ELMAGIN CAPITAL, LLC v. CHEN et al

Filing 343

(REDACTED) MEMORANDUM AND/OR OPINION. SIGNED BY Circuit Judge STEPHANOS BIBAS ON 8/26/22. 9/16/22 ENTERED AND COPIE E-MAILED. (SEE ECF #338)(mbh)

Case 2:20-cv-02576
Case 2:20-cv-02576
Document
Document
338 *SEALED*
343 Filed 09/16/22
Filed 08/26/22
Page 1
Page
of 111 of 11

IN THE UNITED STATES DISTRICT COURT
FOR THE EASTERN DISTRICT OF PENNSYLVANIA
ELMAGIN CAPITAL, LLC,
Plaintiff,
v.
No. 20-2576

CHAO CHEN; KARL PETTY;
ENTERGRID LLC; and
ENTERGRID FUND I LLC,
Defendants.

MEMORANDUM OPINION
August 26, 2022
In most cases that go to trial, the jury has the final say. That makes sense. Finality
limits litigation and conserves judicial resources. So when a party asks the court to
overturn the verdict or grant a new trial, it shoulders a heavy burden.
Elmagin asks for both, yet fails to carry its burden on either. Thus, I decline to
grant judgment in its favor or order a new trial.
I. BACKGROUND
Dr. Chao Chen helped found Elmagin Capital. There, he helped develop the
algorithmic trading strategies that the company used in wholesale electricity
markets. The company did well.

Case 2:20-cv-02576
Case 2:20-cv-02576
Document
Document
338 *SEALED*
343 Filed 09/16/22
Filed 08/26/22
Page 2
Page
of 112 of 11

Eventually, relationships soured, and Dr. Chen left. With the help of another
former Elmagin employee, Dr. Karl Petty, he formed two new companies to compete
in the same market: Entergrid LLC and Entergrid Fund I LLC.
Elmagin was upset. It believed the duo had ripped off its trading strategies.
Compl. D.I. 1. So it sued Dr. Chen, Dr. Petty, and their new companies for tradesecret infringement. Id. ¶¶ 65–101. Plus, it said the former employees had breached
their contracts, so it sued them for that too. Id. ¶¶102–17.
The jury rejected all Elmagin’s claims. D.I. 287. Unhappy with that result, it has
filed various motions, two of which are pending before me. Elmagin first asks that I
reconsider my denial of its motion for judgment as a matter of law on its contract
claim against Dr. Chen. D.I. 290; Trial Tr. Day 9, at 41:23–42:7 (denying the motion
at trial). It then asks for a new trial. D.I. 328.
I may reconsider my earlier denial of Elmagin’s motion for judgment as a matter
of law if Elmagin shows that: (1) the controlling law has changed; (2) it has new
evidence; or (3) it is necessary to correct a clear error or “prevent manifest injustice.”
Max’s Seafood Cafe ex rel. Lou-Ann, Inc. v. Quinteros, 176 F.3d 669, 677 (3d Cir. 1999).
And I should grant Elmagin a new trial if “the great weight of the evidence cuts
against the verdict and a miscarriage of justice would result” otherwise. Leonard v.
Stemtech Int’l Inc., 834 F.3d 376, 386 (3d Cir. 2016) (internal quotation marks
omitted).
Elmagin lobs a slew of theories in support. D.I. 329. But none persuades. So I deny
both motions.

2

Case 2:20-cv-02576
Case 2:20-cv-02576
Document
Document
338 *SEALED*
343 Filed 09/16/22
Filed 08/26/22
Page 3
Page
of 113 of 11

II. CONTRACT CLAIM
Now, on to the merits. Dr. Chen did not clearly violate his confidentiality
agreement. In a contract with Elmagin, Dr. Chen agreed not to “use … [or] disclose
… any Confidential Information.” D.I. 60-10 ¶ 2(b). And earlier in this case, this Court
held that this agreement was “enforceable to the extent [that it] protect[s] Elmagin’s
trading strategies and any other confidential information related to the strategies.”
D.I. 94, at 20. At trial, the jury found that Dr. Chen had not used or disclosed any of
Elmagin’s confidential information. Elmagin now says the jury got this dead wrong.
In support, Elmagin points to many notes that Dr. Petty took of his conversations
with Dr. Chen. These, it says, prove that Dr. Chen violated his duty not to disclose
confidential information.
1. Elmagin’s

method. Elmagin claims that Dr. Chen

“[d]isclosed Elmagin’s insight that predictions ‘get[ ] better’ by
.” D.I. 329, at 7 (quoting PTX15, at
ENAEO-1-006685). True enough. But the key question is whether this counted as
“Confidential Information” under the parties’ agreement. D.I. 60-10 ¶ 2(a)–(b).
A reasonable jury could conclude that it did not. At trial, one of Defendants’
experts, Dr. Cristian Zarcu, testified that traders use “

…

throughout the investment world.” Trial Tr. Day 6, at 174:19–20. So the jury could
have reasonably concluded that using

was not confidential.

Elmagin counters that its tactic was novel in the power-trading world. D.I. 329, at
8. Plus, it says, Dr. Chen himself admitted that Elmagin considered this step

3

Case 2:20-cv-02576
Case 2:20-cv-02576
Document
Document
338 *SEALED*
343 Filed 09/16/22
Filed 08/26/22
Page 4
Page
of 114 of 11

confidential. Trial Tr. Day 5, at 161:19–23. Still, the jury was entitled to credit Dr.
Zarcu’s testimony over others’. And though it may be probative that Elmagin
considered this approach confidential, it is not dispositive. See Mellon Bank, N.A. v.
Aetna Bus. Credit, Inc., 619 F.2d 1001, 1009 (3d Cir. 1980) (noting that “the words
[that parties] use in their written contract” govern). The jury’s job was to decide what
counted as confidential based on the words in Dr. Chen’s contract, not the intentions
the parties harbored.
. Next, Elmagin

2. Elmagin’s use of
charges that Dr. Chen “[d]isclosed [its] use of
to ‘

and
.’ ” D.I. 329, at 7 (quoting PTX15 at

EN-AEO-1-006703). In short, he disclosed that Elmagin uses

But again, a reasonable jury could find that this tactic was not confidential. Dr.
Zarcu testified that “
.” Trial Tr. Day
6, at 179:8–13. And Dr. Craig Pirrong, Elmagin’s expert, acknowledged that
. Trial Tr. Day 5, at 270:6–9.
3. Elmagin’s strategy for finding the best financial transmission rights. Elmagin
also alleges that Dr. Chen breached his contract when he “[d]isclosed Elmagin’s
method for finding the

.’ ” D.I. 329, at 8

(quoting PTX15 at EN-AEO-1-006687). But as Dr. Pirrong testified, strategies that

4

Case 2:20-cv-02576
Case 2:20-cv-02576
Document
Document
338 *SEALED*
343 Filed 09/16/22
Filed 08/26/22
Page 5
Page
of 115 of 11

are “betting, trying to

… would

want to

.”

Trial Tr. Day 5, at 266:16–19. The jury was free to conclude that Elmagin’s method
of reaching this common goal was common, too.
4. Elmagin’s bid pricing. Elmagin next faults Dr. Chen for “[d]isclos[ing]
Elmagin’s Faber pricing strategy … ‘At [E]lmagin:
.’ ” D.I. 329, at 8 (quoting PTX15 at EN-AEO-1-006676). No matter. Trial
testimony suggested that bid data is public. Dr. Petty said as much. Trial Tr. Day 6,
at 70:9–17; see also Trial Tr. Day 6, at 187 (Dr. Zarcu testifying that markets “give
you the name of the participants [in the markets]”).
5. Elmagin’s use of

. Elmagin also complains that Dr. Chen “[d]isclosed that

at Elmagin, [his] employee ‘use[d] to look for
.’ ” D.I. 329, at 8 (quoting PTX15 at EN-AEO-1-006678).

. Given this, the jury
was not confidential

could have reasonably found that
either.

6. Elmagin’s commodity choice. Elmagin also points out that Dr. Chen disclosed
the following (captured in Dr. Petty’s notes as bullet points): “At elmagin •
•

” D.I. 329, at 8 (quoting

5

Case 2:20-cv-02576
Case 2:20-cv-02576
Document
Document
338 *SEALED*
343 Filed 09/16/22
Filed 08/26/22
Page 6
Page
of 116 of 11

PTX15 at EN-AEO-1-006688). That Elmagin traded financial transmission rights
was no secret. See Trial Tr. Day 6, at 70:9–17; see also Trial Tr. Day 6, at 187:5–12.
And Elmagin never explains what the notes about

mean, much less why that

information is confidential. Without more, I cannot say that these notes warrant
overturning the jury’s verdict or granting a new trial.
7. Elmagin portfolio-construction method. Last, Elmagin says that Dr. Chen
“[d]isclosed [its] technique of ‘portfolio construction’ by ‘
.’ ” D.I. 329, at 8 (quoting PTX15 at EN-AEO-1006704). But
as with Elmagin’s other volleys, this one falls short. Other testimony at trial
suggested that

techniques were not novel. See, e.g., Trial Tr.

Day 6, at 177:11–178:7 (discussing

).

None of Elmagin’s examples warrants granting a new trial or reconsidering my
previous ruling. So I deny its motions on breach of contract.
III. TRADE SECRET MISAPPROPRIATION
Elmagin says it is entitled to a new trial on its claim that Defendants
misappropriated its Breck strategy when creating their Hydra strategy. In support,
Elmagin distills Breck into four high-level steps and compares those to Hydra’s steps.
D.I. 329, at 15–16. Because the strategies are similar at this high level, Elmagin
claims, “there is no rational way to find that Defendants did not misappropriate
[Breck].” D.I. 329, at 18. I disagree.

6

Case 2:20-cv-02576
Case 2:20-cv-02576
Document
Document
338 *SEALED*
343 Filed 09/16/22
Filed 08/26/22
Page 7
Page
of 117 of 11

The jury had to decide whether Defendants had used or disclosed Elmagin’s trade
secrets. D.I. 288, at 9–10 (jury instructions on misappropriation). It could infer “use,”
I instructed, if it found that the parties’ strategies were “similar enough.” Id. at 10.
At trial, the jury heard evidence that the two strategies differed meaningfully. For
one, there was testimony that they winnow the universe of financial transmission
rights differently. Breck uses
232:1. Hydra uses

. Trial Tr. Day 2, at 231:4–
. Trial Tr. Day 5, at 78–82. While Breck

, Hydra begins with

. Trial

Tr. Day 2, at 233; Trial Tr. Day 5, at 81:4–16. For filtering, Hydra uses
, a technique that Dr. Chen believes was not used by Elmagin. Trial Tr. Day
5, at 83:8–84:22, 90:12–21. And to conduct that filtering, Breck uses a “
. Trial Tr. Day 3, at 32–
34; Trial Tr. Day 5, at 83–85 These differences were enough to support the jury’s
verdict.
Not so fast, says Elmagin. Framed at its chosen level of generality, the strategies
are the same. Thus, Elmagin seems to argue that the only parts of the strategies up
for debate are those that it has highlighted. By pointing to differences outside those
high-level steps, it says, Defendants attack a “strawman trade secret theory.” D.I.
335, at 6. Not so.
The jury was allowed to consider differences in the strategies, even those
differences outside the steps now highlighted by Elmagin. Elmagin’s citations to the
master-of-the-complaint doctrine do not change this. It may not now distill its

7

Case 2:20-cv-02576
Case 2:20-cv-02576
Document
Document
338 *SEALED*
343 Filed 09/16/22
Filed 08/26/22
Page 8
Page
of 118 of 11

complex strategies to four key components and then cry foul when defendants point
to other differences. Those are fair game.
Plus, the jury heard testimony from Dr. Zarcu that the strategies are “nowhere
near identical.” Trial. Tr. Day 6, at 187:20–21. He explained that while the strategies
had “commonalities at a very high level,” there was a “vast difference in [their]
results.” Id. at 188. Indeed, the parties’ bids had only a 1–6% overlap. Trial Tr. Day
6, at 186–88. And Elmagin’s own expert, Dr. Pirrong, also emphasized that results
matter when assessing similarity. Strategies are the same, he explained, when they
“perform[ ] the same sort of tasks in the same way[,] leading to … the same if not —
or similar if not exactly identical results.” Trial Tr. Day 5, at 267:21–68:2.
Elmagin is now saying that Defendants stole its grand strategy, so the jury was
wrong to also consider its battle plan and its tactics. But all three levels mattered to
its claim. Elmagin cannot now disregard the two it thinks unimportant. Plus, there
was ample evidence that the strategies differed. So I deny Elmagin a new trial on this
account too.
IV. INDEPENDENT DEVELOPMENT AND EVIDENTIARY RULINGS
Elmagin objects to two last parts of the trial. First, it attacks as against the great
weight of the evidence the jury’s finding that Defendants independently devised their
strategies. And second, it claims that Defendants’ experts should not have been
allowed to testify. I disagree.
1. Independent development. I instructed the jury that “Elmagin must also prove
… that Defendants did not independently develop their allegedly misappropriated

8

Case 2:20-cv-02576
Case 2:20-cv-02576
Document
Document
338 *SEALED*
343 Filed 09/16/22
Filed 08/26/22
Page 9
Page
of 119 of 11

strategies.” D.I. 284, at 10. And in its verdict sheet, the jury found that Elmagin had
not proven “that Defendants did not independently develop each of Entergrid’s
trading strategies.” D.I. 287 ¶¶ 1.3, 3.3.
At trial, Elmagin quibbled only with the burden of proof. In its view, Defendants
should have borne it. Trial Tr. Day 8, at 4. Now, it adds new complaints: the jury’s
finding was “against the great weight of the evidence” and independent development
is “irrelevant.” D.I. 329, at 22. But even if it is right on both points, and even if I were
to overlook its earlier failure to object, I still would not grant a new trial.
If the jury had not been instructed on independent development, it would have
reached the same result on Elmagin’s trade-secret claims. Indeed, the jury was so
skeptical of these claims that it found that Elmagin had brought them in bad faith.
It believed that there was “a complete lack of evidence supporting” Elmagin’s claims.
D.I. 287, at 9 (emphasis added). And though I treated that finding as advisory and
rejected it, it still shows that the jury’s trade-secret findings did not hinge only on
independent development. D.I. 322, at 2. The jury simply did not buy Elmagin’s story.
Since any error on independent development did not prejudice Elmagin, I will not
grant it a new trial for this reason either. See Fed. R. Civ. P. 61.
2. Defendants’ experts. Start with Dr. Mason. He testified that the parties’ source
codes differed. Trial Tr. Day 7, at 108:10–10:18. Elmagin says this confused the jury
and responded to a “strawman argument” because it has never claimed the source
codes are similar or that Defendants stole them. Again, I disagree.

9

Case 2:20-cv-02576
Case 2:20-cv-02576
Document
Document
338 *SEALED*
343 Filed 09/16/22
Filed 08/26/22
Page 10
Page
of 11
10 of 11

For one, Elmagin’s counsel did intimate that Dr. Chen had done something
suspicious with Elmagin’s source code:
Now it was interesting to hear testimony yesterday from Mr. Mason. … And
Mr. Mason said, well, I didn’t see any code checked in by Dr. Chen after August
of 2017. Now, the documents don’t stop at August of 2017. So it seems clear to
me, and you’ll have to decide for yourself, but it seems clear to me that there
was access to all of this information. So why wasn’t the code there?
Trial Tr. Day 8, at 32:1–9. And even if Elmagin did not allege that Defendants had
stolen its source code, it points to nowhere in the trial record where it disavowed that
theory for the jury. Instead, Elmagin argued that Defendants had “misappropriated
[its trade secrets] by [ ] use.” Trial Tr. Day 8, at 25. Through Dr. Mason, Defendants
tried to foreclose one way the jury could have found they “used” Elmagin’s trading
strategies—copying its code.
Nor does Elmagin explain how Dr. Mason’s testimony prejudiced it. It had opening
and closing arguments to explain that it was not alleging code stealing. And it had
cross-examination to impeach Dr. Mason and attack his testimony’s relevance.
Next, consider Dr. Zarcu. He compared the parties’ trades and testified that they
had little overlap. See, e.g., Trial. Tr. Day 6, at 187–88. Elmagin protests that this
testimony was “irrelevant and confusingly prejudicial.” D.I. 329, at 29. First, Dr.
Zarcu conceded that his analysis compared all the parties’ trades, instead of the
specific pairs at issue (Breck/Hydra and Faber/Gryphon). Trial Tr. Day 6 at 203–04.
Plus, it says, he failed to consider that financial transmission rights are correlated.
But Elmagin forfeited these arguments. True, Elmagin objected in limine to Dr.
Zarcu’s testimony. D.I. 64-1. But it did so on other grounds, not these. It claimed that
Dr. Zarcu did not know enough about the field to be an expert and faulted him for
10

Case 2:20-cv-02576
Case 2:20-cv-02576
Document
Document
338 *SEALED*
343 Filed 09/16/22
Filed 08/26/22
Page 11
Page
of 11
11 of 11

comparing the strategies based only “on high level summary descriptions.” Id. at 6.
But by failing to raise the arguments it poses now, it forfeited them. See Fed. R. Evid.
103(a); see also Kiss v. Kmart Corp., 2001 WL 568974, at *6 (E.D. Pa. May 22, 2001).
*****
Elmagin had its day in court. It lost. I should upend that result only if Elmagin can
show that the jury or this court seriously erred. It cannot. So I deny both Elmagin’s
motion to reconsider my earlier denial of its motion for judgment as a matter of law
and its motion for a new trial.

11