4 F.4th 454
7th Cir.2021Background
- Michael Coscia, principal of Panther Trading, was convicted by a jury of six counts of commodities fraud and six counts of spoofing for 2011 high-frequency trading using algorithms that placed large orders meant to be cancelled and small orders intended to execute.
- Government proof included testimony from Coscia’s programmer (Jeremiah Park), ICE and CME data summary charts, testimony from affected traders, Coscia’s CFTC deposition, and expert rebuttal (Hank Bessembinder). The Seventh Circuit affirmed the convictions on direct appeal.
- After the direct appeal, Coscia moved for a new trial based on post-trial ICE/CME audit-trail data and later indictments of other traders, arguing the new data showed errors in charts presented to the jury and that he was not an outlier.
- Coscia separately filed a §2255 petition alleging ineffective assistance: (1) Sullivan & Cromwell had undisclosed conflicts of interest (representation of ICE, and prior representation of D.E. Shaw and Citadel) that adversely affected counsel; and (2) counsel failed to challenge or investigate the statistical evidence.
- The district court denied both the new-trial motion and the §2255 petition. The Seventh Circuit affirmed: it held the post‑trial data and later indictments were, at most, impeaching or cumulative and that Coscia failed to show adverse effect or Strickland prejudice from counsel’s conduct.
Issues
| Issue | Plaintiff's Argument | Defendant's Argument | Held |
|---|---|---|---|
| New trial — post‑trial ICE/CME data | Post‑trial audit trails show errors and additional traders with similar stats; data could not have been obtained earlier | Defense lacked due diligence; new data is mainly impeachment/cumulative and does not undermine verdict | Denied — district court did not abuse discretion; errors were minor or discoverable; evidence not likely to produce acquittal |
| New trial — later indictments of other traders | Subsequent prosecutions show Coscia was not unique/outlier, undermining Government’s theory of intent | Other prosecutions irrelevant; others’ misconduct does not excuse Coscia’s conduct | Denied — subsequent indictments not material to guilt; Coscia’s intent was independently proven |
| §2255 — conflict of interest (ICE, D.E. Shaw, Citadel) | Sullivan & Cromwell concurrently or previously represented government witnesses; nondisclosure created an actual conflict that affected trial strategy | Any conflicts were either non‑existent as to some witnesses or did not adversely affect counsel’s performance | Mixed: Court found a conflict with ICE but no adverse effect; claims about D.E. Shaw and Citadel were not shown to create an actual conflict |
| §2255 — ineffective assistance (independent of conflict) | Counsel failed to challenge, investigate, or impeach statistical summaries and other witnesses; different strategy likely to change outcome | Trial strategy to concede the trading pattern and argue legality/good faith was reasonable; overwhelming evidence (Park, Coscia’s own testimony) made prejudice unlikely | Denied — counsel’s performance fell within reasonable strategic choices; Coscia failed both deficiency and prejudice prongs; no evidentiary hearing warranted |
Key Cases Cited
- United States v. Coscia, 866 F.3d 782 (7th Cir. 2017) (affirming spoofing conviction and analyzing anti‑spoofing statute)
- Cuyler v. Sullivan, 446 U.S. 335 (1980) (standard for showing an actual conflict of interest: must show counsel actively represented conflicting interests and an adverse effect)
- Strickland v. Washington, 466 U.S. 668 (1984) (two‑prong test for ineffective assistance: deficient performance and prejudice)
- Mickens v. Taylor, 535 U.S. 162 (2002) (clarifies actual conflict concept and need to show adverse effect)
- United States v. Reyes, 542 F.3d 588 (7th Cir. 2008) (Rule 33 new‑trial standard for newly discovered evidence)
- United States v. O’Malley, 833 F.3d 810 (7th Cir. 2016) (criteria for newly discovered evidence under Rule 33)