The Inert Device That Burns Every Hand It Touches

On the evening of January 5, 2021, someone in a gray hooded sweatshirt, dark gloves, jeans, and a particular model of Nike Air Max Speed Turf sneakers walked through several blocks of southeast Washington, D.C. carrying a backpack. Between approximately 7:34 p.m. and 8:18 p.m., the person placed two pipe bombs — one under a park bench behind the Democratic National Committee headquarters, one in an alley behind the Capitol Hill Club near the Republican National Committee. The devices were discovered the next afternoon. Neither detonated.

The investigation ran for fifty-eight months. The Federal Bureau of Investigation collected approximately three million data points, conducted more than a thousand interviews, subpoenaed records from hardware stores and shoe retailers, and offered a $500,000 reward. No suspect was arrested for nearly five years.

On November 5, 2025, Steve Baker, an independent journalist recently hired by Blaze Media, began publicly teasing his investigation's findings during an appearance on Glenn Beck's podcast. The next morning, November 6, the FBI administered a polygraph examination to Shauni Kerkhoff, a former U.S. Capitol Police officer now employed as campus security at the Central Intelligence Agency. She was asked two questions: "Did you place those pipe bombs?" and "Did you place those pipe bombs that evening?" According to the bureau's own polygraph examiner, she failed the examination. The examiner documented her "very controlled reaction to the news of her failing the polygraph and seemingly rehearsed responses."

On November 7, the FBI formally opened an investigative file on Kerkhoff and began issuing subpoenas for her electronic accounts. The same day, Ed Martin — a senior DOJ official who had served as a pardon attorney and as director of the Justice Department's Weaponization Working Group — publicly denounced online rumors identifying Kerkhoff in a post on X.

Baker and Hanneman published the following morning, November 8. Their article, "Former Capitol Police officer a forensic match for Jan. 6 pipe bomber, sources say," identified Kerkhoff by name, published her photograph, and claimed a forensic gait analysis matched her walking pattern to the pipe bomb suspect at a confidence level of 94 to 98 percent.

On November 12, the same day the FBI opened its formal investigation of the eventual defendant in this case, the bureau attempted to question Kerkhoff and her boyfriend at their residence. Per Kerkhoff's own complaint, the attempt involved a caravan of FBI vehicles including a bomb-disposal truck, an FBI helicopter overhead, and agents in full tactical gear.

On December 4, 2025, the FBI arrested Brian Cole Jr., 30, of Woodbridge, Virginia, at his family home. He was charged with transporting an explosive device and attempted malicious destruction by means of explosive materials. A superseding indictment later added charges of using a weapon of mass destruction and committing an act of terrorism while armed. He has pleaded not guilty.

On December 5, 2025, Blaze News retracted its November 8 article.

On April 1, 2026, Cole's defense attorneys filed a motion in U.S. District Court requesting subpoenas for Kerkhoff, her boyfriend Daniel Dickert — also a U.S. Capitol Police officer — and Dr. Michael Nirenberg, a board-certified forensic podiatrist. The motion stated the defense's intent to preserve the right to argue at trial that "Ms. Kerkhoff — not Mr. Cole — placed the pipe bombs." The motion was publicly docketed, then taken down and refiled under seal. The Department of Justice is seeking sanctions against Cole's attorneys for the initial filing.

On April 21, 2026, Kerkhoff filed a 124-page defamation suit in the Eastern District of Virginia against Blaze Media, Steve Baker, Joseph Hanneman, and Veritas Regnat LLC — a new journalistic outlet Baker and Hanneman founded after being removed from Blaze.

She is represented by Clare Locke, the Alexandria boutique that is the most effective plaintiff-side defamation firm operating in the United States. The firm secured the $787.5 million Dominion Voting Systems settlement from Fox News in 2023 — believed to be the largest defamation settlement involving a media company in U.S. history — and continues to represent Dominion in active cases against Newsmax, One America News Network, Sidney Powell, Rudy Giuliani, MyPillow, and Patrick Byrne. Its prior client list includes University of Virginia Dean Nicole Eramo against Rolling Stone ($3 million jury verdict); Gov. Sarah Palin against the New York Times (appellate reversal of dismissal); hedge fund manager Daniel Michalow against DE Shaw ($52.125 million FINRA-record award); British Muslim reformer Maajid Nawaz against the Southern Poverty Law Center ($3.375 million settlement); Dr. Fred Eshelman against Puma Biotechnology ($26 million verdict); former Penn State president Graham Spanier in a suit against investigator Louis Freeh; Matt Lauer in defense against sexual assault allegations; and Chiquita Brands against the Cincinnati Enquirer.

The list bears inspection. What stands out is not partisan alignment — the clients span the political spectrum — but a consistent posture: plaintiffs who were public institutions, public officials, or large corporate interests pursuing outlets and accusers that challenged them. Dominion against Fox and Newsmax. A university dean against Rolling Stone. A sitting governor against the New York Times. A Penn State president against a critical investigator. An institution-defending trade group against a metropolitan newspaper. Matt Lauer against accusers. It is not accurate to call Clare Locke a political firm. It is accurate to say that its practice has consistently represented institutional, corporate, and elite plaintiffs pursuing outlets and individuals who accused them — across the political spectrum, with case-selection criteria that appear to prioritize plaintiff-strong legal posture over ideological alignment. That is a pattern worth naming before interpreting what their appearance in this case tells us. For now: they took Kerkhoff's case.

The criminal case against Cole is ongoing. Cole is detained pending trial. The pipe bomb investigation is officially closed on Kerkhoff and officially solved with respect to Cole.

None of those sentences can be simplified without losing accuracy.

The mainstream account of this story tells it as two events that happened sequentially: first, the Blaze published a reckless article naming an innocent woman — a framing we will examine for historical validity later in the piece. Then, the FBI arrested the real bomber and the Blaze's mistake was exposed. One followed the other. Cause and effect, straightforward.

The record shows two investigative tracks running inside the same building in the same month, in the same bureau, under the same leadership. One of them produced a polygraph administered two days before the Blaze article was published, a formal case file opened one day before it published, month-long surveillance, subpoenas for electronic accounts, interviews with the subject's boyfriend and dog walker, a joint interview of the subject with an Assistant U.S. Attorney present, surveillance footage sent for gait analysis, and a request to the subject's current employer for her records. The other produced the arrest the bureau chose to announce.

Both tracks targeted different people. Both tracks ran simultaneously. Both tracks were part of the same pipe bomb investigation.

The first track pre-dated the Blaze article itself, though it did not pre-date the allegations underlying it. Baker had teased his findings on Glenn Beck's podcast on November 5, the day before the polygraph. Per Kerkhoff's own complaint, Baker had shared his allegations with staff for Director of National Intelligence Tulsi Gabbard prior to the FBI's action. What the bureau responded to, then, was not a published news story but a set of allegations that had already begun to circulate through institutional and media channels. What is notable is not that the bureau acted in an informational vacuum — it did not — but that the posture it adopted toward Kerkhoff was formally investigative, not reactively dismissive. The polygraph was on November 6. The file opened on November 7. Ed Martin publicly denounced the identification on November 7. The Blaze article published on November 8. The tactical-caravan residential visit happened November 12.

Then, on November 13, 2025 — the day after the FBI's joint interview of Kerkhoff and Dickert with the U.S. Attorney's Office — FBI Deputy Director Dan Bongino posted on X that reporting about "prior persons of interest is grossly inaccurate and serves only to mislead the public." That same week, according to Rep. Thomas Massie (R-KY), Bongino called Massie privately and said the bureau was "currently working to clear" Kerkhoff, and that he would personally finance defamation lawsuits against reporters covering the story. The FBI's formal investigation of Kerkhoff remained open for another seven weeks. It closed on or around January 6, 2026 — thirty-three days after Cole was arrested and charged.

A responsible investigation of a private citizen merits thoroughness, and two months is not in itself excessive. What is notable is not the duration but the posture. The joint interview with an AUSA present — a former FBI agent told Blaze News this was "definitely a bit unusual" — is not the posture of a bureau responding to a false media lead. It is the posture of a bureau pursuing a working theory with prosecutorial intent. That posture continued for weeks after the alleged real bomber was in custody. And the public characterization of the entire track as "grossly inaccurate" reporting about a non-existent person of interest did not reflect what was happening inside the bureau. That characterization managed the news cycle. The investigation continued anyway.

The bureau's public position was that Kerkhoff had been falsely accused by a reckless media outlet and was never a genuine suspect. The bureau's operational posture, in the same weeks, was that Kerkhoff was a person of interest serious enough to merit a polygraph, surveillance, subpoenas, and a joint interview with prosecutors. These two positions are not consistent with each other. One of them — at minimum — misrepresents what the bureau was actually doing.

We are not going to pretend we know whether Kerkhoff planted the bombs. There are facts in the record that make the question a live one, and they deserve to be named rather than waved at.

The bureau's own polygraph examiner documented Kerkhoff's failed November 6 examination and her "very controlled reaction to the news of her failing" and her "seemingly rehearsed responses." Dr. Michael Nirenberg — a board-certified forensic podiatrist, co-author of the standard English-language textbook on the methodology (Forensic Gait Analysis: Principles and Practice, Taylor & Francis), and a prior consultant to the FBI on gait-analysis cases — produced, in combination with a separate software analysis layer, a gait match of 94 percent at the software stage and 96 to 98 percent at his own visual assessment. Two additional analysts concurred. The comparison used two reference samples: Capitol Police CCTV footage of Kerkhoff on duty, and 2017 footage of her playing professional soccer. A 5'7" height matched the FBI's own suspect description. Capitol Police training supplied the geographic familiarity the placement pattern implies — including the bomber's escape through a hidden gate at St. Peter's Catholic Church that required site knowledge to locate. A 2015 right-tibia surgery is documented; its implications for her 2021 gait are contested. Kerkhoff's April 2026 complaint asserts she has fully recovered, citing four sub-four-hour marathons since 2016 and a USCP female physical fitness record. Baker and Hanneman, writing in Veritas Regnat on April 6, 2026, maintain they have hours of video — harvested from Capitol Police CCTV of her on-duty on January 6, 2021 and from her short professional soccer career — showing what a medical term of art calls a "circumduction gait," the distinctive leg-swinging-outward pattern that characterized the hoodie suspect's walk on the FBI-released placement footage. A CIA employment transition in mid-2021 sits behind an SF-86 clearance application whose timing Cole's defense is currently trying to subpoena.

The marathon detail is doing a peculiar double duty in the public record of this case, and it deserves its own moment. Kerkhoff's defense, as articulated in her complaint and her attorneys' January 2026 pre-litigation letter, rests substantially on the claim that she is a marathon runner — specifically, that her completion of four sub-four-hour marathons since 2016 is evidence she does not walk with any kind of limp. Marshall Yates, the former FBI congressional liaison to Massie, is on the record — via Baker's subsequent reporting — as having said, while reviewing video of the hoodie suspect in Massie's congressional office, "We need to be looking for a marathon runner." He said it because marathon runners exhibit a set of distinct physical traits — a particular cadence, a particular stride mechanic, a particular way of carrying the hips — that were visible in the hoodie suspect's forty-four-minute walk. That is, the internal FBI-adjacent profile of the bomber, developed from the placement footage before any public identification of Kerkhoff existed, was: "marathon runner." Kerkhoff's defense of herself against the identification is: "I am a marathon runner." The profile the investigative source arrived at from the video and the alibi the subject is offering to rebut the identification from the video are, if both are true, the same profile. This does not prove Kerkhoff planted the bombs. It does establish that the single most public plank of her defense is the exact characteristic the bomber's movements were read as exhibiting.

A brief word about what gait analysis is, because the Blaze's reporting has been widely characterized as if it rested on something closer to internet hobbyism than on a real forensic discipline. It is neither. Gait analysis sits somewhere between fingerprint analysis and facial recognition on the spectrum of forensic reliability. It is more established than popular coverage recognizes, less established than DNA or fingerprinting, admissible as evidence in U.S. courts, and in active use by credentialed practitioners including — on prior FBI-assisted cases — Nirenberg himself. What it does not yet have is a peer-reviewed literature supporting confidence figures at the precision the Blaze's "94 to 98 percent" framing implied. The methodology is real. The specific confidence number outran what the underlying science reliably supports — a point we will return to when we score Baker's reporting. A body of evidence that does not meet the standard of proof is not the same as a body of evidence that amounts to nothing. What the bureau formally investigated for two months sat between those two poles.

We are also not going to pretend those facts constitute proof. They constitute a set of questions that the bureau's formal investigation took seriously enough to pursue for two months, that the bureau's public statements dismissed as non-existent, that the defense is being actively prevented from raising at trial, and that no mainstream outlet has catalogued in one place.

Broadside is not going to solve this case. We are not equipped to, and more importantly, we are not positioned to. The people positioned to solve it — the FBI's own investigators, congressional oversight, the Cole defense team with full discovery access — have variously chosen not to, been actively prevented from doing so, or been publicly characterized as cranks for trying. What interests us is different. It is the fragility of the official account, the asymmetry of institutional response to anyone who examines that account, and what that asymmetry could possibly point to — even if, in the end, it points only to the fact of the asymmetry itself.

This is the part of the story that does not fit any partisan frame.

Rep. Thomas Massie, a Republican congressman with speech-and-debate immunity, has publicly alleged that Ashan Benedict, an FBI official, threatened a criminal investigation of Massie's congressional staff if Massie did not "play ball" on the pipe bomb investigation. Massie has described a meeting in his own congressional office in which Benedict repeatedly interrupted, answered for, or "qualified" the answers of a U.S. Capitol Police officer Massie was attempting to interview. This officer was the member of the Capitol Police counter-surveillance team who physically discovered the pipe bomb behind the DNC on the afternoon of January 6, 2021. Massie has sought a transcribed interview with him. The bureau has not facilitated one. The officer has still not given a transcribed interview to any congressional body about what he saw that day.

Kyle Seraphin, a former FBI Special Agent, has stated on the record that his team was assigned to surveil a Falls Church, Virginia residence within days of January 6, 2021, as part of the pipe bomb investigation. The residence belonged to an Air Force civilian employee tied to a DC Metro SmarTrip card allegedly used by the bomber. Seraphin's team proposed a knock-and-talk. The request was denied. The team was pulled off the case that same night. Seraphin gave this account publicly beginning in 2021. He was suspended from the FBI in August 2022. He is now a podcaster.

The Falls Church subject — an Air Force civilian employee tied to DC Metro SmarTrip use — is, on the available record, unlikely to be Cole. Cole's residence is Woodbridge, not Falls Church. He has no known Air Force connection. The license plate reader hit placing his 2017 Nissan Sentra near Capitol Hill at 7:10 p.m. on January 5, 2021 is consistent with a drive-in, not with Metro use. What Seraphin's team was surveilling within days of the bombing — four years before any public identification of any suspect — was a different person of interest entirely. That thread was terminated the same night the knock-and-talk was proposed. Whatever the bureau's reasons, they have not been publicly explained in the four years since. The current "languished for four years" framing does not engage with this thread at all.

An anonymous FBI supervisory whistleblower raised concerns in late 2025 that the Falls Church surveillance was shut down after two days at the doorstep of a person subsequently identified through gait analysis as possibly matching the bomber. The whistleblower's attorney, Kurt Siuzdak, sent Massie and Loudermilk a formal letter on November 13, 2025 — the same day as Bongino's public X post — warning that any attempt by the bureau to identify the whistleblower would violate federal whistleblower protections. Siuzdak's letter stated that identifying the whistleblower "serves only one purpose, which is to allow FBI management to retaliate."

Steve D'Antuono, the former Assistant Director in Charge of the FBI Washington Field Office, testified to Congress in June 2023 that the sixty-minute kitchen timer used on the devices could not have detonated them seventeen hours after placement. He testified: "Maybe they weren't supposed to go off. We can't — we don't know." D'Antuono is no longer at the bureau. His testimony was cited in a September 2024 letter from House Republicans to then-Director Christopher Wray. The letter went unanswered. The current FBI leadership does not publicly reference D'Antuono's testimony when describing the devices. The pending superseding indictment against Cole charges him with using a weapon of mass destruction.

Rep. Barry Loudermilk (R-GA), chair of the House Select Subcommittee to Investigate the Remaining Questions Surrounding January 6, has submitted formal questions to the FBI about when Cole's phone number was included in the bureau's "numbers of interest" list and why investigation of it ceased at some point. The FBI has not publicly answered. During the Biden administration, Loudermilk was himself the subject of an FBI investigation related to constituent tours he gave before January 6. That investigation closed without charges.

An independent video analyst who publishes under the name Armitas submitted a twenty-six-page report on the pipe bomb footage to Loudermilk's subcommittee. Armitas had been feeding findings to an FBI Special Agent at the Washington Field Office since March 2025. His analysis identified specific evidence the FBI had not publicly addressed: the bomber's apparent first attempt to plant a device at the Congressional Black Caucus Institute, the interrupted placement at a congressional dormitory on C Street, the escape through a hidden gate at St. Peter's Catholic Church that required site familiarity to locate. The report was submitted to House Judiciary counsel on October 4, 2025 and reported publicly by Joseph Hanneman and Steve Baker in the Blaze on October 9, 2025 — a full month before the November eruption this piece has been documenting. When Deputy Director Bongino dismissed "prior persons of interest" reporting as "grossly inaccurate" on November 13, he was dismissing analysis that had been public, and in the hands of congressional counsel, for weeks.

Armitas's more recent work is more consequential. Scrubbing the feeds of dozens of Capitol Police CCTV cameras, he identified three blue Nissan Sentras on Capitol Hill between roughly 7 and 8:10 p.m. on January 5, 2021. He disqualified two — one driven by a white individual with a white passenger (Cole is black), one with fog lamps engaged that are not present on Cole's vehicle per National Motor Vehicle Title Information System records. The third was a visual match for Cole's 2017 Nissan Sentra. At 8:03 p.m., security cameras captured that Sentra traveling north on 3rd Street SW more than a mile from the bomb sites. At 8:04 p.m., it turned east on C Street SW and entered the I-395 on-ramp, traveling west — away from Capitol Hill. At that exact window, the government alleges Cole was on foot near Folger Park between the DNC placement and the RNC placement. It is a six-to-seven minute drive from 3rd Street SW to Folger Park. Cole cannot be on foot near Folger Park and driving west on I-395 more than a mile away at the same time. The Nissan finding is not corroboration for the government's case. It is, if the visual identification holds, exculpatory.

Armitas's analysis of the DNC placement video identified a second problem. The first pipe bomb's sixty-minute kitchen timer required two full turns of the dial to arm, with the alligator clips detached and reattached. The video — released by the FBI itself — shows the device pulled from the backpack and placed under the bench in a single motion, with no dial manipulation visible. The Department of Justice's detention memo states Cole "set the timer on the first device to the maximum duration (60 minutes) and planted the device." The video the FBI released does not show that. The timer was never set.

Per Armitas's analysis of the bodycam footage, Cole's walk, as captured on Prince William County Police bodycam footage from a 2024 traffic accident, is visibly distinct from the hoodie suspect's walk — lumbering, feet outward, no cadence. The hoodie suspect, per the FBI's own video and Dr. Nirenberg's forensic analysis, walked with a consistent gait featuring a particular knee flexion pattern and step rhythm. Nirenberg's characterization of that hoodie-suspect gait is, in effect, the forensic baseline against which both potential suspects have been compared. The same methodology that produced a 94 to 98 percent match to Kerkhoff produces, when applied to Cole, no match at all. Neither comparison has been formally adjudicated. Both rest on the same underlying method and the same released video.

Armitas has asked not to be publicly named, citing security concerns. The mainstream characterization of his work, in outlets including The Bulwark, has centered on his anonymous-online status and the fact that his social media avatar is Jaheira, a druid character from the 1998 computer role-playing game Baldur's Gate. None of the mainstream coverage has addressed his Nissan finding, his timer finding, his walk-comparison finding, or the Department of Justice's documentary contradictions with the video evidence the bureau itself released.

Steve Baker is an independent journalist who covered the January 6 Capitol events without press credentials — as did approximately sixty other journalists and photographers in the crowd that day. He filmed the events, licensed his footage to mainstream outlets, and subsequently turned his video over to the FBI in response to a grand jury subpoena. In March 2024 — three years after the event — he was arrested in Dallas, handcuffed, placed in leg irons, and charged with four misdemeanors. The FBI affidavit specifically cited Baker's "use of loud, threatening, or abusive language" — recorded questions to officers including "You gonna use that thing on us?" — as the basis for the speech-based misdemeanor charges under 40 USC § 5104(e)(2), which criminalizes disruptive speech inside a Capitol Building. The core act being prosecuted was recorded speech. Elijah Schaffer, then a Blaze commentator, entered former Speaker Pelosi's office on January 6 while wearing an official Capitol press credential and was never charged. Baker was. Baker pleaded guilty in November 2024, stating he did so "to avoid the shaming exercise of a trial." Rep. Jim Jordan, as chair of the House Judiciary Committee, opened an investigation into Baker's treatment, writing that it "smacks of harassment and selective treatment for a disfavored criminal defendant." Before sentencing, Baker was pardoned by President Trump on January 20, 2025. He joined Blaze Media and reported on the pipe bomb investigation, including the November 8, 2025 article identifying Kerkhoff. He was terminated by Blaze after the retraction. He is a named defendant in the defamation suit filed this week.

Joseph Hanneman, Baker's co-author, previously covered January 6 exclusively for The Epoch Times. He joined Blaze. He resigned after Baker's firing. He is a named defendant in the defamation suit.

Julie Kelly, an independent journalist who has pursued her own investigation of the pipe bomb case, publicly criticized the Blaze's gait analysis as "weak evidence" — meaning she is not on the Blaze's side of the specific dispute, but remains on the "unresolved questions" side of the broader investigation. Her work is routinely characterized by mainstream outlets as conspiratorial.

Brian Cole's defense attorneys filed a motion requesting discovery materials that would allow them to present an alternative-suspect theory. The motion was publicly docketed, then yanked and refiled under seal. The Department of Justice is seeking sanctions against them for the public filing. A status hearing is pending.

That is the list, to date, of people who have asked specific factual questions about the pipe bomb investigation. Now the other list.

Dan Bongino publicly called the Blaze reporting "grossly inaccurate" on November 13, 2025, while the bureau he was deputy director of was actively investigating the subject of that reporting. Bongino had previously, on his own podcast in 2024, described the pipe bomb as "an inside job" about which he had "zero doubt." No institutional consequence has attached to either statement.

Pam Bondi, as U.S. Attorney General, announced the Cole arrest on December 4, 2025, and stated the case had "languished for four years until Director Patel and Deputy Director Bongino came to the FBI." The statement credited the new administration with breaking the case and implicitly criticized the previous administration for the delay. No institutional consequence has attached.

John Nantz, a retired FBI Supervisory Special Agent, testified to Loudermilk's subcommittee on January 15, 2026, in support of the Patel/Bongino framing. He attacked independent analysts of the case as "internet cranks utilizing hobbyist analysis" and the Kerkhoff-adjacent theory as a "grifting class" conspiracy. He has a Substack and a column at Townhall. No institutional consequence.

Mainstream outlets including CBS News, NBC News, the Washington Post, the New York Times, The Bulwark, and Status have characterized Kerkhoff as "cleared" and the questions as "debunked" without noting the bureau's own contemporaneous investigation, the failed polygraph, or the pending sealed motion. Several of these outlets had the underlying facts available to them. No institutional consequence. No retractions.

The pattern is not ambiguous. On one side of the line, asking specific factual questions about this investigation has produced: threatened prosecution of congressional staff, termination of employment, defamation litigation backed by elite counsel, FBI suspension, the need for legal anonymity, public characterization as a crank, discovery sanctions, and — in Baker's case — a federal prosecution premised on the defendant's recorded speech. On the other side, defending the official framing has produced promotions, federal appointments, continued career trajectories, and invited congressional testimony.

A rational observer watching this pattern — a mid-career journalist, a congressional staffer considering whether to press a lead, an FBI agent weighing a whistleblower disclosure — would draw the obvious inference.

That is the structural finding of this investigation. Not who placed the pipe bombs. Not whether Kerkhoff was involved. Not whether Cole is guilty as charged. The finding is this: in the pipe bomb investigation, as in a small number of other cases over the last four years, the cost of asking specific factual questions falls reliably on the questioners, and the cost of defending the official account is zero. The distribution of consequences is the evidence. It tells us what the ecosystem is calibrated to protect, which is not the same thing as what it claims to be protecting.

That finding does not require anyone to be coordinating. It requires no conspiracy and no meeting in a room. It requires only that each individual actor make the rational career choice in the face of the incentives visible to them. The pattern produces itself. Which is the most durable kind of pattern there is.

The piece so far has moved from chronology to structure to pattern. Now the claim narrows. We are not going to speculate. We are going to do the one thing Broadside can do ironclad with the available record: set the two cases next to each other and read what the comparison produces.

Broadside does not accuse private citizens of crimes. Broadside does not declare criminal defendants guilty or innocent. What Broadside does is insist that two people being investigated as the suspect in the same bombing receive the same standard of evidentiary treatment and the same standard of public accounting. In this case, they have not.

On the Kerkhoff side, the public record includes: a failed polygraph administered by the FBI's own examiner on November 6, 2025, with documented examiner observations of "rehearsed responses"; a forensic gait analysis conducted with a software layer (94%) and a credentialed forensic podiatrist layer (96-98%) using two reference samples — Capitol Police CCTV of her on duty and 2017 footage of her playing professional soccer; a 2015 right-tibia surgery documented by contemporaneous Temple News reporting and acknowledged by Kerkhoff, whose implications for her 2021 gait are contested — the Blaze's analysts found a persistent limp consistent with the hoodie suspect's walk, while Kerkhoff's complaint cites her subsequent running of four sub-four-hour marathons since 2016 and a USCP female physical fitness record as evidence she does not walk with a limp; a 5'7" height match to the FBI's suspect description; USCP training that supplies the geographic familiarity the bomber's movements suggest, including the use of a hidden gate at St. Peter's Catholic Church that required site knowledge to locate; employment as a USCP officer on the grounds the bomber navigated; and a CIA employment transition in mid-2021 whose SF-86 application timing remains under subpoena by Cole's defense.

On the Cole side, the public record includes: a confession obtained during a four-hour FBI interrogation following the December 4 arrest, reportedly given without counsel present, by a defendant who is autistic; cell phone tower pings placing his phone in the general area of the RNC and DNC during the relevant window; a license plate reader hit on his 2017 Nissan Sentra at 7:10 p.m. less than half a mile from the area where the hoodie suspect was first observed; credit card records showing purchases of items "consistent with" pipe bomb components between 2018 and 2020.

The items, per the FBI affidavit, were these: galvanized pipe and end caps (Home Depot, October 2019 and October 2020); a 9-volt battery connector (Micro Center); two white kitchen timers (Walmart, June 3, 2020); a pair of safety glasses; wire strippers; work gloves. None of those purchases, in isolation or in aggregate, is unusual. Galvanized pipe and end caps are plumbing hardware — what a homeowner buys to replace a section of pipe under a sink. A 9-volt battery connector is an electronics-hobbyist component sold at every consumer electronics store in the country, used in any project from a science-fair buzzer to a door chime. Kitchen timers are kitchen timers. Safety glasses, wire strippers, and work gloves are in every hardware aisle, bought every day by people who are swapping out an outlet, pruning roses, or patching drywall. Pipe bombs are fabricated from ordinary materials — that is what makes them pipe bombs — and so a purchase history "consistent with pipe bomb components" is, in most cases, also consistent with a slightly handy adult living in a house. The same framing would describe the purchase history of a substantial share of the American homeowner population over any two-year window. What the affidavit establishes is that Cole, like several million other Americans, bought things that can be used to make a pipe bomb. What it does not establish is that he bought things that cannot be explained by ordinary household use.

What the affidavit does not cite is, if anything, more telling than what it does. There is no black powder, no smokeless powder, no propellant of any kind in the traced purchase record. There are no blasting caps, no commercial detonators, no fuse material, no matchhead harvesting, no regulated or watchlisted precursor chemicals — no ammonium nitrate, no potassium chlorate, no aluminum powder, no anything that appears on ATF or DHS precursor watch lists. There are no specialty tools. No drill, no pipe cutter, no pipe threader, no vice. No soldering iron, no solder, no heat-shrink tubing, no transistors or relays. No anonymized purchases. All seven cited transactions are traceable credit-card purchases in Cole's own name at chain retailers. The energetic filler — the material that actually makes a pipe bomb a pipe bomb rather than a piece of threaded steel — has no traced purchase at all. The comparison cases are instructive here. The Boston Marathon bombers were caught partly through fireworks purchases, which is where their black powder came from. Ted Kaczynski was traced through specialty wire and unique typewriter ribbon orders. Timothy McVeigh's fertilizer and racing-fuel purchases were the smoking-gun evidence in his case because the quantities were inconsistent with any ordinary use — forty 50-pound bags of ammonium nitrate is not something a hobbyist buys for a flowerbed. Cole's purchase record shows no analog. It shows a homeowner's hardware-aisle receipts.

That difference matters, because in a prosecution built on a short list of circumstantial indicia plus a contested confession, the weight assigned to each indicium has to bear the weight the prosecution places on it. "Consistent with pipe bomb components" is doing a particular kind of rhetorical work when the affidavit has no inculpatory specificity to offer on the one material that would actually distinguish a bomb-assembly purchase pattern from a homeowner's. If that material existed in the record — a fuse order, a precursor chemical, a quantity of anything inconsistent with ordinary use — the affidavit would say so. It does not.

That is the evidence being treated as dispositive of Cole's guilt and treated as having "cleared" Kerkhoff. The comparison is not neutral.

The record also contains findings that cut against the government's current theory of the case — findings the government has not publicly addressed and mainstream coverage has not catalogued.

Armitas's analysis of the CCTV camera feeds identified what appears to be Cole's Nissan Sentra traveling away from Capitol Hill at 8:03-8:04 p.m. on January 5, more than a mile from the bomb sites, entering the I-395 on-ramp heading west — at precisely the window when the government alleges Cole was on foot between the DNC placement and the RNC placement. Cole cannot be in both places. If the visual identification of the Sentra is correct, the government's timeline does not hold.

Armitas's analysis of the DNC placement video — released by the FBI itself — showed the first pipe bomb being taken from a backpack and placed under the bush in a single motion, with no visible manipulation of the kitchen timer dial. The timer requires two full turns and the detachment-reattachment of alligator clips to be armed. The DOJ's detention memo states Cole "set the timer on the first device to the maximum duration (60 minutes) and planted the device." The video shows that did not happen. The timer was never set. The full report detailing this finding, along with the Sentra incompatibility and the device-orientation inconsistency between placement on January 5 and discovery on January 6, is publicly available in Hanneman and Baker's October 9, 2025 Blaze Media piece summarizing the Armitas submission to congressional counsel.

Cole's walk, as captured on Prince William County bodycam footage from a 2024 traffic accident, bears no visible resemblance to the hoodie suspect's walk. Per Armitas's analysis, Cole's gait is lumbering, feet pointed outward, with no discernible cadence. The hoodie suspect's walk, per the FBI's own released footage and Dr. Nirenberg's forensic analysis, shows a consistent rhythm and a specific knee flexion pattern.

Steve D'Antuono, the FBI's own former Washington Field Office Assistant Director in Charge, testified under oath to Congress in June 2023 that the sixty-minute timer could not have detonated the devices seventeen hours after placement — meaning the devices as configured could not have functioned as weapons during the January 6 Capitol riots. The current superseding indictment charges Cole with using a weapon of mass destruction. That charge and D'Antuono's sworn testimony are documentarily in conflict.

Every finding in the preceding four paragraphs is (a) based on publicly available evidence — CCTV footage, FBI-released video, bodycam footage, sworn congressional testimony — and (b) unaddressed in any mainstream coverage of the Cole prosecution as of this writing.

The alibi that is credited with clearing Kerkhoff is a video reportedly showing her playing with her dog during the evening of January 5, 2021. That video has not been publicly produced by any outlet. Its metadata has not been authenticated. Its existence is known to the public only through a November 25, 2025 CBS News report attributed to anonymous sources familiar with the investigation. That report appeared during the same seven-week window in which the FBI's formal investigation of Kerkhoff remained open. The same bureau that was still investigating her was, through anonymous sources, briefing CBS that she was alibied.

Compare this to the public standard the mainstream press applies to other claims in this case. Armitas's CCTV-sourced findings — the Sentra at 8:03, the timer that was never set, the walk mismatch — rest on footage the FBI has itself released or Capitol Police cameras have publicly captured. These findings are independently reviewable. They have been dismissed as "hobbyist analysis."

One private citizen has been cleared by a video nobody can see, sourced through anonymous leaks from a bureau still investigating her. One defendant's exculpatory evidence — the same kind of Capitol Police video the FBI itself has relied on for the charging documents — is being suppressed by the Department of Justice through a motion to seal and a sanctions motion against his own attorneys. The standard of evidentiary seriousness is not the same.

The framing in which Baker's November 8 article constituted a uniquely reckless act of journalism does not survive comparison to how major outlets have handled analogous cases in the recent historical record.

In 1996, the Atlanta Journal-Constitution named Richard Jewell as the Olympic Park bomber on the basis of an anonymous FBI leak and a criminal profile. No direct comment was sought from Jewell. The article asserted he fit the "lone bomber" profile. NBC, CNN, and the New York Post amplified the identification. Jewell was innocent. NBC, CNN, and NYPost settled his subsequent defamation suits for undisclosed amounts. AJC fought the case and won on the ground that their article was "substantially true" because the FBI was looking at Jewell — regardless of whether that was based on solid evidence. No AJC reporter was fired. The reporter, Kathy Scruggs, continued at AJC. A Clint Eastwood film was eventually made about what happened to Jewell.

In 2002, Nicholas Kristof of the New York Times published six columns naming Steven Hatfill — first as "Dr. Z," then dropping the pseudonym — as the likely anthrax mailer. Hatfill was innocent. The bureau searched his apartment with news helicopters overhead. The DOJ eventually paid Hatfill $5.85 million to settle his Privacy Act claim. Kristof won summary judgment in the subsequent defamation case on public-figure grounds. He kept his column, won a Pulitzer the following year, and remained at the Times for another two decades. No firings.

In 1999, the New York Times published a sustained run of reporting implicating Wen Ho Lee as a Chinese nuclear spy. Lee was held in solitary for nine months. He eventually pleaded guilty to one minor count. A federal judge apologized. Lee received $1.6 million in combined settlements from the federal government and five news organizations including the New York Times. The Times published a qualified "From the Editors" note. No firings.

In 2013, Reddit and BuzzFeed amplified the false identification of Sunil Tripathi — a missing Brown University student — as a Boston Marathon bombing suspect. Tripathi was dead by suicide before any of this. Reddit issued a blog apology. BuzzFeed continued operations. No firings.

Now Baker. He published an article naming Kerkhoff on November 8, 2025, sourced to a credentialed forensic podiatrist (Nirenberg, co-author of the field's standard textbook), a software gait-analysis layer that produced a 94% match, the podiatrist's visual assessment that raised the match to 96-98%, two additional analysts who concurred, and multiple intelligence-adjacent sources. He did not seek comment from Kerkhoff before publication. He published her photograph. The article was retracted on December 5. Baker was terminated. Hanneman resigned. Within six months, they are named in a 124-page defamation suit filed by the most effective plaintiff-side defamation firm in the United States, whose practice specializes in representing institutional plaintiffs against outlets that accuse them. The complaint cites Veritas Regnat — the new outlet they founded after their firings — as evidence of continued defamatory intent.

The Baker sourcing included, on the record, a credentialed forensic podiatrist with a published textbook on the methodology — a level of expert grounding absent from AJC's 1996 lone-profile leak, from Kristof's anonymous-source columns, from the New York Times' espionage reporting on Lee, and dramatically more than the 2013 Tripathi crowd-sourced identification. The failure — no pre-publication comment sought from the subject — was a failure the mainstream press has committed repeatedly in analogous high-profile cases. The consequences Baker has absorbed are categorically more severe.

The question is not whether Baker's article was responsible journalism by every standard. The question is whether the consequences being imposed on him and his colleagues are proportional to his journalistic failures, or proportional to something else. The historical record answers that question. It is not the journalistic failures being punished.

Several mainstream outlets — including CBS News, NBC News, the Washington Post, the New York Times, The Bulwark, and Status — published characterizations of Kerkhoff as "cleared" by the FBI and the Blaze's reporting as "debunked" during the same weeks in which the FBI's formal investigation of Kerkhoff remained open, her polygraph failure was documented in the bureau's own records, and her joint interview with an AUSA had just been conducted. Several of these outlets had access to the underlying facts. CBS's own sourcing — the anonymous leak describing the puppy video — came from inside the same bureau whose investigation of Kerkhoff was still active.

No correction has been issued. No retraction has been published. No coverage has been revised to note the parallel investigation, the sealed motion, the polygraph, or the pending defense subpoenas. The word "cleared" continues to appear in reporting on this case.

The outlets that got this wrong were not working from the record. They were working from the briefings they were given. When the briefings and the record diverged, they reported the briefings. This is not specific to this case. It is how institutional journalism functions in proximity to federal law enforcement. In this case, its failure mode was made visible because the contradiction between briefing and record was legible in the FBI's own documented investigative conduct.

What follows is an accounting by named actor of what each said, what each did, and where the public record and the operational record diverge. Kerkhoff and Cole are not scored here. Kerkhoff is a subject of investigation whose innocence is not Broadside's to declare. Cole is a criminal defendant whose guilt is for a jury to determine. The actors scored are the public officials, institutions, and journalists whose public statements and conduct are themselves part of the public record — the people whose positions were adopted professionally and can be evaluated professionally.

The language has been calibrated. "Accurate" is reserved for claims the record supports on inspection. "Misrepresented" marks a gap between what was publicly said and what was operationally done. "Factually inaccurate" marks a specific claim contradicted by documentary evidence. "Improper" means an action that fails a procedural, legal, or ethical standard regardless of motive. "Procedural failure" means a professional standard was not met. "Documented contradiction" means two things an actor has stated or done are in conflict on the record. "Suspicious" is reserved for cases where the action is on the record and is consistent with improper motive, but where intent cannot be established from what is publicly available. "Not asking" and "Not answering" are assigned to questions an actor has a professional duty or opportunity to address and has not. The point of this vocabulary is to avoid the one thing that would make this piece easier and less honest: a flat accusation of bad faith where the record supports only the narrower finding.

The scorecards begin with the two individual officials most publicly associated with the bureau's November 2025 posture, then move to the two institutions themselves, then to the journalist whose reporting triggered the public phase of the case, then to the coverage ecosystem that responded to it.

Dan Bongino, then-Deputy Director of the FBI, entered the case with a public record of having publicly diagnosed the investigation's failure. That diagnosis — on his own podcast in 2024, calling the case "an inside job" about which he had "zero doubt" — proved partially borne out by his own bureau's subsequent disclosure that the Cole arrest was possible using data the FBI had held since 2021-2022. What changed in November 2025 was not his diagnosis but his posture. He went from publicly criticizing an investigation he did not run to publicly defending a set of narratives about an investigation he now did run. The gap between those two postures is where the record diverges from the framing.

The Bongino scorecard is notable not for any single item but for the shape of its four rows read together. A public official with an accurate prior diagnosis of institutional dysfunction adopts, once in office, the exact institutional posture his prior diagnosis condemned. The Massie-attributed private offer, whether or not it is independently verified in due course, sits in the file uncontested. Whatever the operational justification for the parallel Kerkhoff track, there is no operational justification for publicly denying its existence while it was running.

Pam Bondi as U.S. Attorney General made the public case for the Cole arrest as a clean, evidence-supported closure of a long-cold investigation. The case did follow a re-examination of evidence in the bureau's possession. The closure framing that re-examination produced, however, did work the record does not support. It attributed the case's break to new leadership using language — "languished for four years" — that implied prior investigative failure while eliding the more uncomfortable reading: the evidence was available, the connection was not made, and whether that was negligence, prioritization, or something else is unresolved by the public record. That framing was not a lie. It was a choice about what to foreground. The subsequent prosecutorial posture, however, is a choice of a different character.

The Bondi scorecard's "Improper" row matters more than any of the rhetorical ones. The DOJ is entitled to prosecute. It is not entitled, in a prosecution of the scale and political weight of this one, to move to prevent the defense from raising an alternative-suspect theory and to move to seal the defense's own filing that raised it, without the kind of good cause a court ultimately will have to evaluate on the record. Both actions are within the DOJ's procedural reach. The question the record raises is not whether the DOJ can do this. It is why the DOJ is doing this in a case the department publicly describes as an open-and-shut product of rigorous investigative work.

The Federal Bureau of Investigation as institution is the actor the piece has spent the most time with, because the bureau is the one whose operational record and public framing have the largest and most legible gap. The bureau's charging-document evidence against Cole is real. The bureau's operational posture toward Kerkhoff was also real. The bureau's public denial of the latter was not accurate. And the bureau's own released video contradicts the bureau's own detention memo on a specific point the government is prosecuting Cole under a WMD charge in reliance upon.

The bureau's scorecard is the one where the calibrated language carries the most weight. "Misrepresented" is not "lied." "Documented contradiction" is not "fraud." These are the precise words the record supports. They are also, when read together, a significantly more serious accounting than the piece could deliver by escalating the language. The record is strongest when it is allowed to speak in the terms that it itself documents.

The Department of Justice as institution inherits and ratifies the bureau's posture, but also adds to it. A bureau can produce a charging document that contradicts its own video. A DOJ has to then decide whether to litigate on that charging document. The DOJ has decided to litigate, to pursue sanctions against the defense for raising the contradiction, and to seal the filing in which the contradiction was raised. Those are three choices, not one.

The DOJ scorecard is the point at which the piece has to say plainly: the weapon-of-mass-destruction charge is not commensurate with the forensic record the bureau has itself produced. Either the prior testimony was wrong and the current charge is right, in which case the prior testimony has to be addressed. Or the current charge is overstated relative to the forensic record, in which case the question the record invites is why the charge exists. The answer the architecture of the case suggests is: to create plea leverage. That is the reading the record supports. It is not a claim about anyone's state of mind. It is a claim about what the record looks like from outside.

Steve Baker, journalist, is the actor whose scorecard comes with the most framing baggage. The mainstream consensus on Baker is that his article was a uniquely reckless act of journalism that produced a uniquely deserved set of consequences. The record does not support the uniqueness claim on either end. His methodology was not unique; his sourcing was, if anything, more rigorous than the precedents. His failure — not seeking pre-publication comment — was a failure the mainstream press has repeatedly committed. And the consequences that have attached to him are, by any measure against the precedents, categorically more severe. The scorecard therefore has to do two things at once: hold him responsible for what he did do wrong, and decline to hold him responsible for a consequence distribution the record shows is not a measure of his fault.

The Baker scorecard's bottom row is the one that generalizes. A journalist can fail in an identifiable, documented way and still not be responsible for what the ecosystem around him makes of that failure. That is the asymmetry the piece has been tracking from the beginning. It is visible nowhere more clearly than in the comparison of Baker's November 8 article to the 1996, 1999, 2002, and 2013 precedents, because in those precedents the failure was the same and the consequence was nothing.

The mainstream coverage, collectively, is the actor hardest to hold accountable because it is distributed. No single reporter wrote the "cleared" framing. CBS, NBC, the Washington Post, the New York Times, The Bulwark, and Status each published their own version of it, independently, sourced from briefings that were flowing in parallel from the same bureau. That distributed authorship is also what makes the pattern visible: when six major outlets all adopt the same characterization of a case while the bureau's operational record contradicts that characterization, the pattern is not any individual outlet's error.

The mainstream-coverage scorecard lands in the same place the Baker scorecard does, from the opposite direction. Baker is held to a standard that the precedents show was not applied to the mainstream press when the mainstream press made the same error. And the mainstream press, in this case, made a different error — a "cleared" framing not grounded in the record — and has faced no comparable consequence for it. The two scorecards together describe a single ecosystem with two different rules for two different kinds of journalism, applied to the same case.

The two cases are not receiving equal treatment. That is not a question. It is a documented observation. The question — the only question Broadside is prepared to ask directly — is what that unequal treatment is in service of.

The documented argument closes here.

Every claim to this point has been anchored to the record: court filings, sworn congressional testimony, publicly released video, publicly available CCTV, named sources on the record, and published reporting with traceable citations. The findings the piece has made do not depend on motive. They depend on the observable divergence between what was publicly said and what was operationally done, and on the observable asymmetry of consequence between people who asked specific factual questions and people who defended the official account. Those findings stand on their own.

What follows is different. Broadside is now going to do something it has refused to do for the first six thousand words of this piece: reason from the record to hypotheses about what might produce it. This section is labeled speculation. It is not labeled speculation to diminish it. It is labeled speculation to honor the reader — so that nothing that follows can be mistaken for a claim Broadside has proven.

If you only wanted the proven part, this is the exit. The asymmetry is documented. That finding stands regardless of whether any of what follows fits.

If you want to press the question the record raises but does not settle — why the asymmetry is shaped as it is; what such a shape could serve — read on.

What the piece has documented, stripped to its essentials: a federal investigation that failed to produce a suspect for fifty-eight months suddenly produced two parallel tracks in the same month, one of which was publicly denied while it was operationally active. The bureau's own video evidence contradicts its own charging documents. Its own former lead investigator's sworn testimony contradicts its own current superseding indictment. A credentialed forensic analysis of a different suspect is being suppressed. The defense motion that would introduce that analysis is under seal with sanctions pending against the lawyers who filed it. The pattern of friction against anyone asking these questions is consistent across journalists, whistleblowers, sitting congressmen, former FBI personnel, independent analysts, and the defense itself.

Something is producing this shape. The question is what.

We will walk through five hypotheses. Each one is tested against the same checklist of facts. None will be declared true. The goal is to see which ones fit and which ones do not.

Hypothesis 1: Ordinary institutional inertia. Cases get mishandled. Investigations get deprioritized. The FBI had three million data lines and the connection to Cole was missed for four years — mistakes happen. This hypothesis explains the fifty-eight-month delay, the need for a "red team" re-review, and possibly the general reduction of the case's priority under the prior administration. It does not explain why the bureau opened a formal investigation of Kerkhoff days before any media coverage existed. It does not explain the polygraph on November 6, the subpoenas on November 7, the joint AUSA interview on November 12. It does not explain the public denial of those steps on November 13 while the investigation continued. It does not explain why the DOJ's detention memo contradicts the FBI's own released video about whether Cole set the timer. It does not explain the sealed defense motion. It does not explain the cross-administration pattern of friction against questioners — Seraphin was suspended in 2022, under the Biden-era bureau, for concerns he first raised in 2021, and D'Antuono's 2023 sworn testimony remains unrebutted under both administrations. Institutional inertia is always available as an explanation for federal investigations going slowly. It does not produce the specific anomalies this case has produced.

Hypothesis 2: Political pressure to close a high-profile case. The Trump-era bureau needed a win in a case that had embarrassed the previous administration. The "red team" surfaced Cole. The prosecutorial aggression, the WMD charge, and the public messaging are all consistent with an institution motivated to close a long-embarrassing case in a way that serves its leadership's political narrative. This hypothesis explains the Cole arrest timing, the Bondi and Patel press conference framing ("languished for four years"), the speed of the charging documents, and Bongino's public posture. It is a recognizable pattern — politically motivated closure of a case through a suspect who generates a confession. This is also the pattern in which plea bargains typically feature: charge aggressively, offer a reduction for a plea, close the case without testing the evidence in open court. Baker's own 2024 misdemeanor plea — which he says he entered "to avoid the shaming exercise of a trial" — is a structural precedent for exactly this mechanism. This hypothesis explains a lot. What it does not explain is why the Kerkhoff track existed at all. If the bureau was simply building a politically useful closure around Cole, investigating Kerkhoff with the full apparatus — polygraph, surveillance, subpoenas, joint AUSA interview — is a parallel exposure the bureau did not need. It also does not explain the Falls Church surveillance shutdown in 2021, which pre-dates any Trump-era political pressure to close the case. And it does not explain why the FBI's own video contradicts its own charging documents. A politically motivated closure does not require internal documentary self-contradiction.

Hypothesis 3: Protecting a specific person for reasons unrelated to their guilt or innocence. Kerkhoff is a current CIA employee. She was previously a USCP officer who testified at January 6 defendant trials. The bureau may have institutional reasons to protect her that are not about whether she planted the bombs — reasons about what her prosecution or thorough public exoneration would require exposing: her security clearance history, her colleagues at CIA, the process by which she was hired, or what a full evidentiary proceeding about her 2021 whereabouts would surface about any USCP-CIA personnel pipeline. This hypothesis explains the parallel investigation being conducted quietly while the public narrative denied its existence. It explains the anonymous "puppy video" leak through CBS rather than a public evidentiary release — a private citizen's innocence is a CBS scoop; a CIA employee's alibi becomes a question of why she was investigated by the FBI. It explains why the defense's request for her SF-86 is being resisted. It does not, however, explain the internal video contradictions in the Cole prosecution. Protecting Kerkhoff does not require the DOJ to misrepresent what its own video shows about the timer. It does not explain the Falls Church surveillance shutdown in 2021. It does not explain the fifty-eight-month delay in surfacing Cole. And it does not explain the pattern of friction against questioners that pre-dates any Kerkhoff-specific concern — Seraphin's surveillance concerns are from 2021, long before any public identification of Kerkhoff existed.

Hypothesis 4: The operational-event frame. The devices were not what they were publicly presented to be. The timer on the first bomb was not set because it was never intended to be set. The bomb was placed to be found, not to detonate. The bomber chose locations that maximized the narrative impact of the discovery — national party headquarters on the eve of the certification vote — and route features that required geographic familiarity, including a hidden escape gate only locals would know. The devices functioned as theater: a pretext for pulling security resources, a diversion, a plausible context for something adjacent that needed cover. The bureau's internal "training exercise" theory — which Baker's FBI sources say was floated and then quickly abandoned once he went public with the rumor — is the version of this hypothesis that has been reported, however imperfectly sourced. In this frame, the placer was chosen not for ideology but for operational suitability: geographic familiarity, physical match to the intended public description, access to USCP grounds and camera blind spots, and a trajectory that would move them out of public scrutiny afterward. A USCP officer with soccer-forged athletic gait and an active SF-86 clearance application whose subsequent move to the CIA — in this frame — would function as defensive repositioning behind institutional walls that make subsequent investigation harder. This hypothesis explains a lot. It explains the pre-January-5 planning implied by the hidden-gate escape route. It explains the inert devices. It explains why D'Antuono's testimony has been left unrebutted rather than addressed — addressing it would require explaining whether the devices were intended to function as weapons. It explains the Falls Church surveillance shutdown in 2021: a bureau discovering an operational thread no bureau wants publicly litigated. It explains the fifty-eight-month non-investigation under both administrations. It explains the parallel Kerkhoff track in November 2025, the bureau's public denial of the same, the sealed defense motion, and the friction pattern against questioners. It also explains why the Cole prosecution reads as the building of a closure narrative rather than the pursuit of a truth: if the operational frame is correct, the case cannot be truthfully resolved at trial, because a truthful resolution would expose the frame. It requires a plea.This hypothesis is the one Broadside cannot investigate further. It requires documents the government controls and will not release: internal FBI communications about the "training exercise" theory, the Falls Church surveillance shutdown order, the polygraph chart on Kerkhoff, the SF-86 date, the DNC interior security footage, the identity and cooperation of the FBI whistleblower, and testimony from the USCP officer who found the DNC bomb. Without those documents, the hypothesis remains unfalsifiable from outside the institution. It fits the record. It cannot be proven from the record.

Hypothesis 5: Some combination of the above. Institutional inertia, political closure pressure, protection of specific individuals, and operational-event cover are not mutually exclusive. The record can be produced by a mix. A fifty-eight-month delay that began as genuine investigative dysfunction could acquire protective motivations over time as particular leads pointed toward particular people. A political-closure posture in November 2025 can coexist with a simultaneous operational concern that required the parallel track to be conducted quietly. The operational frame does not require everyone in the bureau to know about it; institutions protect things compartmentally, and the protection does not need to be coordinated from the top. This is the most boring hypothesis — multiple ordinary institutional pathologies compounded over four years to produce an extraordinarily abnormal case — and it is the one that most closely matches how institutions actually fail. It is also indistinguishable from hypothesis 4 from the outside, because the operational observable of both hypotheses is identical: a sequence of individually explicable choices that, in aggregate, narrow what can be publicly examined. The difference between hypothesis 4 and hypothesis 5 is whether anyone at any level intended the narrowing. That difference may never be knowable.

The table is not a proof. It is an honest map of where the hypotheses fit and where they do not. Hypothesis 4 covers the most of the record. It is also the one that cannot be tested without documents the people holding them will not produce. Hypothesis 5 covers slightly less of the record but requires fewer specific claims. Hypotheses 1, 2, and 3 each cover significant slices of the record but leave major anomalies unexplained.

The most important observation from this mapping is not which hypothesis is correct. It is that every hypothesis that fits more than half the record requires, as a component, a choice by institutional actors to narrow what can be publicly examined. The shape of the record itself rules out the simple "everyone is doing their job" explanation. That is not speculation. That is what the record forces us to conclude. The speculation begins only when we try to name why.

And the endgame mechanism — the plea bargain — works across all the non-trivial hypotheses. Political closure wants a plea because it closes the case in a way that serves the narrative. Specific-person protection wants a plea because it forecloses discovery. The operational frame requires a plea because a trial would force the kind of evidentiary examination the frame cannot survive. The institutional-combination hypothesis produces a plea as a matter of path dependence: once discovery starts to expose any single pathology, every other pathology's protector has an interest in closing the case quickly. Cole is an autistic defendant with public defenders facing a weapon-of-mass-destruction charge whose underlying factual predicate is contradicted by video the bureau itself has released. Every structural incentive in his case points toward a plea. If he takes one, nothing Broadside has written in this piece gets tested in open court. The sealed motion stays sealed. D'Antuono's testimony stays unrebutted. Nirenberg does not testify. Kerkhoff is not subpoenaed. The SF-86 does not surface. The Falls Church surveillance never gets explained. The DNC interior security video stays suppressed. The case closes. The questions do not.

That is the endgame the architecture produces regardless of which hypothesis is correct.

The speculation closes here. The documented observation remains: two cases, two standards, a consistent pattern of institutional friction against anyone asking specific factual questions, and an endgame mechanism that forecloses every question the piece has raised without any of them being litigated in open court.

What would change this is not more speculation. It is specific documents and specific testimony.

The Standard Form 86. Kerkhoff's clearance application. If dated before January 5, 2021, it establishes she had an active security clearance investigation at the time of the bomb placement — a fact that bears on motive analysis, though not dispositively. If dated after, the CIA transition reads differently. The defense is subpoenaing it. The court's response is pending.

The polygraph chart. Kerkhoff's November 6, 2025 polygraph examination records. Currently held by the FBI. Would be subject to defense discovery at a full Cole trial.

The alleged puppy video and its metadata. Not publicly produced. Currently known only through anonymous leaks to CBS News. Congressional subpoena could compel its production.

Dr. Michael Nirenberg's forensic analysis. The defense is subpoenaing it. His testimony would establish what the Blaze's methodology was and was not capable of concluding.

The FBI's answer to Rep. Loudermilk's "numbers of interest" question. Submitted formally. Not publicly answered.

Steve D'Antuono's full transcribed 2023 testimony. House Judiciary Committee archive. The "inert devices" line is already public. The context around it is not.

The DNC interior security video from January 6, 2021. Would establish whether the device was visible during the Secret Service sweep. Currently not publicly released.

The Falls Church surveillance shutdown order. Would identify the official who terminated the 2021 surveillance of "Person of Interest 3" and the reason given. Currently held by the FBI.

The identity and testimony of the Capitol Police counter-surveillance officer who physically discovered the DNC bomb. Congressional subpoena. He has still not given a transcribed interview to any body.

The FBI whistleblower's full disclosure. Currently protected by counsel against retaliation attempts. May or may not surface depending on congressional protection.

If any of these surface in a form that either confirms or refutes the hypotheses above, Broadside will update this piece and say so. Until they do, the piece stands as written. Two cases, two standards, documented asymmetry, and a closing architecture designed to prevent any of this from being tested.

The bombs did not go off. The investigation has. It has gone off in the hands of everyone who tried to hold it.