Best Buy Co., Inc., is a leading retailer of consumer electronic products and services. - Plaintiffs sued Best Buy and three of its executives (collectively, “defendants”) *777alleging they violated Rule 10b-5 of the federal securities laws1 by making fraudulent or recklessly misleading public statements. Plaintiffs alleged that statements in a press release and ensuing conference call with securities analysts on September 14, 2010, artificially inflated and maintained Best Buy’s publicly traded stock price until the misstatements were disclosed when Best Buy reported its quarterly earnings on December 14. After the district court dismissed claims relating'to the press release, plaintiffs moved to certify their claims relating to the conference 'call as a class action. Plaintiffs relied on the fraud-on-the-market presumption to satisfy the Rule 23 requirement that common questions predominate in proving their Rule 10b-5 claim. See Basic, Inc. v. Levinson, 485 U.S. 224, 241-47, 108 S.Ct. 978, 99 L.Ed.2d 194 (1988). Defendant:? contended they rebutted the presumption; plaintiffs responded that rebuttal-evidence was not admissible at the class certification stage.
The district court stayed the class certification motion until the. Supreme Court resolved that issue in Halliburton Co. v. Erica P. John Fund, Inc., — U.S. -, 134 S.Ct. 2398, 2414-16, 189 L.Ed.2d 339 (2014) (Halliburton II). The district court then certified a class consisting of all purchasers of Best Buy stock between September 14 and December 14, 2010, concluding that common questions predominate because defendants failed to rebut the Basic presumption by establishing that the challenged statements did not impact Best Buy’s publicly-traded stock price: We granted defendants permission to appeal this interlocutory ruling. See Fed.R.Civ.P. 23(f). Addressing an issue of first impression, we conclude the district court misapplied the price impact analysis mandated by Halliburton II and therefore reverse.
I. Factual Background.
On September 14, 2010, Best Buy issued a press release summarizing its reported financial performance for the- second quarter of its 2011 fiscal year, which ended on August 28, The press release, issued at 8:00 A.M., before the stock market opened, announced that Best Buy was increasing its, full-year earnings per share (EPS) guidance by ten cents to ,$3.55-$3.70, Best Buy’s common stock (BBY), which closed the prior, day at $34.65, opened at 9:30 A.M. at $37,25, up 7.5%.
At 10:00 A.M. on September 14, Best Buy’s Chief Executive Officer, Brian J. Dunn, and its Chief Financial Officer, Jim Muehlbauer, held a .conference call with analysts. The call began with a reminder to investors . that the executives’ discussion “may contain forward-looking .statements, which are-.subject to risks and uncertainties,” and that the company’s SEC filings contained additional information about these factors. During the call, Muehl-bauer said:, (1) “looking at the results for the first half of fiscal 2011, while there are many moving pieces that we manage, like always, we are. pleased that our earnings are essentially in line with our original expectations for the year”; and (2) “Overall, we are pleased that we are on track to deliver and exceed our annual EPS guidance.” (Emphases added.) BBY closed on September 14 at $36.73; 21.3 million shares were traded that day.
At 8:00 A.M. on December 14, Best Buy issued-a press release reporting a decline iñ fiscal third quarter sales and announcing that it had reduced the FY 2011 EPS guidance to $3.20-$3.40. - In a conference call after the market opened that day, *778Dunn and Muehlbauer discussed the “lower than expected” third quarter sales. BBY’s price, which had risen after September 14 (though not in a straight line) to close at. $41.70 on December 13, closed at $35.52 on December 14, down 14.8%; 64.7 million shares were traded that day.
Plaintiffs filed this action on February 18, 2011. On March 20, 2012, the district court dismissed the Amended Complaint with prejudice, concluding that plaintiffs failed to allege sufficient facts showing that misrepresentations on September 14, specifically including the conference call statement that Best Buy was “on track to deliver and exceed our annual EPS guidance,” were not forward looking statements protected by the PLSRA’s Safe Harbor provision.2 The court denied plaintiffs’ request to further amend their complaint. However, in October 2012, after a further hearing, the court granted plaintiffs’ motion for leave to file a First Amended Class Action Complaint.
The First Amended Class Action Complaint alleged that defendants made three actionable false and misleading statements on September 14, 2010: the statement in the press release increasing Best Buy’s FY 2Ó11 EPS guidance to $3.55-$3.70, and two statements in the conference call: “we are ón'track to deliver and exceed our annual EPS guidance,” and “our earnings are essentially in line with our original expectations for the year.” In an Order dated August 5, 2013, the district court dismissed the claim based' on the EPS guidance statement in the press release because it was forward-looking, and accompanied by meaningful cautionary statements in B.est Buy’s Form 8-K and Form 10-K SEC filings. However, the court denied defendants’ motion to dismiss the claim based on the conference call statements because they were “not forward-looking and are, therefore, actionable as a statement of present condition.” The court denied defendants’ request to certify that interlocutory ruling for immediate appeal. Thus, not presently before us is the question whether those statements, “when read in context, cannot meaningfully be distinguished from the future projection of which they are a part.” Avaya, 564 F.3d at 255.3 Plaintiffs then filed the motion for class certification here at issue.
II. Framing the Price Impact Issue.
Plaintiffs in Rule 10b-5 fraud actions must prove six elements: “(1) a material misrepresentation or omission by the defendant; (2) scienter; (3) a connection between the misrepresentation or omission and the purchase or sale of a security; (4) reliance upon the misrepresentation or omission; (5) economic loss; and (6) loss causation.” Amgen Inc. v. Conn. Ret. Plans & Tt. Funds, — U.S. -, 133 S.Ct. 1184, 1192, 185 L.Ed.2d 308 (2013) (quotation omitted). To be certified, a proposed class must show that “questions of law or fact common to class members predominate 'over any questions affecting only individual members.” Fed,R.Civ.P. *77928(b)(3). Defendants contest the district court’s Rule 23 predominance ruling.4 .A district court abuses its discretion if it certifies a class that does not meet this requirement. Halvorson v. Auto-Owners Ins. Co., 718 F.3d 773, 780 (8th Cir.2013), We review the grant of class certification for abuse of discretion, but review legal questions de novo. In re St. Jude Med., Inc., 425 F.3d 1116, 1119. (8th Cir.2005).
In Basic, 485 U.S. at 245, 108 S.Ct. 978, the Supreme Court observed that requiring each investor to prove individual reliance on an alleged fraudulent misrepresentation “would place an unnecessarily unrealistic evidentiary burden on the Rule 10b-5 plaintiff who has traded on an impersonal market,” because each investor would have to prove “how he would have acted ... if the misrepresentation had not been made.” Concluding this would be contrary to the statutory purpose of protecting investors trading on weli-developed securities markets, the Court held that Rule 10b-5 plaintiffs may invoke a rebut-table fraud-on-the-market presumption of reliance.
The presumption is based on the theory “that the market price of shares traded on well-developed markets reflects all publicly available information,” and therefore “[a]n investor who buys or sells stock at the price set by the market does so in reliance on the integrity of that price.” Id. at 246-47, 108 S.Ct. 978. The presumption applies if a Rule 10b-5 plaintiff demonstrates “(1) that the alleged misrepresentations were publicly known, (2) that they were material, (3) that the stock traded in an efficient market, and (4) that the plaintiff traded the stock between the time the misrepresentations were made and when the truth was revealed.” Halliburton II, 134 S.Ct. at 2408; see Basic, 485 U.S. at 248-50, 108 S.Ct. 978. If a defendant rebuts the presumption, “a Rule 10b-5 suit cannot proceed as a class action” because individual questions of reliance will predominate over common questions of law and fact. Halliburton II, 134 S.Ct. at 2416. Thus, establishing the Basic presumption is critical to a Rule 10b-5 plaintiffs Rule 23 burden to establish that a class action should be certified.
Plaintiffs’ motion for class certification relied on Basic’s fraud-on-the-market presumption to prove ’the reliance element of théir Rule T0b-5 claims. In support, plaintiffs submitted a report by their expert, Bjorn I. Steinholt, to demonstrate that Best Buy’s stock traded in an efficient' market. Based on an “event study,”5 he concluded that the BBY stock price increased in reaction to the three allegedly misleading statements on September 14. However, the' report did not differentiate between the effects of the 8:00 A.M. press release and the 10:00 A.M. conference call.
Defendants countered with an event study by their expert, Kenneth Lehn, who found that the price increase on September 14 occurred after the press release but before. the call. As the price before the call was almost identical to the September 14 closing price, Lehn concluded the “on track” and “in line” conference call statements “had no discernible impact on Best Buy’s stock price.” In response, Steinholt submitted a revised opinion. He agreed that the conference call statements did not immediately increase the stock price be*780cause “the economic substance” was disclosed in the press release, and thus “by the time the 2Q11 conference call started, the economic substance of the alleged misrepresentations was largely reflected in Best Buy’s stock price.” However, he opined, “even if the 2011 EPS guidance given in- the 2Q11 [press] release is not actionable, this does not mean that investors did not give it greht weight.” He then concluded that the price decline on December 14 demonstrated that the conference call statements fraudulently maintained the BBY stock price until the “corrective disclosure” on December Í4.'
Defendants do not contest, that the alleged conference call misrepresentations were publicly made, that Best Buy’s stock traded in an efficient public market,, or that the district court erred in defining the class time periqd. (The district court removed from the certified class those who. bought BBY stock after the non-actionable press release but before the conference call.) Materiality is another “essential predicate” to invoking the fraud-on-the-market presumption. Amgen, 133 S.Ct. at 1195. However, the Supreme Court concluded in Amgen, “[b]ecause the question of materiality is common to the class, and because a failure of proof on that issue would not result in questions ‘affecting only individual members’ predominating,” the Rule iOb-5 plaintiff “was not required to prdve the materiality of [defendant’s] alleged misrepresentations- and omissions at the class-certification stage.” Id. at 1197.
As Amgen left materiality for the merits stage of a Rule 10b-5 class action, it is clear that plaintiffs made the prima facie showing required to invoke the presumption of reliance for the purpose of class certification. But the protracted Halliburton litigation clarified that there is an additional question at the class certification stage — whether the Rule 10b-5 defendant can rebut the Basic presumption with evidence showing an absence of price impact, that is, evidence that “severs the link between the alleged misrepresentation and either the price received (or paid) by the plaintiff,' or his decision to trade at a fair market price.” Halliburton II, 134 S.Ct. at 2408, quoting Basic, 485 U.S. at 248, 108 S.Ct. 978; see Local 703 v. Regions Fin. Corp., 762 F.3d 1248, 1259 (11th Cir.2014).
In Erica P. John Fund, Inc. v. Halliburton Co., 563 U.S. 804, 131 S.Ct. 2179, 180 L.Ed.2d 24 (2011) (Halliburton I), the Supreme Court reviewed a circuit court ruling that Rule 10b-5 plaintiffs must prove loss causation to invoke the Basic presumption. In reversing, the Court distinguished between price impact (“transaction causation”) and loss causation. Id. at 2186. The Basic presumption addresses whether the plaintiff investor relied on defendant’s publicly loiown misrepresentations “at the time of the relevant [initial] transaction.” Loss causation, on the other hand,' requires plaintiff to prove that the misrepresentation “also caused a subsequent economic loss” that may instead have in fact been caused by other factors. Id. (emphasis in original); see generally Dura Pharm., Inc. v. Broudo, 544 U.S. 336, 343, 125 S.Ct. 1627, 161 L.Ed.2d 577 (2005) (listing factors such as “changed economic circumstances, changed investor expectations, new industry-specific or firm-specific facts, conditions, or other events”). The Court concluded that “[l]oss causation has no logical- connection to the facts necessary -to establish the efficient market predicate to [Basic’s ] fraud-on-the-market theory.” Halliburton I, 131 S.Ct. at 2186.
On remand from Halliburton I, the circuit court affirmed class certification, concluding that defendant could not use price impact evidence to rebut the presumption of reliance at the class certification stage *781because it does not bear on the Rule 23(b)(3) predominance issue. The Supreme Court granted certiorari to consider “whether we should overrule or modify Basic’s presumption of reliance and, if not, whether defendants should nonetheless be afforded an opportunity in securities class action eases to rebut the presumption at the class certification stage, by showing a lack of price impact.” Halliburton II, 134 S.Ct. at 2405. The Court unanimously vacated the grant of class certification and remanded for further proceedings. Addressing the first question, the majority opinion by Chief Justice Roberts declined to overrule Basic’s fraud-on-the-market presumption of reliance in Rule 10b-5 class action cases. Id. at 2407-13. Three Justices, concurring in the judgment, concluded that “Basic should be overruled.” Id. at 2418 (Thomas, J., concurring).
Turning to the second question,, the Court partially rejected Halliburton’s contention that Rule 10b-5 plaintiffs must “prove that a defendant’s misrepresentation actually affected the stock price — so-called ‘price impact’ — in order to invoke the Basic presumption” at the class certification stage:
Because market efficiency is not a yes- or-no proposition, a public, material misrepresentation might not affect a stock’s price even in a generally efficient market. But ... Basic never suggested otherwise; that is why it affords defendants an opportunity to rebut the presumption by showing, among other things, that the particular misrepresentation at issue did not affect the stock’s market price.
Id. at 2413-14. The Court then reversed the circuit court’s holding that evidence of lack of price impact, like evidence of materiality, should be left to the merits stage because it does not bear on the predominance requirement:
Price impact is different [than materiality]____ [T]hat [a misrepresentation] had price impact ... is Basic’s fundamental premise. It thus has eveiything to do with the issue of predominance at the class certification stage. That is why, if reliance is to be shown through the Basic presumption, the publicity and market ’efficiency prerequisites must be proved '[by plaintiffs] before class certification.
* # # ■ * if!
Our choice in this case, then, is not between allowing price impact evidence at the class certification stage or relegating it to the merits. Evidence of price impact will be before the court at the certification stage in any event. The choice, rather,, is between limiting the price impact inquiry before class certification tó indirect evidence, or allowing consideration of direct evidence as well. As explained, we see no reason to artificially limit the inquiry at the certification stage to indirect evidence of price impact. Defendants may seek to defeat the Basic presumption at that stage through direct as well as indirect price impact evidence. -,
... [Defendants must be afforded an opportunity before class certification to defeat the presumption through’ evidence thát an .alleged misrepresentation did not actually affect the market price of the stock.
Id. at 2416-17. In a brief concurring opinion, three Justices noted that “the Court recognizes that it is incumbent upon the defendant to show the absence of price impact.” Id. at 2417 (Ginsbürg, J., concurring).
III. Discussion.
On August 6, 2014, after awaiting the decision in Halliburton II; the district court granted plaintiffs’ motion for class *782certification. The court recognized that Halliburton II permits Rule 10b-5 defendants to rebut the fraud-on-the-market presumption at the class certification stage with evidence of no price impact but concluded that defendants did not adequately rebut the presumption. The court agreed with expert Steinholt’s revised opinion: “Even though the stock price may have been inflated prior to the earnings phone conference,” the court reasoned, “the alleged misrepresentations could have further inflated the price, prolonged the inflation of the price, or slowed the rate of fall.” The court held that “price impact can be shown by a decrease in price following a revelation of the fraud,” and defendants “have not offered evidence to show that Best Buy’s stock price did not decrease when the truth was revealed.”
We agree with the district court that, when plaintiffs presented a prima fade case that the Basic presumption applies to their claims, defendants had the burden to come forward with evidence showing a lack of price impact. See Fed.R.Evid. 301 (“the party against whom a presumption is directed has the burden of producing evidence to rebut the presumption”). But what the district court ignored, in our view, is that defendants did present strong evidence on this issue — the opinion of plaintiffs’ own expert. Steinholt opined that the “economic substance” of the non-fraudulent press release statements and the alleged misrepresentations in the immediately following conference call was “virtually the same,” and that the two “would have been expected to be interpreted similarly by investors.” His event study showed that the forward-looking EPS guidance in the press release had an immediate impact on the BBY market price, whereas the confirming statements in the conference call two hours later had no additional price impact.
Steinholt opined that investors gave the EPS guidance in the press release “great weight.” Combined with the absence of further price impact following the conference call, this was direct evidence that investors did not rely on the executives’ confirming statements two hours later. Defendants’ expert, Lehn, agreed with this analysis. And it is consistent with common sense. In the September 14 conference call, CFO Muehlbauer said Best Buy was “on track” to meet the EPS guidance announced just two hours earlier. This was not a case where a third party professional confirmed the accuracy of a company’s inaccurate financial statements, as in McIntire v. China MediaExpress Holdings, Inc., 38 F.Supp.3d 415, 434 (S.D.N.Y.2014). Earnings projections are statements of what a company is “on track” to do; thus, the Best Buy executives’ conference call statements added nothing to what was already public. As in In re Moody’s Corp. Securities Litigation, 274 F.R.D. 480, 493 (S.D.N.Y.2011), “Plaintiffs’ own expert conclude[d] that there is no link between the price of [Best Buy’s] stock and any of the alleged misrepresentations because these misrepresentations just reflected the ‘status quo.’ ”
This overwhelming' evidence of no “front-end” price impact rebutted the Basic presumption. Plaintiffs’ contrary theory — that the conference call statements effected a gradual increase in stock price between September and December — was contrary to the efficient market hypothesis on which the Basic presumption of reliance is based. See, e.g., In re Xcelera.com Sec. Litig., 430 F.3d 503, 508 (1st Cir.2005) (an efficient market rapidly reflects all publicly available information). Plaintiffs further argued, and the district court agreed, that BBY’s price drop after the December 14 “corrective disclosure” was evidence of the requisite price impact — maintaining *783an inflated stock price. Cf. FindWhat Investor Group v. FindWhat.com, 658 F.3d 1282, 1314 (11th Cir.2011); Schleicher v. Wendt, 618 F.3d 679, 683 (7th Cir.2010). But that theory provided no evidence that refuted defendants’ overwhelming evidence of no price impact. The allegedly “inflated price” was established by the non-fraudulent press release. Expert Steinholt attributed the entire September 14 price impact to the non-fraudulent EPS guidance in the press release. The substance of the conference call statements two hours later was “virtually the same” and had no immediate impact on. that price, impact the Basic presumption would otherwise presume.
A district court abuses its discretion if it fails to conduct the “rigorous analysis” Rule 23 requires. Powers v. Credit Mgmt. Servs., Inc., 776 F.3d 567, 569 (8th Cir.2015) (quotation omitted). The court “must conduct a limited preliminary inquiry, looking behind the pleadings” to determine whether common issues predominate, which “extends to the resolution of expert disputes 'concerning the import of evidence.” Blades v. Monsanto Co., 400 F.3d 562, 566, 575 (8th Cir.2005). “The class determination generally involves considerations that -are enmeshed in the factual and legal issues comprising the plaintiffs cause of action.” Wal-Mart, 131 S.Ct. at 2551-52 (quotation and alteration omitted). Here, defendants rebutted the Basic presumption by submitting direct evidence (the opinions of both parties’ experts) that severed any link between the alleged conference call misrepresentations and the stock price at which plaintiffs purchased. As plaintiffs presented no contrary evidence of price impact, they failed to satisfy the predominance requirement of Rule 23(b)(3), and the district court abused its discretion by certifying the class.
IV. Conclusion.
The district court’s Order dated August 6, 2014, is reversed and the case is remanded for further proceedings not inconsistent with this' opinion.
. “The Safe Harbor provision, 15 U.S.C. § 78u-5(c), immunizes from [Rule 10b — 5] liability any forward-looking statement, provided that; the statement is identified as such and accompanied by meaningful cautionary language; or is immaterial: or the plaintiff fails to show the statement was made with actual knowledge of its falsehood." Institutional Inv’rs Grp. v. Avaya, Inc., 564 F.3d 242, 254 (3d Cir.2009).
. As the court said in Gissin v. Endres, 739 F.Supp.2d 488, 506 (S.D.N.Y.2010), “Defendants’ statement was made.during an earnings conference call discussing the financial results for the preceding quarter____ Defendants were not making guarantees about the present: they were stating their educated guess about what the preceding quarter's financial -data would mean for the Company’s future.”
. The other often-litigated prerequisites to Rule 23 certification are hot at issue. See generally Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, 131 S.Ct. 2541, 2548-51, 180 L.Ed.2d 374 (2011),
. Defendants define an event study as "a regression analysis which measures the effect of a publicly reported event on a company's stock price.”