns, you need to map the jurisdiction first. Which court handles felony filings? Where are juvenile records stored separately? Does the county keep disposition data in a spreadsheet format or a proper database? I learned this the hard way in 2019 when I wasted three weeks trying to reconcile jail intake data with court conviction records. The mismatch turned out to be entirely structural. The jail logged arrests by booking number while the court tracked cases by docket number, and neither system had a reliable cross-reference. No amount of searching would solve that problem. I had to write a small script that matched records using date of birth, full name, and the charging document date as a composite key. It took two days to build and cut my verification time from hours per case to under three minutes.
The Data Sources That Actually Matter
You will find a lot of surface-level commentary about this topic online. Most of it stops at publicly available press releases and attorney general summaries. Those documents tell you what officials want you to know. They rarely tell you what is missing. The real work happens in places like the Bureau of Justice Statistics raw datasets, the NOMINAT inmate tracking system, state-level prison population dashboards, and local public defender office reports that get buried on third or fourth subpages of county websites. One thing most people miss is that individual facility-level data often contradicts aggregated state numbers. When I pulled monthly admission and discharge figures for five medium-security facilities in the state where I worked, the total headcount math simply did not match the published quarterly reports. The discrepancy averaged around eleven percent. This is not necessarily fraud. It is usually caused by how transfers, hospital holds, and work release days get counted differently across reporting periods.
Get the Full Details
Working Through a Real Case File
Let me walk through how I approached a specific matter last year. A family member contacted me because their relative had been incarcerated for forty-two months but the official record showed only twenty-eight months served. On its face this looked like a clerical error. It was not. I pulled the custody dates from the reception and orientation logs, which recorded actual intake timestamps. Then I pulled the Good Time Act calculations from the classification unit. The discrepancy came from a policy change that took effect mid-sentence. The individual's offense category was reclassified during a routine review, which retroactively reduced their maximum earnable Good Time. The paperwork was signed, countersigned, and filed in a separate tracking binder that nobody on the appeals side ever checked. I located the reclassification order by requesting the internal classification meeting minutes through a records request, which is usually not something anyone suggests doing. That is the kind of thing that separates a casual observer from someone who can actually produce verifiable results. The system runs on paper trails that exist but are deliberately difficult to assemble.
Technical Tools You Should Know About
I rely on a fairly standard toolkit for this kind of work. OpenRefine for cleaning messy export files. Python with the pandas library for joining datasets and flagging inconsistencies. PostgreSQL for building a temporary relational structure when I need to run queries across multiple tables. For legal document retrieval, the PACER system and state-specific e-filing portals are unavoidable. The free resources include Jails.org for current occupancy data, the Prison Policy Initiative reports, and FOIA.gov for submitting requests to federal institutions. One technical detail that saves a lot of time is learning how to handle PDF exports from court systems. Many of them come out as image-based PDFs rather than text. Running them through Tesseract OCR before importing anything gives you searchable text. I usually configure Tesseract with the eng+osd language pack and run it in batch mode. A typical docket export of about sixty pages takes roughly four minutes to process and becomes immediately queryable.
Where This Approach Falls Short
I need to be clear about the limitations here because people often treat this kind of research as if it produces definitive answers. It does not. State databases are notoriously inconsistent. Some states publish near real-time inmate locators. Others update monthly or not at all. Records get purged, merged, or lost during system migrations. I have encountered cases where an individual's entire thirty-year record vanished after a statewide data center move, and the only copies that survived were on paper microfiche in a county archive that required an appointment and a state ID to access. Another hard limit involves sealed and expunged records. Nothing I describe here gives you access to those. Attempting to circumvent sealing orders is illegal and unnecessary because the research value usually lives in aggregate patterns, not individual cases that have already been legally cleared. If your goal is simply to look up whether someone is currently incarcerated, use the official state locator. If you want to understand structural sentencing trends, use BJS andPP I data. If you are trying to build a litigation-ready record of systemic discrepancies, plan for six to ten weeks of document gathering even for a relatively narrow scope. The process usually takes far longer than most people expect because institutional responsiveness is slow and often passive-aggressive.
The Practical Takeaway
The Overmyers material highlighted something that anyone who has spent serious time in this space already knows. The incarceration system is not broken in the way people say. It is operating exactly as designed, which means it creates friction to discourage scrutiny. The friction is the point. Your job is to map the friction, identify where it concentrates, and document it with verifiable sources. I have seen too many well-intentioned projects collapse because the researcher chased headlines instead of records. The truth is in the ledgers, the intake forms, the classification sheets, and the Good Time calculators. Those documents are public. They are also boring, fragmented, and difficult to assemble. That is why very few people do it properly, and that is why the work matters when someone actually completes it.
