No. 24-6374

D.C. No. 2:23-cv-06436WLH-E

Appeal from the United States District Court for the Central District of California Wesley L. Hsu, District Judge, Presiding

Argued and Submitted October 7, 2025 Pasadena, California

Filed April 8, 2026

Before: Johnnie B. Rawlinson, Eric D. Miller, and Anthony D. Johnstone, Circuit Judges.

Opinion by Judge Johnstone

Page 2SUMMARY *

The panel affirmed the district court’s summary judgment for the government in Brothers Market LLC No. 2’s challenge to the final decision of the Food and Nutrition Service of the United States Department of Agriculture (the 'Agency') permanently disqualifying the Market from participating in the federal Supplemental Nutrition Assistance Program (“SNAP”).

The Agency charged the Market with trafficking in SNAP benefits.

The panel held that Plaintiffs—the Market and its owner Brad Brown—failed to raise a genuine dispute over whether transactions flagged by the Agency evidenced SNAP trafficking. The government produced evidence of several suspicious transaction patterns that establish an inference of SNAP trafficking. Although Plaintiffs had several opportunities to refute the suspicious patterns with evidence, they did not. Given the overwhelming evidence of SNAP trafficking and Plaintiffs’ lack of a meaningful showing to the contrary, the panel affirmed the district court’s grant of summary judgment for the government.

Page 3Andrew Z. Tapp (argued), Metropolitan Law Group PLLC, Brandon, Florida, for Plaintiffs-Appellants.

Zakariya K. Varshovi (argued), Assistant United States Attorney; David M. Harris, Assistant United States Attorney, Chief, Civil Division; Joseph T. McNally, Acting United States Attorney; Office of the United States Attorney, United States Department of Justice, Los Angeles, California; for Defendant-Appellee.

JOHNSTONE, Circuit Judge:

Brad Brown owns Brothers Market LLC No. 2 ('the Market'), a small convenience store in downtown Los Angeles. The Food and Nutrition Service of the United States Department of Agriculture (“the Agency”) authorized the Market to participate in the federal Supplemental Nutrition Assistance Program ('SNAP'). Many of the Market's customers receive SNAP benefits and purchase food using an electronic benefits transfer ('EBT') card, which transfers funds from a customer’s SNAP account to a SNAP retailer. The Agency prohibits both customers and retailers from 'trafficking' in SNAP benefits: exchanging benefits for cash or consideration other than eligible food. If the Agency determines that a store trafficked in SNAP benefits, it shall disqualify the store from the program.

In February 2022, the Agency began to detect suspicious patterns in the Market's SNAP transactions. Over sixPage 4 months, the Market recorded more than 600 unusually large transactions, nearly 200 transactions that fully or nearly depleted a SNAP household’s monthly benefits, more than 100 sets of rapid transactions made by the same household, and scores of large transactions for the same dollar value. Based on these transactions and a physical inspection of the Market, the Agency charged it with trafficking in SNAP benefits. After the Market’s response and appeal, the Agency issued a final decision permanently disqualifying the store from participating in SNAP. Plaintiffs sought judicial review of the decision, and the district court granted summary judgment for the government. We affirm because Plaintiffs failed to raise a genuine dispute over whether the flagged transactions evidence SNAP trafficking.

Congress enacted SNAP 'to promote the general welfare . . . by raising levels of nutrition among low-income households.” 7 U.S.C. § 2011. SNAP does so by providing funding to “permit low-income households to obtain a more nutritious diet through normal channels of trade.' Id. The Agency provides these households with funds to buy eligible items from approved stores, see id. § 2013(a)(1), by depositing credits to SNAP households' accounts each month, see id. § 2016(h); 7 C.F.R. § 274.2(b). Those credits may be used at an authorized retailer, much like a debit card: a cashier rings up a purchase, a household member pays for the eligible items with an EBT card, and the funds are electronically transferred from the household's SNAP account to the store's bank account. See 7 U.S.C. § 2016(h); 7 C.F.R. § 274.8(b). A store must apply to participate in Page 5SNAP and agree to comply with the program’s regulations. See 7 U.S.C. § 2018; 7 C.F.R. § 278.1(a).

A store may accept SNAP benefits “only in exchange for eligible food,” 7 C.F.R. § 278.2(a), primarily food for “home consumption,' 7 U.S.C. § 2012(k); see 7 C.F.R. § 271.2 (defining eligible foods). A store may not accept SNAP benefits for ineligible food items, like hot foods and alcoholic beverages, or non-food items like toothpaste or cosmetics, which stores often sell alongside eligible items. See 7 U.S.C. § 2012(k); 7 C.F.R. § 278.2(a). A store also may not accept SNAP benefits for cash. 7 C.F.R. § 278.2(a). If a store exchanges SNAP benefits for anything but SNAPeligible items, it commits SNAP “trafficking.” Id. § 271.2; see 7 U.S.C. § 2024(b). The penalty for SNAP trafficking is permanent disqualification from the program. 7 U.S.C. § 2021(b)(3)(B); see 7 C.F.R. § 278.6(e)(1)(i).

The Agency monitors SNAP transactions for trafficking using its ALERT system. See 7 U.S.C. § 2021(a)(2). 'ALERT' stands for “Anti-Fraud Locator using EBT Retailer Transactions.' This system includes a searchable database that records the household, store, date, time, and amount for each SNAP transaction. If the Agency detects an unusual pattern of SNAP transactions at a store, it may refer the matter to a program specialist, who will arrange for a contractor to visit the store and conduct an on-site investigation. After reviewing the relevant data and the onsite investigation report, the program specialist makes a recommendation to the Agency. If the specialist recommends further action, the Agency will issue a charge letter, detailing the allegations against the store. See 7 C.F.R. § 278.6(b). The store may respond to the allegations and present its own evidence to show that the flagged transactions did not occur or were legitimate. Id.

Page 6§ 278.6(b)(1). The Agency then issues its determination based on the charge letter, the store’s response, and any other information available to the Agency. Id. § 278.6(c). If the Agency determines that '[p]ersonnel of the firm have trafficked” in SNAP benefits, it “shall . . . [d]isqualify” the “firm permanently.” Id. § 278.6(e).

Brad Brown opened the Market in April 2020. Soon, it became an authorized SNAP retailer. By early 2022, the Agency began to detect several transaction patterns at the Market that it deemed suspicious. The Agency conducted an in-person inspection, which showed that the Market was a small convenience store with only 540 square feet of retail space. The Market had one cash register, no shopping carts or baskets, and minimal counter space.

The Market stocked fresh bread, dairy products, canned and dry goods, condiments, baking supplies, coffee, and tea. It also offered bananas and several packaged snacks. At the time of the inspection, the most expensive SNAP-eligible item for sale was a $21.99 jar of instant coffee, but the Market had only two jars in stock.

For the next six months, the Agency monitored the Market's SNAP transactions. The Agency summarized its findings in an ALERT report, which compiled data on the Market’s SNAP transactions and information from the onsite inspection. See 7 U.S.C. § 2021(a)(2); 7 C.F.R. § 278.6(a). In the report, the Agency identified hundreds of suspicious transactions based on 'patterns of unusual, irregular, and inexplicable activity' involving unusually large or otherwise suspicious transactions given the Market’s layout and inventory.

In March 2023, the Agency sent the Market a charge letter, formally charging it with SNAP trafficking and listing all the flagged transactions. In response, the Market asserted that “all transactions conducted in our store have been valid and authorized” and added, “we would never engage in any fraudulent activity.' The Market did not dispute that the flagged transactions occurred but tried to explain the suspicious patterns identified in the charge letter. It noted that it is located near a women's rehabilitation center and that because of COVID-19 pandemic restrictions, the center's residents visited the Market to buy 'all the items they needed, as well as items for their friends” who could not leave their homes. The Market asserted that these residents often completed 'multiple transactions with one [EBT] card,” as the Market’s point-of-sale system allowed cashiers to scan all items and hold them until the customer was ready to pay. The Market also attached several months of bank and credit card statements recording non-itemized purchases from various vendors during the reviewed period. See 7 C.F.R. § 278.6(b)(1).

The Agency, after considering the charge letter and the Market’s response, determined that the Market had trafficked in SNAP benefits and permanently disqualified it from SNAP. See id. § 278.6(c). The Market requested administrative review and submitted receipts, photos, and affidavits from customers and employees attesting to their reliance on the store. The Agency then issued its final decision permanently disqualifying the Market from SNAP.

Plaintiffs sought judicial review of the Agency’s permanent disqualification. See 7 U.S.C. § 2023. The government moved for summary judgment. In opposingPage 8 summary judgment, Plaintiffs relied on the same evidence they had put before the Agency-financial statements, receipts, photos, and affidavits-plus an affidavit from Brown. The district court granted the government’s motion for summary judgment because 'Plaintiffs have failed to raise a material issue of fact as to at least one pattern [of] transactions indicative of trafficking.” Plaintiffs timely appealed.

We have jurisdiction under 28 U.S.C. § 1291 and review de novo the district court's grant of summary judgment. Cohen v. Apple Inc. , 46 F.4th 1012, 1025 (9th Cir. 2022). We consider the facts in the light most favorable to Plaintiffs, the nonmoving parties. See Anderson v. Liberty Lobby, Inc. , 477 U.S. 242, 255 (1986).

A store may seek judicial review of its disqualification from SNAP, 'requesting the court to set aside such determination.' 7 U.S.C. § 2023(a)(13). The district court then reviews the Agency’s decision by conducting a “trial de novo . . . in which the court shall determine the validity” of the disqualification-that is, whether the Agency validly determined that the store trafficked in SNAP benefits. Id. § 2023(a)(15); Wong v. United States , 859 F.2d 129, 132 (9th Cir. 1988) (per curiam). In a trial de novo, the district court “is not limited to the administrative record;” it may consider “any relevant evidence” offered by the store to support its case, “whether or not it has been previously submitted to the agency.' See Kim v. United States , 121 F.3d 1269, 1272 (9th Cir. 1997) (citation omitted). This de novo review does not require a trial: a district court may, where appropriate, Page 9resolve a case at summary judgment based on the administrative record and other evidence before it. See id. at 1276-77; see also Irobe v. USDA , 890 F.3d 371, 379–82 (1st Cir. 2018).

The district court here granted summary judgment for the government based on the administrative record before the Agency and Brown’s affidavit. We must affirm that decision if there are no genuine disputes of material fact and the government is entitled to judgment as a matter of law. Fed. R. Civ. P. 56(a). The government produced evidence of transaction patterns that, given the Market's size and inventory, support the Agency’s finding of SNAP trafficking. Plaintiffs do not dispute that the transactions at issue occurred. So to genuinely dispute that the Market trafficked, Plaintiffs must provide evidence, when viewed in the light most favorable to them, that would allow a factfinder to conclude that the flagged transaction patterns do not support a valid trafficking finding.¹

Page 10To determine whether Plaintiffs have done so, we begin by describing the mechanics of summary judgment. Next, we consider the government’s evidence that supported the Agency’s trafficking determination and Plaintiffs’ attempts to dispute that determination. We then assess whether Plaintiffs show, on the augmented record before the district court, a genuine dispute over whether they trafficked in SNAP benefits.

If a store seeks judicial review of its disqualification from SNAP, it bears the burden of showing by a preponderance of the evidence that the Agency’s trafficking determination is invalid. See Kim , 121 F.3d at 1272. So for the government to establish that it is entitled to summary judgment, it must either (1) produce evidence negating an essential element of the store’s case—that the store has not trafficked, or (2) show that the store does not have enough evidence of an essential element to carry its burden of persuasion at trial—again, that the store has not trafficked. Fed. R. Civ. P. 56(c); see Celotex Corp. v. Catrett , 477 U.S. 317, 322–23 (1986); see also Kim , 121 F.3d at 1272. The government attempts the former, relying on the Market's transaction data and the on-site investigation report to show that the store trafficked in SNAP benefits. 7 U.S.C. § 2021(a)(2); see 7 C.F.R. § 278.6(a).

If the government makes this initial showing, the burden shifts to the store, as the nonmoving party, to show the existence of a genuine dispute of material fact. Anderson , 477 U.S. at 249–50. Because the store bears the burden of Page 11proof at trial, see Kim , 121 F.3d at 1272, it must show that a reasonable factfinder could conclude that the Agency's decision was invalid, 7 U.S.C. § 2023(a)(15). In other words, the store must produce sufficient evidence to allow a finding at trial that it did not engage in SNAP trafficking. Fed. R. Civ. P. 56(c); see Anderson , 477 U.S. at 249 (“[T]here is no issue for trial unless there is sufficient evidence favoring the nonmoving party for a jury to return a verdict for that party.”).

When the Agency flags suspicious transaction patterns to validate its trafficking determination, a store can show that it did not engage in SNAP trafficking by presenting evidence supporting a legitimate explanation for each flagged transaction pattern. For example, a store could present its owner's declaration denying the trafficking allegations based on personal knowledge of the flagged transactions. See, e.g. , Betesfa, Inc. v. United States , 410 F. Supp. 3d 132, 139–40 (D.D.C. 2019) (considering store owner’s declaration denying the trafficking allegations and attesting that he had “been present for nearly all EBT transactions at the store'). Or it could present evidence of customer shopping practices at the store that help explain the flagged transactions. See, e.g. , Kahin v. United States , 101 F. Supp. 2d 1299, 1301, 1303 (S.D. Cal. 2000) (observing that evidence that customers bought “food products in bulk” and obtained “their entire monthly food supply at the beginning of the month” may raise a genuine dispute of material fact). Or it could present receipts, inventory records, or supplier invoices showing that its customers’ purchases were consistent with the store's inventory and total sales. See, e.g. , Betesfa , 410 F. Supp. 3d at 140 (noting that a store’s receipts and invoices may show “customers’ purchases are consistent with [the store's] inventory and total receipts'); see also Page 12Cheema v. United States , 365 F. Supp. 3d 172, 189–90 (D. Mass. 2019) ('Had [the store] submitted invoices to this court showing its ability to sell at such a volume, that evidence may have been enough to survive summary judgment on this set of transactions.”); Irobe , 890 F.3d at 378, 381.

This is not to say that a store must produce all these forms of evidence to defeat summary judgment. But when the government meets its initial burden with evidence of trafficking, the store must produce some evidence to genuinely dispute those facts material to the Agency's trafficking finding. General denials of trafficking or speculative theories of how the patterns emerged will not suffice. See Irobe , 890 F.3d at 380–81.

Nor must a store show a genuine dispute about every flagged transaction in a suspicious pattern. Although a single SNAP violation warrants permanent disqualification, see 7 U.S.C. § 2023, the Agency here does not claim that any single transaction proves trafficking. Instead, it relies on patterns of suspicious transactions to find that the store engaged in SNAP trafficking. In such a case, a store need not rebut every transaction at issue to defeat summary judgment so long as it raises triable issues as to each flagged transaction pattern. See Irobe , 890 F.3d at 380 n.3; Euclid Mkt. Inc. v. United States , 60 F.4th 423, 430 (8th Cir. 2023) (rejecting the 'hardline rule that a store cannot prevail without transaction-specific evidence for each transaction”).

In sum, the government bears the initial burden to show that the record does not raise a genuine dispute of material fact. It may do so by offering its own evidence of trafficking, including patterns of suspicious transactions, or by showing that the store’s opposing evidence cannot carry its burden at Page 13trial. If the government makes this initial showing, the burden shifts to the store to show a genuine dispute of material fact. Put another way, the store must show that a reasonable juror could conclude, based on the materials cited to the court at summary judgment, that it did not engage in SNAP trafficking. The store, however, need not address every transaction in each of the transaction patterns on which the Agency relied. To defeat the government's motion for summary judgment, it is enough for the store to show that there are genuine disputes over whether each of those patterns validates the Agency’s trafficking determination.

B. The government produced evidence of several

suspicious transaction patterns that establish an inference of SNAP trafficking.

The government argues that there is no genuine dispute of material fact as to whether Plaintiffs engaged in SNAP trafficking. It relies primarily on transaction data from the ALERT system, which identified hundreds of transactions constituting “patterns of unusual, irregular, and inexplicable activity' during the six-month period. Those transaction patterns included (1) unusually large transactions, (2) benefits-depleting transactions, (3) rapid transactions, and (4) repeated transactions for the same-dollar value. Given the Market’s characteristics, the government argues, these patterns support the Agency’s trafficking determination. Plaintiffs do not deny that the flagged transactions occurred, but dispute that the patterns of transactions show SNAP trafficking. We consider each pattern in turn.

The Agency flagged 649 unusually large SNAP transactions at the Market during the six-month period.

Page 14These transactions ranged from about $37 to $400, with 38 transactions over $150.

In determining whether a SNAP transaction is unusually large, the Agency considers the store's characteristics and available food stock. In this case, the Agency noted that the Market had only 540 square feet of retail space, one cash register, limited inventory, and no shopping baskets or carts.

The Agency also compared the Market’s SNAP transaction data with those of similar stores. It found that during the six-month review period, the Market's SNAP transactions were larger than those of comparable stores: while the average transaction amount for convenience stores in Los Angeles County was $9.88, the Market's was $14.10. See Irobe , 890 F.3d at 379 (comparing store's unusually large SNAP transactions with average SNAP transaction amount in the area). Given the Market’s characteristics and the large number of high-dollar SNAP transactions, this pattern of unusually large transactions shows trafficking.

The Agency also identified 190 sets of SNAP transactions that fully or nearly depleted a SNAP household’s monthly benefits in a single day. For example, the Market processed two transactions totaling $445.96 within minutes, leaving only $1.15 in monthly benefits for that household. Another time, the Market processed four transactions totaling $351 within five minutes, leaving just $0.99 in benefits. And another time, the Market processed three transactions totaling $304 in about one minute, depleting all the household's monthly benefits.

Page 15The Agency determined that these transactions show SNAP trafficking because “[i]t is unusual that a household would deplete, or very nearly deplete, their entire SNAP balance in one day leaving nothing, or nearly nothing, until their EBT card is replenished with funds.' It cited a government study showing that an average household takes three weeks to spend 90% of its monthly SNAP benefit allotment. See Laura Castner & Juliette Henke, U.S. Dep’t of Agric., Food & Nutrition Serv., Benefit Redemption Patterns in the Supplemental Nutrition Assistance Program , 32-33 (2011). That study also revealed that households typically redeem their SNAP benefits at supermarkets and supercenters—about 4% of households using SNAP never shop at a supermarket. Id. at 70. Thus, it is unusual for a household to spend nearly all its monthly SNAP benefits at a convenience store. Yet the Market conducted almost 200 sales over six months that depleted a household's SNAP benefits, even though several supermarkets were nearby. By contrast, comparable stores in the area conducted only six benefits-depleting transactions during the same period.

Plaintiffs do not dispute that the flagged transactions depleted or nearly depleted those households' monthly SNAP benefits. Nor do they dispute the government study showing that households rarely deplete their monthly SNAP benefits in a single day or at a convenience store. All told, this pattern-nearly 200 SNAP transactions that fully or nearly depleted a SNAP household’s monthly benefits in a single day—shows trafficking.

In addition to large and benefits-depleting transactions, the Agency flagged 127 sets of SNAP transactions conducted within a few minutes. For example, the Market Page 16made two sales to the same household, the first for $101.22 and the second for $115.36, within 17 seconds. The Market also made three large sales to a second household, one minute apart, totaling $304; to a third household, four sales, four minutes apart, totaling $307.71; and to a fourth household, four sales, six minutes apart, totaling $351. Eighteen times, the Market processed multiple transactions from the same household totaling $100 or more in under 30 seconds.

The Agency concluded that these transactions evidence SNAP trafficking because “[m]ultiple transactions, conducted within a set time period, are methods which stores use to avoid single high dollar transactions that cannot be supported and are indicative of trafficking.' According to the Agency, it is implausible that these rapid transactions involved actual purchases of eligible food in such large amounts. Because of the Market's small size and limited inventory, the cashier would need to ring up several eligible food items at a small, crowded counter without a conveyor belt or bagging area. The cashier would need to complete several steps to account for each SNAP transaction, then repeat them a few seconds later. Considering the Market’s characteristics and the multi-step process needed to complete a single transaction, this pattern of rapid transactions shows trafficking.

Lastly, the Agency found that the Market processed several SNAP transactions with the same dollar value. Although the Agency identified hundreds of same-dollarvalue transactions above $30, it noted that these transactions typically clustered at $99 and $100. For example, the Market conducted 23 transactions for $100 and 29 transactions forPage 17 $99, each with varying amounts of cents. While the Market processed 52 transactions for $99 or $100, nearby SNAPauthorized stores processed only three to eight during the same period. These transactions stood out to the Agency because a store’s transactions typically peak at the store’s average amount, then gradually decrease as transaction amounts increase. It is therefore unusual for a store's transactions to spike at specific dollar amounts, much less at amounts far exceeding the store’s average transaction amount. For this reason, the Market’s pattern of same-dollarvalue transactions for $99 or $100 shows trafficking.

The government's evidence of transaction patterns is circumstantial, but its cumulative effect is compelling. See Irobe , 890 F.3d at 380 ('Irregular patterns may emerge in virtually any retail operation, but a drumbeat of irregularities can be highly probative of unlawful conduct.”); cf. Laurins v. Comm’r , 889 F.2d 910, 913 (9th Cir. 1989) (considering, in the tax context, a pattern of unusual conduct to support an inference of fraud). It reveals several suspicious transaction patterns, each comprising scores of undisputed transactions by the Market. Plaintiffs do not meaningfully challenge the government’s evidence. They offer Brown’s affidavit, customer and employee affidavits, financial statements, and a set of receipts. Of the hundreds of flagged SNAP transactions, the receipts correspond to just a handful, while the other materials advance general theories attempting to explain the suspicious patterns. Even when viewed in the light most favorable to Plaintiffs, no reasonable factfinder could conclude from this evidence that the Market did not engage in SNAP trafficking.

Page 18In his affidavit, Brown generally denies the trafficking charges and offers several theories to account for the suspicious transaction patterns. For the unusually large and benefits-depleting transactions, Brown asserts that many of the Market’s customers were nearby shelter residents who used their SNAP benefits to buy large quantities of nonessential food, while others were homeless or transient individuals who purchased frequently or in large quantities. He suspects that many of these customers later resold or traded the food for cash or other items. But Brown does not attest to the legitimacy of any one transaction based on his personal knowledge. Nor does he identify any differences between the Market and other SNAP-authorized stores near the shelter that would explain why the Market conducted so many benefits-depleting transactions compared with those other stores. And Plaintiffs offer no evidence beyond Brown’s affidavit to support their theories about the customers’ shopping behavior.

For the rapid transactions, Brown attests that the Market's cash register can pause and resume transactions and thus “process several high-dollar transactions consecutively.” He says that the Market uses this feature “to allow multiple customers to make selections at the same time.' Yet he does not say that the Market ever used this feature to process rapid transactions for the same SNAP household. Brown's affidavit also does not address the repeated same-dollar-value transactions or their frequency compared with other stores. Without other evidence supporting Brown's affidavit, Plaintiffs cannot establish a genuine issue of material fact. See FTC v. Publ’g Clearing House, Inc. , 104 F.3d 1168, 1171 (9th Cir. 1997) ('A conclusory, self-serving affidavit, lacking detailed facts and Page 19any supporting evidence, is insufficient to create a genuine issue of material fact.”).

The Market's employee and customer affidavits offer little more. To begin, no employee or customer attests that any of the flagged transactions actually took place. Nor do they attest that the Market used its cash register to pause and resume large transactions for SNAP-eligible items. They also do not address the benefits-depleting or same-dollarvalue transactions.

Plaintiffs offer financial statements, too, but the nonitemized statements show only the Market’s total payments to vendors. They do not show how many SNAP-eligible items the Market stocked or how expensive those items were. Similarly, Plaintiffs’ photos of the Market’s inventory and storage space do not show whether the Market had enough SNAP-eligible items to conduct the hundreds of unusually large, benefits-depleting, and rapid transactions during the relevant period.

That leaves the 75 receipts, the only evidence corresponding to any of the flagged transactions. Although most of the receipts record large SNAP transactions, those transactions account for only a small fraction of the flagged large transactions, leaving hundreds unaccounted for. The receipts also do not correspond to any of the flagged benefits-depleting transactions. And while the receipts correspond to some rapid transactions, they document only one out of the 52 transactions for $99 or $100: a sale for $99.99 consisting of two 36-count boxes of Twix candy bars for $80, six 'Deluxe cream cookies?' for $18, and one “Jack” for $1.99.

Although plaintiffs are not required to produce evidence for every flagged transaction, these few receipts are not Page 20enough to create a genuine dispute when hundreds of undisputed transactions support the suspicious patterns at issue. See Zellmer v. Meta Platforms, Inc. , 104 F.4th 1117, 1122 (9th Cir. 2024) (“The mere existence of a scintilla of evidence in support of the plaintiff's position will be insufficient; there must be evidence on which the jury could reasonably find for the plaintiff.' (quoting Anderson , 477 U.S. at 252)). Even if a jury were to credit the receipts as establishing some legitimate transactions, the remaining evidence would still show several patterns of suspicious transactions on which the Agency’s finding rests. Without more, a reasonable factfinder could not conclude from the Plaintiffs' scattershot business records that the remaining transactions were legitimate. They do not create a genuine dispute over whether the Market engaged in SNAP trafficking. See Irobe , 890 F.3d at 381 (“Nothing about the purchases evidenced by the receipts impugns the agency’s finding that, on many occasions, the Store trafficked in SNAP benefits.”).

The record shows that the Market processed hundreds of unusually large transactions, nearly 200 benefits-depleting transactions, more than 100 rapid transactions, and scores of large same-dollar-value transactions. Although Plaintiffs had several opportunities to refute these suspicious patterns with evidence, they did not. The Market’s invoices do not show that it stocked enough eligible food for these transactions. Its receipts do not show that it accounted for eligible food in enough of these transactions. Its employees do not say that they sold eligible food in any of these transactions. And its customers do not say that they bought eligible food in any of these transactions. What little material evidence Plaintiffs do cite is not enough to create a genuine dispute over whether Page 21the Agency's finding of trafficking is valid. Given the overwhelming evidence of SNAP trafficking and Plaintiffs’ lack of a meaningful showing to the contrary, we affirm the district court’s grant of summary judgment for the government.²