Leech faces a five-count indictment returned on November 25, 2024, charging Investment Adviser Fraud, Securities Fraud, Commodity Trading Advisor Fraud, Commodities Fraud, and False Statements. The government alleges that from January 2021 through October 2023, Leech executed Treasury futures and options trades, waited to observe how they performed, and then allocated profitable trades to a favored strategy called Macro Opportunities while routing losing trades to the Core Strategies — all without disclosing the practice to clients. The government also alleges that Leech gave false sworn testimony to the SEC when he said he typically had an allocation in mind at the time he placed a trade.
The court held that none of the proposed expert testimony warranted exclusion under Federal Rules of Evidence 702, 403, or 704(b), or under Federal Rule of Criminal Procedure 16. Judge Gregory H. Woods of the Southern District of New York issued the ruling on April 24, 2026.
The government's two experts — Professor Lauren Cohen of Harvard Business School and Professor Petter Kolm of NYU's Courant Institute — propose to testify that statistical analyses of WAMCO's trade blotter show a pronounced difference in first-day returns between trades allocated to Macro Opps and those allocated to the Core Strategies, that the difference disappears when Leech lacked discretion to choose between the two strategies, and that permutation testing results are inconsistent with random chance as an explanation. The defense moved to exclude that testimony on five grounds: that it impermissibly opined on Leech's mental state, that the underlying return calculations rested on flawed assumptions about allocation timing, that the professors lacked practical fixed-income portfolio management experience, that comparisons to a post-October 2023 trading period were unreliable, and that Cohen's rebuttal disclosure was improper bolstering rather than genuine rebuttal.
The ruling rejected each argument. On the mental-state question under Rule 704(b) — which bars experts in criminal cases from stating whether a defendant had the mental state constituting an element of the offense — the court drew a line between opining on conduct and opining on intent. Describing results as consistent with Leech using first-day performance information to make allocation decisions, or stating that performance differentials do not appear to be driven by portfolio requirements, stops short of expressly stating the inference that Leech acted knowingly and with intent to defraud — the last inferential step the rule reserves for the jury. The court distinguished a ruling by Judge Hellerstein in United States v. Hwang, where an objection was sustained to testimony that trading was consistent with the intent to minimize price impact, noting that Cohen and Kolm do not propose to use the word intent or its derivatives. On reliability, the court concluded that using end-of-day prices or blotter-recorded allocation times as proxies for actual allocation prices was not so unrealistic as to require exclusion, and that the defense's evidence of a lag between Leech's oral allocation instructions and their entry into WAMCO's systems presented a factual dispute for the jury rather than a methodological defect warranting preclusion.
The defense's own experts fared equally well. Charles River Associates vice president Aaron Dolgoff and NYU Stern professor Bruce Tuckman designed and conducted what the parties called the CRA Exercise: a blind allocation experiment in which Tuckman — without knowing Leech's actual allocations or first-day performance — allocated a sample of Leech's trades to representative Macro Opps and Core Plus accounts using duration and convexity targets he developed independently. Dolgoff proposes to testify that Tuckman's allocations matched Leech's at a statistically significant rate of 56.8 percent, and that Tuckman's allocations nonetheless generated statistically significant first-day performance differences between the two strategies. The government moved to exclude the CRA Exercise as unreliable, unrealistic in design, and likely to mislead the jury into thinking Tuckman was replicating Leech's decision-making. The court concluded that the Exercise was designed not to mimic Leech's thinking but to test whether lawful, performance-blind allocation decisions could produce the kind of first-day performance differential the government treats as evidence of fraud — a purpose the court found sufficiently reliable and relevant to survive Daubert scrutiny. The government's concession in its reply that Professor Kolm had made calculation and interpretation errors in critiquing Tuckman's allocation rationales did not change the outcome; the court concluded that Dolgoff's declaration adequately supported the reliability of Tuckman's window-level methodology.
The court did flag three areas of language concern without excluding any testimony. It ordered the parties to meet and confer on alternative phrasing for the government experts' use of using in connection with allocation decisions, and for Tuckman's description of his duration and convexity targets as proxies for Leech's market views — language the court warned could veer toward implying testimony about Leech's good faith. The court also cautioned that framing the government's case as the null hypothesis of the CRA Exercise, as the defense did in its opposition brief, is an inapt use of a statistical term of art that the court expects to sustain objections to at trial absent scientific justification. A status letter on the meet-and-confer and any proposed limiting instruction regarding the post-October 2023 trading period are due two weeks from the date of the opinion.