Do <i>Daubert</i> Motions Really Work?

In May 2016, PriceWaterhouseCoopers (PwC) published a report studying Daubert trends and outcomes for the years 2000-2015. The PwC report focused exclusively on financial experts, but used a comparable sample size of 2,014 cases.

These two studies are remarkable for the number of decisions analyzed by the researchers. A plethora of articles, commentaries, and blogs are available on nearly every aspect of Daubert practice, but virtually all of them are either more narrow in scope or draw conclusions from fewer decisions. Indeed, the authors of the Searle report assert that their 2015 study is “the most comprehensive view of Daubert practice in federal civil litigation to date.” Id. at 2. Even so, this study does not purport to include all the federal court Daubert decisions between 2003 and 2014. Four years — 2006, 20011, 2012, and 2013 — were heavily sampled, but “only a handful” of cases came from 2003, 2004, or 2014. Id. at 4, fn 17.

The study sample in PwC's report is comparably sized, but the international accounting giant also could contend for the “most comprehensive view” since the 2016 publication is a follow up to PwC's 2014 report analyzing 7,299 cases — albeit all involving financial testimony. See http://pwc.to/2cq57F3.

Before delving into the insights afforded by these large data sets, a different kind of publication deserves mention at the outset, because it is a cornerstone for almost any meaningful discussion of Daubert practice.

The Federal Judicial Center's Reference Manual for Scientific Evidence, first published in 1994 and currently in its third edition, is a treasure trove of authoritative and instructive essays by respected scientists, engineers, doctors, statisticians, accountants and other professionals. The refined and expanded Reference Manual (3d) is a first-rate and reliable treatise on the subjects most likely to be the target of a Daubert challenge. Even better, the Reference Manual is probably on the book shelf of nearly every federal district judge in the country.

(We will return to this important reference, and look at some of the mechanics of successful Daubert practice, next month in Part Two of this article.)

What Does the Data Say? Searle

Since plaintiffs bear the burden of proof in nearly every civil case, and many lawsuits require admissible opinion testimony to carry that burden, it is not surprising that the vast majority of Daubert motions are made by defendants. In the Searle study, defendants made 71% of the motions, while plaintiffs made only 29%.

In fact, almost half of the cases analyzed in the Searle report involved multiple Daubert motions. Based on an average of 2.1 motions per case, it seems likely that parties anticipating a Daubert motion against one or more opinion witnesses often elect to challenge at least one witness on the other side in an effort to keep a level playing field.

Nearly two-thirds of the motions challenged medical or technical opinions. “Medical” witnesses — defined to include physicians, psychologists, toxicologists, and allied medical professionals — made up a hefty 31% of the Daubert challenges in the Searle study. “Engineering” witnesses, including all types of technical and environmental testimony, were close behind at 24%. Accountants comprised 10% of the challenged witnesses and economists another 5%.

These data likely ring true with attorneys who routinely make motions to exclude unreliable opinions. Medical science, engineering, and financial witnesses giving opinions on damages often can be professional witnesses for hire and, unfortunately, arguably are the most likely to omit significant facts or employ questionable methods. These witnesses depend on repeat business, and may bend the rules just to stay employed.

But the burden of proof makes a difference, too. Plaintiffs usually do not have the same make-it-or-break-it motivation when challenging opinion testimony offered by defense witnesses. Daubert rulings that eliminate or greatly impair essential elements of a plaintiff's case can lead directly to summary judgment for the defendant. By contrast, a successful challenge of a defense witness usually does not result in summary judgment for the plaintiff. With more at stake, therefore, defendants can be more motivated to challenge admissibility of opinion testimony.

Tort lawsuits draw the most Daubert challenges. Given the frequency of challenges to medical and technical testimony compared to challenges to witnesses in other fields, the prevalence of tort cases is not surprising. Just over 50% of the cases analyzed in the Searle report were tort actions, even though tort claims comprise only 30% of the total civil caseload in federal court.

Similarly, challenges to accountants and business experts are most prevalent in contract cases; legal witnesses are the most prevalent witnesses in IP litigation; and, police or law enforcement witnesses are most prevalent in civil rights cases.

According to the Searle study, the success rate of Daubert motions depends in the first instance on the moving party. For defendants, the “batting average” for obtaining at least some relief as the result of a Daubert challenge is .500. The “batting average” for outright wins — i.e., a home run that grants all the relief requested — is .250. That's exactly half the win rate for partial relief.

Understandably, the authors of the Searle report use true percentages, not “batting averages.” Defendants obtain at least partial relief 50% of the time and complete relief 25% of the time. But conversion of the study's data to baseball nomenclature is a useful way to understand the meaning of success.

Setting aside the mathematics of calculating the differences between singles, triples, and strike-outs, a baseball player with a .500 batting average is unquestionably destined for the Hall of Fame. Even a player who carries a .250 average, however — especially if s/he is also a “golden glove” — can be a Most Valuable Player. If success rates in Daubert practice are assayed like baseball batting averages, we will feel better about this particular corner of the litigation field.

And plaintiffs' success rates are not significantly lower than defendants'. The average rate of a civil plaintiff obtaining some relief due to a Daubert challenge is 40%, while home runs — obtaining all the relief requested — average 18%. Interestingly, plaintiffs' 18% “home run” rate is probably close to what trial lawyers on both sides perceive to be the general success rate of Daubert motions. But the Searle report's data support a more optimistic view.

The second major success factor in Daubert motions is the type of the lawsuit. Not all causes of action are equal, and the type of lawsuit in which the witness is testifying can double — or halve — the success rate. When defendants challenge, the likelihood of at least some relief is over 50% in disputes involving contracts, torts, civil rights, and RICO. According to the Searle data, a defendant's best chance of at least partial success is in an antitrust case. But only 35 antitrust cases were analyzed, so the assurance provided by a large data set is not really present.

Plaintiffs' Daubert motions do not succeed as much as 50% of the time in any kind of lawsuit. The best chances for a partial grant of relief occur in contract and real property disputes and, to a lesser degree, in claims involving torts and labor disputes. Notably, plaintiffs obtain a full grant of relief in only 16% of tort cases.

Today, the timing of Daubert challenges is most often fixed by the court's scheduling order, and they usually must be made no later than the deadline fixed for summary judgment motions. Going back 15 or 20 years, however, when many federal courts were still experimenting with scheduling requirements we now take for granted, Daubert motions might be made before, with, or after motions for summary judgment.

Unchanged is the length of time between filing and ruling. Daubert challenges easily linger for 80 to 90 days in tort cases, for example; but the average duration can be as long as 100 to 150 days in labor, real property, and securities disputes. The Searle study identified 41 environmental cases where the “average duration” of a Daubert motion was a jaw-dropping 203 days. These intervals are useful to know when setting client expectations.

PwC's Study

Turning to PwC's latest study of Daubert motions, the authors' case selection trigger was not the court's citation to Daubert, but was instead Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999): “Our study is limited to written opinions citing Kumho Tire.” PwC, fn 2. Kumho, of course, extended the rationale developed in 1993 in Daubert to assess the admissibility of epidemiological opinions to non-scientific testimony of the kind provided by engineers — and also by accountants, economists, and appraisers.

In some ways, PwC's focus on financial witnesses furnishes a clearer picture of Daubert success rates. Financial witnesses typically are not called upon to provide a legal element crucial to establishing a cognizable cause of action. Although “damage” caused by alleged misconduct is a required ingredient for relief, opinion testimony is usually not the only way to establish damages. Numbers frequently can be put on the blackboard without the necessity of an “expert.” Accordingly, the motivations for challenging financial opinion testimony are less likely to be driven by who has the burden of proof and are more likely a function of whether of the witness used reliable methods.

PwC's report confirms this view. Year in and year out, lack of reliability is the most common reason for excluding opinion testimony on financial subjects. Analyzing 2,014 cases (all citing Kumho) between 2000 and 2015, PwC identified 896 cases in which the financial testimony was partially or completely excluded — an average exclusion rate of 44%.

Moreover, an exclusion rate of financial witnesses in the 40th percentile is common across all types of cases. Only a few percentage points separate securities litigation (47%) from product liability (48%), or intellectual property (49%) from bankruptcy (47%).

In one respect, however, PwC's analysis of financial witnesses reflects the same distribution of challenges, as between plaintiffs and defendants, seen in Searle's analysis of every other kind of expert. About twice as many Daubert challenges are made against plaintiffs' witnesses as defendants' witnesses. In 2015, 63% of the Daubert challenges were against witnesses for the plaintiffs and 37% were against witnesses for the defendants.

Notably, PwC found that, after reliability, “relevance” was the second most common reason for exclusion of financial testimony. FRE 702(a) requires that opinion testimony fit the case at hand, i.e., “help the trier of fact to understand the evidence or to determine a fact in issue.” PwC reported that, when a financial expert is excluded for lack of relevance, the proposed testimony is either not helpful to the trier of fact or is beyond the scope of the witness's competence.

An aspect of Daubert practice covered by PwC's analysis but not found in the Searle study is the rate of disposition on appeal. PwC began tracking appellate decisions only five years ago, restricting its analysis to trial court rulings only before 2011.

Between 2011 and 2015, a majority of the 64 appeals of trial court rulings on financial witnesses were affirmed. If the trial court allowed the testimony, the affirmance rate was 89%. If the testimony was partially excluded, the affirmance rate was 80%. And if the testimony was totally excluded, the affirmance rate was still a respectable 58%. Again, these percentages are useful to have when setting client expectations about the proverbial question of “chances on appeal.”

*****
John L. Tate is a Fellow of the American College of Trial Lawyers and a member of the southeastern regional law firm Stites & Harbison PLLC. He practices from the firm's office in Louisville, KY.

More than 20 years into the Daubert era, a surprising number of litigators still have doubts and disagreements about the effectiveness of motions in limine challenging the admissibility of federal court opinion testimony under Federal Rules of Evidence (FRE) 702. Among the concerns commonly expressed by the trial bar is the perception that so-called Daubert motions are a long shot at best, often not worth the time and effort.

Two recent studies shed new light on these attitudes.

In October 2015, the Searle Civil Justice Institute, part of the Law & Economics Center at George Mason School of Law, published an empirical examination, titled Timing and Disposition of Daubert Motions in Federal District Courts (hereafter “Searle”). Covering the years 2003-2014, this report examined the outcomes of over 2,127 Daubert motions from 91 federal district courts.

In May 2016, PriceWaterhouseCoopers (PwC) published a report studying Daubert trends and outcomes for the years 2000-2015. The PwC report focused exclusively on financial experts, but used a comparable sample size of 2,014 cases.

These two studies are remarkable for the number of decisions analyzed by the researchers. A plethora of articles, commentaries, and blogs are available on nearly every aspect of Daubert practice, but virtually all of them are either more narrow in scope or draw conclusions from fewer decisions. Indeed, the authors of the Searle report assert that their 2015 study is “the most comprehensive view of Daubert practice in federal civil litigation to date.” Id. at 2. Even so, this study does not purport to include all the federal court Daubert decisions between 2003 and 2014. Four years — 2006, 20011, 2012, and 2013 — were heavily sampled, but “only a handful” of cases came from 2003, 2004, or 2014. Id. at 4, fn 17.

The study sample in PwC's report is comparably sized, but the international accounting giant also could contend for the “most comprehensive view” since the 2016 publication is a follow up to PwC's 2014 report analyzing 7,299 cases — albeit all involving financial testimony. See http://pwc.to/2cq57F3.

Before delving into the insights afforded by these large data sets, a different kind of publication deserves mention at the outset, because it is a cornerstone for almost any meaningful discussion of Daubert practice.

The Federal Judicial Center's Reference Manual for Scientific Evidence, first published in 1994 and currently in its third edition, is a treasure trove of authoritative and instructive essays by respected scientists, engineers, doctors, statisticians, accountants and other professionals. The refined and expanded Reference Manual (3d) is a first-rate and reliable treatise on the subjects most likely to be the target of a Daubert challenge. Even better, the Reference Manual is probably on the book shelf of nearly every federal district judge in the country.

(We will return to this important reference, and look at some of the mechanics of successful Daubert practice, next month in Part Two of this article.)

What Does the Data Say? Searle

Since plaintiffs bear the burden of proof in nearly every civil case, and many lawsuits require admissible opinion testimony to carry that burden, it is not surprising that the vast majority of Daubert motions are made by defendants. In the Searle study, defendants made 71% of the motions, while plaintiffs made only 29%.

In fact, almost half of the cases analyzed in the Searle report involved multiple Daubert motions. Based on an average of 2.1 motions per case, it seems likely that parties anticipating a Daubert motion against one or more opinion witnesses often elect to challenge at least one witness on the other side in an effort to keep a level playing field.

Nearly two-thirds of the motions challenged medical or technical opinions. “Medical” witnesses — defined to include physicians, psychologists, toxicologists, and allied medical professionals — made up a hefty 31% of the Daubert challenges in the Searle study. “Engineering” witnesses, including all types of technical and environmental testimony, were close behind at 24%. Accountants comprised 10% of the challenged witnesses and economists another 5%.

These data likely ring true with attorneys who routinely make motions to exclude unreliable opinions. Medical science, engineering, and financial witnesses giving opinions on damages often can be professional witnesses for hire and, unfortunately, arguably are the most likely to omit significant facts or employ questionable methods. These witnesses depend on repeat business, and may bend the rules just to stay employed.

But the burden of proof makes a difference, too. Plaintiffs usually do not have the same make-it-or-break-it motivation when challenging opinion testimony offered by defense witnesses. Daubert rulings that eliminate or greatly impair essential elements of a plaintiff's case can lead directly to summary judgment for the defendant. By contrast, a successful challenge of a defense witness usually does not result in summary judgment for the plaintiff. With more at stake, therefore, defendants can be more motivated to challenge admissibility of opinion testimony.

Tort lawsuits draw the most Daubert challenges. Given the frequency of challenges to medical and technical testimony compared to challenges to witnesses in other fields, the prevalence of tort cases is not surprising. Just over 50% of the cases analyzed in the Searle report were tort actions, even though tort claims comprise only 30% of the total civil caseload in federal court.

Similarly, challenges to accountants and business experts are most prevalent in contract cases; legal witnesses are the most prevalent witnesses in IP litigation; and, police or law enforcement witnesses are most prevalent in civil rights cases.

According to the Searle study, the success rate of Daubert motions depends in the first instance on the moving party. For defendants, the “batting average” for obtaining at least some relief as the result of a Daubert challenge is .500. The “batting average” for outright wins — i.e., a home run that grants all the relief requested — is .250. That's exactly half the win rate for partial relief.

Understandably, the authors of the Searle report use true percentages, not “batting averages.” Defendants obtain at least partial relief 50% of the time and complete relief 25% of the time. But conversion of the study's data to baseball nomenclature is a useful way to understand the meaning of success.

Setting aside the mathematics of calculating the differences between singles, triples, and strike-outs, a baseball player with a .500 batting average is unquestionably destined for the Hall of Fame. Even a player who carries a .250 average, however — especially if s/he is also a “golden glove” — can be a Most Valuable Player. If success rates in Daubert practice are assayed like baseball batting averages, we will feel better about this particular corner of the litigation field.

And plaintiffs' success rates are not significantly lower than defendants'. The average rate of a civil plaintiff obtaining some relief due to a Daubert challenge is 40%, while home runs — obtaining all the relief requested — average 18%. Interestingly, plaintiffs' 18% “home run” rate is probably close to what trial lawyers on both sides perceive to be the general success rate of Daubert motions. But the Searle report's data support a more optimistic view.

The second major success factor in Daubert motions is the type of the lawsuit. Not all causes of action are equal, and the type of lawsuit in which the witness is testifying can double — or halve — the success rate. When defendants challenge, the likelihood of at least some relief is over 50% in disputes involving contracts, torts, civil rights, and RICO. According to the Searle data, a defendant's best chance of at least partial success is in an antitrust case. But only 35 antitrust cases were analyzed, so the assurance provided by a large data set is not really present.

Plaintiffs' Daubert motions do not succeed as much as 50% of the time in any kind of lawsuit. The best chances for a partial grant of relief occur in contract and real property disputes and, to a lesser degree, in claims involving torts and labor disputes. Notably, plaintiffs obtain a full grant of relief in only 16% of tort cases.

Today, the timing of Daubert challenges is most often fixed by the court's scheduling order, and they usually must be made no later than the deadline fixed for summary judgment motions. Going back 15 or 20 years, however, when many federal courts were still experimenting with scheduling requirements we now take for granted, Daubert motions might be made before, with, or after motions for summary judgment.

Unchanged is the length of time between filing and ruling. Daubert challenges easily linger for 80 to 90 days in tort cases, for example; but the average duration can be as long as 100 to 150 days in labor, real property, and securities disputes. The Searle study identified 41 environmental cases where the “average duration” of a Daubert motion was a jaw-dropping 203 days. These intervals are useful to know when setting client expectations.

PwC's Study

Turning to PwC's latest study of Daubert motions, the authors' case selection trigger was not the court's citation to Daubert , but was instead Kumho Tire Co. v. Carmichael , 526 U.S. 137 (1999): “Our study is limited to written opinions citing Kumho Tire .” PwC, fn 2. Kumho, of course, extended the rationale developed in 1993 in Daubert to assess the admissibility of epidemiological opinions to non-scientific testimony of the kind provided by engineers — and also by accountants, economists, and appraisers.

In some ways, PwC's focus on financial witnesses furnishes a clearer picture of Daubert success rates. Financial witnesses typically are not called upon to provide a legal element crucial to establishing a cognizable cause of action. Although “damage” caused by alleged misconduct is a required ingredient for relief, opinion testimony is usually not the only way to establish damages. Numbers frequently can be put on the blackboard without the necessity of an “expert.” Accordingly, the motivations for challenging financial opinion testimony are less likely to be driven by who has the burden of proof and are more likely a function of whether of the witness used reliable methods.

PwC's report confirms this view. Year in and year out, lack of reliability is the most common reason for excluding opinion testimony on financial subjects. Analyzing 2,014 cases (all citing Kumho) between 2000 and 2015, PwC identified 896 cases in which the financial testimony was partially or completely excluded — an average exclusion rate of 44%.

Moreover, an exclusion rate of financial witnesses in the 40th percentile is common across all types of cases. Only a few percentage points separate securities litigation (47%) from product liability (48%), or intellectual property (49%) from bankruptcy (47%).

In one respect, however, PwC's analysis of financial witnesses reflects the same distribution of challenges, as between plaintiffs and defendants, seen in Searle's analysis of every other kind of expert. About twice as many Daubert challenges are made against plaintiffs' witnesses as defendants' witnesses. In 2015, 63% of the Daubert challenges were against witnesses for the plaintiffs and 37% were against witnesses for the defendants.

Notably, PwC found that, after reliability, “relevance” was the second most common reason for exclusion of financial testimony. FRE 702(a) requires that opinion testimony fit the case at hand, i.e., “help the trier of fact to understand the evidence or to determine a fact in issue.” PwC reported that, when a financial expert is excluded for lack of relevance, the proposed testimony is either not helpful to the trier of fact or is beyond the scope of the witness's competence.

An aspect of Daubert practice covered by PwC's analysis but not found in the Searle study is the rate of disposition on appeal. PwC began tracking appellate decisions only five years ago, restricting its analysis to trial court rulings only before 2011.

Between 2011 and 2015, a majority of the 64 appeals of trial court rulings on financial witnesses were affirmed. If the trial court allowed the testimony, the affirmance rate was 89%. If the testimony was partially excluded, the affirmance rate was 80%. And if the testimony was totally excluded, the affirmance rate was still a respectable 58%. Again, these percentages are useful to have when setting client expectations about the proverbial question of “chances on appeal.”

*****
John L. Tate is a Fellow of the American College of Trial Lawyers and a member of the southeastern regional law firm Stites & Harbison PLLC. He practices from the firm's office in Louisville, KY.