Cross-Trial Analysis
How the government's case evolved from Trial I to Trial II — and what defense counsel should expect.
Trial 2 (February–March 2025) represents a significantly refined and strengthened government presentation compared to Trial 1. Based on what is known about Trial 1 from the user-provided reference points, the following evolutions are evident. First, the FIFO money-tracing methodology that was ruled lay opinion in Trial 1 was replaced in Trial 2 with a more robust forensic accounting presentation: Roase, Blackwell, and Jansma each addressed discrete pieces of the financial picture rather than presenting a single comprehensive FIFO trace, likely insulating the methodology from the same objection. Second, Kara Lomen — who was never called in Trial 1 — remains absent from the Trial 2 witness list, suggesting the government made a deliberate strategic choice not to use her as a witness (possibly because the defense found effective impeachment material that the government chose to avoid in the retrial). Third, while Roase's hidden column methodology issues damaged her in Trial 1, Trial 2's cross-examination grade on Roase suggests prior defense counsel did not successfully replicate the Trial 1 attack — this may mean the government repaired the methodology flaw, or that the prior defense team was simply less prepared to execute the same cross; Defense counsel must assume the flaw was repaired and build the attack from scratch. Fourth, the Sysco food delivery evidence that was excluded in Trial 1 was admitted in Trial 2 (D3-0109 through D3-0127) — but its impact was limited because the defense did not develop it into a comprehensive food-distribution verification framework; in Trial 1 the exclusion may have been more damaging because the defense had built a strategy around it. Fifth, the two full acquittals in Trial 1 appear to have prompted the government to tighten and intensify the cooperating witness program: Trial 2 featured a larger number of cooperators, each corroborating different aspects of the scheme with more specificity about direct interactions with named defendants. Sixth, the mid-cross supplemental wire fraud jury instruction in Vol IV (clarifying that the wire need only be 'in furtherance of' the scheme) suggests the government may have encountered a jury instruction problem in Trial 1 that they corrected procedurally in Trial 2 before the jury deliberated. Seventh, the government's documentary preparation was visibly more aggressive in Trial 2: pre-admitting exhibits like the Shimbirka bank records months before the relevant defense witness appeared, and having exhibits ready to immediately contradict defendant testimony — suggesting Trial 1 taught them to front-load documentary preparation for anticipated defense moves. The overall picture is of a prosecution team that studied Trial 1's weaknesses carefully and systematically addressed each one, while the defense teams in Trial 2 were arguably less effective than whatever defense succeeded in obtaining two acquittals in Trial 1.
- The site witness parade was effective at establishing specific locations where the claimed volumes were implausible — particularly The Landing (97,000 meals, park manager saw zero) and Tot Park (demolished before the claim period).
- Hadith Ahmed's testimony on kickbacks and post-search obstruction, while impeached, was corroborated by documentary evidence (checks, meeting records) and was likely credited by the jury as to the convicted defendants.
- The Abdimajid Nur 'easy trick' email (F-6) was never convincingly explained and likely anchored the conspiracy conviction.
- The Hayat Nur 'Master Document' email (D-42) for editable fake invoices was similarly devastating.
- The Afrique pitch deck (G-110) — though sent to Shariff rather than from him — was visually compelling and established a premeditated plan.
- Financial charts (M-series, N-series) were visually compelling even if methodologically flawed — juries remember charts.
- The government's financial evidence was voluminous and the defense did not have an expert to counter it (DeSanto was excluded).
- The juror bribery, while not the government's doing, almost certainly contributed to guilty verdicts by causing all defendants to be remanded and jury to be sequestered before deliberations concluded.
- The RANDBETWEEN(7,17) formula in roster spreadsheets was the single most compelling evidence of systematic, deliberate fraud — it is mathematically impossible to argue negligence or mistake when a formula is embedded in a document to generate fake data.
- Pole camera evidence at Safari showing approximately 40 daily visitors against 6,000 claimed meals per day was visually simple and devastating; the jury could understand this without any expert explanation.
- Cooperating witness coordination across multiple volumes created an interlocking web: each cooperator corroborated a different aspect of the scheme (kickbacks, meal count inflation, Bock's personal direction), making it harder to attack any single cooperator without the others holding up.
- SA Wilmer's phone forensics — particularly the text analysis showing 'extremely few' Bock-Said texts about food but 'numerous' about checks — distilled the entire defense narrative into a single falsifiable proposition.
- The password-protected spreadsheets with password '1729CCBOCK$' were a powerful smoking-gun exhibit: the password itself contained Bock's name, proving personal connection to the secret accounting system.
- Calling both defendants to the stand was a government gift: Bock was impeached on every significant claim, and Said made a demonstrably false statement about the Sysco contract. The government was fully prepared with pre-admitted exhibits to destroy both witnesses in real time.
- The government's documentary preparation was superior: they had already admitted Shimbirka LLC bank records months before Liban testified, positioned to ambush the defense witness with fully prepared financial cross-examination.
- The Lake Street corridor analysis — 21 sites across 1.8 miles claiming 59,000 meals/day exceeding total MPS enrollment — was a geographic impossibility argument that required no expert and was intuitively compelling.
- Juxtaposing food program money receipt dates with luxury purchase dates (truck purchased day after $140,000 deposit; Mercedes two weeks later; $1.1M house thereafter) told the fraud story through simple arithmetic.
- The Benadir Hall celebration video of operators celebrating Bock for winning the TRO fight personalized and humanized the criminal enterprise in a way documents cannot.
- The government successfully excluded the vast majority of defense documentary evidence through FRE 803(6) business records objections, leaving the defense narrative largely unsupported by documentary proof.
- The CLiCS accountability gap: the government never called a sponsor witness, yet sponsors alone had CLiCS access and submitted all claims. Sapone's closing argument made this the central theme — 'Where is Kara Lomen? This is a gaping hole in this prosecution.'
- Parks' roster methodology was technically flawed and never challenged (missed opportunity by defense, not government's failing — but it is exploitable in future trials).
- Roase's M-series hidden column and excluded vendor invoices left the financial analysis vulnerable.
- FIFO vs. LIFO disparity was established through Blackwell's own concessions — government's money laundering evidence rests on a methodology that produces different results than the alternative.
- The government's opening statement called Bushra Wholesalers a 'shell company' — government's own witnesses refuted this characterization.
- Government chose not to interview Kara Lomen, not to search Bushra warehouse, and not to subpoena Said Farah's email — three evidentiary gaps that defense hammered.
- MDE's 8-9 month continued payment after FBI referral undermines the narrative that the fraud was obvious.
- The COVID-19 waiver defense was present but defense did not deploy it with full technical precision — leaving the government's physical-presence arguments partially unaddressed.
- Thompson's speaking objection ('They can call a forensic accountant if they want to') was a prosecutorial misconduct moment that resulted in a mistrial motion, court instruction to re-hear, and preserved appellate issue.
- SA Kary had not reviewed all available Sysco delivery footage before testifying — the defense successfully introduced footage (D3-0109, D3-0115, D3-0121, D3-0127) showing deliveries he claimed to have been unaware of. This was a significant investigative gap that a better-prepared defense could have exploited more fully.
- No pole camera existed at the Taylor Street site — the government's primary surveillance evidence methodology had a geographic gap that defense identified but did not develop fully.
- The government could not directly connect Bock to the creation of any specific fraudulent roster — only to approving and funding the sites that produced them. The 'willful blindness' inference may be vulnerable.
- The Exhibit 853 metadata controversy was raised publicly by defense but not developed — suggesting a potential authentication vulnerability in the Z5 email that was left unexplored due to defense unpreparedness.
- The government's treatment of the Abshir brothers courthouse hallway incident (approaching cooperating witness Jama) was potentially inflammatory and may have complicated the government's position if defense had raised witness intimidation as a defense narrative.
- Thompson's 'don't lie to this jury' remark to Said on cross was admonished as potentially improper, and Montez preserved a prosecutorial misconduct objection — while the court denied the motion, this is a viable appellate issue and Defense counsel should anticipate and prepare for similar conduct.
- The government was admonished multiple times for editorial commentary ('It is a mystery to me,' 'It sounds like a real hardship job') — Judge Brasel will regulate argumentative questioning if defense objects promptly.
- The USDA regulatory waiver argument (D1-224, Q19 — preserved in Vol XI) was never fully adjudicated — this represents a potentially viable legal defense that was inadequately developed by prior defense counsel.
- MDE's own institutional credibility was damaged: CLiCS auto-approved claims without human review, MDE human reviewers had approved 5,000/day at Safari at one point, MDE was held in contempt of state court, and MDE's stop-pay order was legally invalidated by a court order — this 'regulatory failure enabled the fraud' narrative was available but largely abandoned by defense.
- ZERO cross-examination of Joshua Parks — the single largest missed opportunity in the trial. His roster analysis comparing SFSP open site submissions to school enrollment records is fundamentally flawed, and he was a brand-new agent still in training. Every defense attorney declined to cross. This must not happen in Trial 2.
- Mass stipulated admission of 100+ email exhibits (E-series) in Vol VIII without review — a significant missed opportunity that allowed government to introduce a substantial volume of documentary evidence without challenge.
- Jill DeSanto (defense forensic accountant) was excluded as unavailable on the final day of evidence — representing a logistical failure that left the defense without independent corroboration of food purchase volumes. Must not recur.
- COVID-19 waiver defense was present but not deployed with regulatory precision in closings. No defense attorney made a sustained, technical argument that USDA's non-congregate waiver, parental pickup waiver, and SFSP open-site rules collectively eliminated the physical presence requirement. Government witnesses who testified 'no children were seen' made regulatorily incoherent arguments that were never fully challenged.
- Kara Lomen was extensively referenced but never made the centerpiece of a unified defense narrative. The defense had: (1) her 'turn off the errors' text; (2) Kary's admission she was never interviewed; (3) the Bell finding naming her as a conspirator while she remained uncharged; (4) multiple texts showing Farah collecting money directly from Lomen. These facts were not assembled into a coherent argument that the person who controlled the submission system was never held accountable.
- The Dar Al-Farooq mosque site was a missed opportunity — the Imam's email (C-319) reporting that sites were registered without mosque knowledge was introduced by the government but never fully exploited to show that Kara Lomen registered sites without operator knowledge.
- Vaaler (diaspora expert) reviewed no case evidence — rendering his testimony academic and ineffective. A properly prepared expert would have connected cultural practices to specific transaction facts in evidence.
- Jacob Steen (land-use attorney) was a witness that backfired — his testimony that Shariff never disclosed the food program to him became the government's most effective consciousness-of-guilt moment. The Steen decision should have been vetted more carefully.
- The Exhibit 853 metadata controversy was raised publicly without preparation — if a genuine authentication flaw exists in the Z5 email, a forensic digital expert could have turned this into a significant challenge to a wire fraud count. Instead it became an embarrassing misstep.
- Prior counsel repeatedly attempted to admit defense documentary exhibits (D1-665, D1-904, D1-660, D1-224, D1-996, D1-781, D1-699, D1-983) that were excluded as hearsay — a pattern of unpreparedness that resulted in multiple court admonishments. These should have been litigated through pre-trial motions in limine, not offered for the first time at trial.
- Liban was called without pre-addressing his $195,000 in Shimbirka LLC payments on direct examination — allowing Thompson to present this as a devastating revelation. Any cooperator or witness with financial connections to scheme entities must be frontloaded on direct.
- No defense handwriting expert was called to challenge the forged board minutes, the SIR Boxing submission document, or the forged signatures on A5/A80/A77 — leaving government characterizations of forgery unchallenged.
- No defense forensic accountant was called to challenge Roase's 4.1% food expenditure rate, Blackwell's kickback tracing methodology, or Jansma's money laundering tracing — leaving the government's entire financial narrative unchallenged.
- The fictitious board of directors issue was devastating and foreseeable — defense should have presented evidence of what Bock's actual governance structure was (informal but functional) with supporting communications before Thompson destroyed her on this point.
- Defense did not adequately challenge Parks's methodology on the name-matching analysis — the private school, charter school, and home-school exclusion was identified but not developed into a structured counter-analysis.
- Bock's decision to testify was strategically catastrophic. Her cross-examination resulted in impeachment on virtually every major claim she made. The decision to testify appears to have been made without adequate preparation for the volume of documentary cross-examination Thompson deployed.
- Said's decision to testify was equally catastrophic, compounded by making a provably false statement about the Sysco contract in front of the jury.
- The CACFP/SFSP regulatory waiver framework was underexplored — the argument that USDA COVID waivers created ambiguous compliance requirements was preserved but never developed into a full regulatory defense.
- Defense missed the opportunity to call MDE officials adversarially to establish the institutional failures that enabled the fraud — instead, MDE witnesses were called by the government and cross-examined only moderately effectively.
The most important lesson from prior defense performance is that this case cannot be won through reactive cross-examination of government witnesses alone. The documentary record is so extensive and so damaging that every defense strategy must be built from the documents forward: identify every exhibit the government will introduce, develop documentary counter-narratives with pre-admitted defense exhibits secured through pretrial motions, and never put a client on the stand without verifying that every factual claim the client will make is either supported by an admitted exhibit or cannot be contradicted by one. The defense was repeatedly unprepared — for the Shimbirka bank records, for the text message evidence, for the board member testimony — suggesting a failure of comprehensive pre-trial document review. Defense counsel must review every government exhibit before making any strategic decision.