MANHATTAN (LN) — U.S. District Judge Gregory H. Woods on Monday refused to bar any of the six experts lined up for trial in the government's case against former Western Asset Management Chief Investment Officer S. Kenneth Leech II, ruling that competing statistical analyses of Leech's Treasury futures allocation decisions all clear the reliability threshold under Federal Rule of Evidence 702.
On November 25, 2024, a grand jury indicted Leech on five counts: Investment Adviser Fraud, Securities Fraud, Commodity Trading Advisor Fraud, Commodities Fraud, and False Statements. The government alleges he ran a cherry-picking scheme in which he waited to see how Treasury futures and options trades performed before routing winners to a favored strategy called Macro Opportunities and losers to the Core Strategies he managed for other clients.
At the center of the evidentiary fight are dueling statistical portraits of Leech's allocation decisions across more than 33,000 trades spanning a 33-month period — a trade blotter that the defense noted runs to 595,236 rows and 48 columns, totaling 28,571,328 cells of data.
The government's experts, Harvard Business School Professor Lauren Cohen and NYU Courant Institute Professor Petter Kolm, proposed to testify that trades Leech allocated to Macro Opps had an average first-day return of $243,000, while trades he sent to the Core Strategies averaged negative $309,000 — a gap they said was inconsistent with random chance and consistent with Leech using first-day performance information when making allocation decisions.
The defense countered with a blind allocation exercise designed by Charles River Associates Vice President Aaron Dolgoff and conducted by NYU Stern Professor Bruce Tuckman, a former Chief Economist of the Commodity Futures Trading Commission. Tuckman, given portfolio data but no information about first-day performance or Leech's actual decisions, matched Leech's allocations 56.8 percent of the time — a result Dolgoff said was statistically significantly different from a 50 percent match rate and offered as confirmatory evidence of overlap between the two traders' allocation decisions.
Woods denied the government's argument that the Tuckman exercise should be excluded because it did not satisfy the four reliability factors articulated in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993). The court held that the Daubert factors, developed for scientific testimony, did not map neatly onto an exercise grounded in professional expertise, and that the design, execution, and analysis of the exercise reflected sufficient intellectual rigor. The government's critiques — that Tuckman could not abstain from trading, could not split trades 50/50, and set artificial duration targets — went to weight, not admissibility.
Woods was equally unpersuaded by the defense's bid to exclude Cohen and Kolm. The defense argued the experts' use of words like "using," "driven," and "influenced" in describing Leech's allocation decisions crossed the line drawn by Federal Rule of Evidence 704(b), which bars experts in criminal cases from opining on a defendant's mental state. Woods disagreed, drawing a distinction between testimony that a pattern is consistent with a defendant using information — permissible — and testimony that expressly states the defendant acted with intent to defraud — not permissible. He cited Judge Hellerstein's rulings in United States v. Hwang, 22-cr-240, as an example of where that line falls, noting that in Hwang, Judge Hellerstein sustained an objection to testimony stating whether the use of certain trading strategies was consistent with the intent to minimize price impact, while Cohen and Kolm proposed no such language.
But Woods did not give either side a complete pass. He ordered both sides to meet and confer on mutually agreeable trial language, with a status letter due within two weeks of the order, warning that testimony framed carelessly could draw sustained objections mid-trial. He flagged the defense's use of the word "proxy" to describe Tuckman's duration targets as potentially implying Tuckman was standing in Leech's shoes — a concern analogous to the government's use of "using" in connection with "allocation decisions." He also put the defense on notice that framing the government's case as a "null hypothesis" to be refuted by the CRA Exercise would likely draw a sustained objection absent scientific justification for the term.
Trial is scheduled to proceed with all six experts — Cohen, Kolm, Dolgoff, Tuckman, and Bryant University Finance Chair Kevin Maloney, with Kolm also having a sur-rebuttal disclosure — leaving the jury to sort out whose regression tells the truer story of what Leech did in the hours between executing a trade and deciding where it would land.
Note: The PACER docket caption for this case lists "The United States of America et al v. The Children's Village Inc.," No. 7:23-cv-02703 (S.D.N.Y.), but the opinion concerns United States v. S. Kenneth Leech II. The correct case caption and docket number should be confirmed before publication.