OPINION

RONALD LEE GILMAN, Circuit Judge.

This case arises out of a traffic accident involving a vehicle driven by Angela Dortch and a Con-Way Transportation Services, Inc. tractor-trailer driven by Loren Fowler. The accident left Fowler unhurt, but Dortch suffered permanently disabling injuries that caused her to lose all memory of the collision.

Dortch alleges that Fowler's tractor-trailer was in her lane when the accident occurred, and that Con-Way had continued to employ Fowler despite his poor driving record. She brought suit against both Fowler and Con-Way, alleging negligent causation on the part of Fowler (attributable to Con-Way under the doctrine of respondeat superior) and negligent supervision and retention on the part of Con-Way.

The district court granted summary judgment in favor of Con-Way on the negligent-supervision-and-retention claim, and a jury found for Fowler (and thus Con-Way) on the underlying negligence claim regarding the cause of the accident. For the reasons set forth below, we AFFIRM the judgment of the district court.

I. BACKGROUND

A. Factual background

The accident occurred on the curve of a narrow two-lane highway in Louisville, Kentucky. Rain that day had caused the road to be slick. Dortch was driving her sport utility vehicle (SUV), heading east. Fowler was driving a Con-Way tractor-trailer, heading west. The two vehicles collided head on. Jolted by the collision, Fowler's tractor trailer swung sharply to the left, where it crashed into a rock wall on the south side of the highway. Dortch's SUV spun around and came to a stop in the middle of the road, its front left side shattered.

Emergency personnel arrived shortly after the accident, followed roughly 35 minutes later by a team of police officers that included Louisville policeman Clarence Beauford, a trained accident reconstructionist. Officer Beauford and his team interviewed Fowler and examined the scene of the accident, taking pictures and inspecting the road, the surrounding area, and the vehicles. Beauford did not attempt a full accident reconstruction and did not personally take any measurements.

According to his testimony at trial, Officer Beauford found a deep gouge in the pavement that ran from the left front wheel of Dortch's wrecked SUV. This large gouge (the first gouge) traced backward and ended in the tractor-trailer's lane. Mainly relying on this first gouge, Office Beauford concluded that Dortch's SUV crossed into Fowler's lane, causing the accident.

Some time after the accident, an expert hired by Dortch — retired Michigan State Police Officer Thomas Bereza — examined the highway and the vehicles. He determined that the accident occurred when Fowler's tractor-trailer crossed into Dortch's lane. Bereza based this determination on his discovery of a different gouge in the highway's surface, slightly east of the crash site, which was entirely in Dortch's lane. According to Bereza, this gouge (the second gouge) was created by the accident and marked the actual point where the vehicles first collided.

Con-Way employed its own experts, including William Cloyd III, to investigate the accident. Relying on the photographs of the accident and the highway (which had subsequently been repaved), and on his inspection of the vehicles, Cloyd concluded that the second gouge was not caused by the collision between Fowler's and Dortch's vehicles, opining that the first and second gouges differed in size, depth, color, and direction. At trial, both Officer Beauford and Cloyd testified that the accident happened in Fowler's lane, as did an additional Con-Way expert witness, Frank Entwisle, as well as Fowler himself. Bereza was the lone occurrence witness called by Dortch, and his conclusion that the accident began in Dortch's lane, based in large part on the second gouge, was criticized by Cloyd and Entwisle. The jury returned a verdict in favor of both Fowler and Con-Way.

B. Procedural history

Dortch filed suit in the United States District Court for the Western District of Kentucky, and the parties consented to have Magistrate Judge James Moyer preside over the case. Jurisdiction was based on diversity of citizenship pursuant to 28 U.S.C. § 1332.

During discovery, Dortch learned that Fowler had seven prior traffic accidents while driving for Con-Way. This caused her to request further information about these prior accidents. When Con-Way refused to comply with the request, Dortch filed a motion to compel. The district court ruled that Con-Way could withhold information about those accidents that "did not involve bodily injury or occur on public roadways." As a result, Dortch received information on only two of Fowler's seven prior accidents.

Con-Way subsequently moved for summary judgment on Dortch's negligent-supervision-and-retention claim, arguing that Dortch lacked enough evidence to proceed, and characterizing Fowler's undisclosed accidents as involving nothing more than the loading and unloading of cargo. In response, Dortch cited the two accidents that Con-Way had revealed, as well as Con-Way's internal rating of Fowler's driver safety as "marginal," the lowest rating given by the company. Con-Way gave Fowler this rating four months before the accident with Dortch. Dortch again requested discovery of Fowler's undisclosed prior accidents. After rejecting this request, the court granted summary judgment to Con-Way on Dortch's negligent-supervision-and-retention claim.

Before trial, in an effort to disprove Bereza's assertion that the second gouge was created by the accident between Dortch and Fowler, Con-Way researched the history of all reported accidents in the area of the collision, hoping to find evidence of a previous; crash that had caused the second gouge. Con-Way discovered reports on a number of accidents in the area, but could find no conclusive evidence that the second gouge predated the accident. As a result, Con-Way filed a motion in limine to exclude its failure to find evidence that the second gouge was there before the accident, arguing that the negative result of its research was irrelevant. Over Dortch's objection, the district court granted the motion, preventing Dortch from cross-examining Con-Way's witnesses about its previous-crash research.

The district court, on the other hand, denied a motion in limine filed by Dortch that sought to prevent the introduction of Officer Beauford's accident report. Dortch argued that other evidence in the case showed that the facts as set forth in the report were untrustworthy and that its conclusions were unreliable. This report was admitted into evidence at trial.

The jury returned a verdict in favor of Fowler and Con-Way on Dortch's claim of negligent causation. This appeal followed.

II. ANALYSIS

A. Standard of review

This court reviews both the district court's discovery and evidentiary rulings under the abuse-of-discretion standard. United States v. Quinn, 230 F.3d 862, 866 (6th Cir. 2000) (discovery rulings); United States v. Wagner, 382 F.3d 598, 616 (6th Cir. 2004) (evidentiary rulings). When reviewing a district court's decision under this standard, we will "reverse only if we are firmly convinced of a mistake that affects substantial rights and amounts to more than harmless error." Pressman v. Franklin Nat'l Bank, 384 F.3d 182, 187 (6th Cir. 2004) (citation and internal quotation marks omitted).

We review de novo a district court's grant of summary judgment. Allen v. Highlands Hosp. Corp., 545 F.3d 387, 393 (6th Cir. 2008). Summary judgment is proper where no genuine issue of material fact exists and the moving party is entitled to judgment as a matter of law. Fed.R.Civ.P. 56(c).

B. Evidentiary rulings

1. Admissibility of research on the origin of the second gouge

The trial boiled down to whether the accident occurred in Dortch's or Fowler's lane of travel. Central to that inquiry is whether either of the two gouges (one in each lane) was caused by the underlying accident. Con-Way presented persuasive evidence that the first gouge in Fowler's lane of travel was caused by the accident, and therefore Dortch was at fault for crossing over the dividing line and causing the accident. Dortch countered with her own expert who testified that the second gouge (in her lane) was caused by the underlying accident. In an effort to discredit Dortch's expert, Con-Way undertook an extensive search of past accident records in the hopes of finding some evidence of a past accident that might have caused the second gouge. It failed to find any such evidence to corroborate its theory.

Dortch now challenges the district court's evidentiary ruling precluding her from cross-examining Con-Way's witnesses about the absence of record evidence corroborating its theory that a previous accident caused the second gouge. We conclude that the district court abused its discretion in preventing Dortch from pursuing this line of questioning because it was relevant to a central issue at trial. Nonetheless, we find that the error was harmless.

The standard for relevancy is "extremely liberal" under the Federal Rules of Evidence. See United States v. Whittington, 455 F.3d 736, 738 (6th Cir. 2006). Evidence is relevant if it has "any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence." Fed.R.Evid. 401. Just as positive evidence of a past accident that could have created the second gouge would have been admissible as tending to support Con-Way (had such evidence been found), the absence of any such evidence in the accident records makes it less likely that the second gouge predated the underlying accident in this case. Put another way, the absence of any past accident record is exactly what you would expect to find if Dortch's theory that the gouge was caused by this accident is correct. Although it may not be very strong evidence, it is certainly some evidence in Dortch's favor. She therefore should have been permitted to inquire about it under Rule 401.

Con-Way counter's that "[e]vidence of this type could hardly establish that it was more probable than not that the [second] gouge preexisted the Dortch accident or who crossed the center line first." This argument, however, mistakenly conflates the standard for evidentiary sufficiency with the Rule 401 standard for relevance. There is no doubt that Con-Way's fruitless records search, standing alone, does not make it more probable than not that either party crossed the center line first. But a piece of evidence does not need to carry a party's evidentiary burden in order to be relevant; it simply has to advance the ball. As one leading commentator has explained:

It is enough if the item could reasonably show that a fact is slightly more probable than it would appear without that evidence. Even after the probative force of the evidence is spent, the proposition for which it is offered still can seem quite improbable. Thus, the common objection that the inference for which the fact is offered "does not necessarily follow" is untenable. It poses a standard of conclusiveness that very few single items of circumstantial evidence could ever meet. A brick is not a wall.

Edward W. Cleary et al., McCormick on Evidence § 185, at 542-43.

As this court has previously noted, "the mosaic of evidence that comprises the record before a jury includes both the evidence and the lack of evidence on material matters." United States v. Poindexter, 942 F.2d 354, 360 (6th Cir. 1991) (emphasis in original). In Poindexter, a criminal defendant wanted to tell the jury that a government fingerprint expert dusted a drug container for prints but did not find any of defendant's prints. The government objected that the absence of the defendant's fingerprints on the drug container was not conclusive either way, and the trial court sustained the objection, commenting "it didn't prove that your client didn't handle them." Id. at 359. This court concluded that the trial court had abused its discretion in excluding this line of inquiry:

The legitimacy of the inference [defendant's] counsel wished to bring to the jury's attention — the absence of evidence of [defendant's] fingerprints on an article containing contraband that had been dusted for fingerprints and which [defendant] was charged with possessing — did not depend upon the conclusiveness of the inference. Neither did it depend upon the necessity that it "prove anything," merely that it had the tendency to do so. Fed.R.Evid. 401.

Id. at 360.

We note the importance of distinguishing the present case — where Con-Way searched the accident records and found nothing to corroborate its preexisting second-gouge theory — from a hypothetical case where nobody searched the accident records. If there had been no investigation and we knew nothing about the records of past accidents, then that fact would support neither party. It would be irrelevant under Rule 401. Here, we do know something about the accident records. We know they are extensive and document numerous past accidents on the roadway in question, yet reveal no evidence of a past accident that caused the second gouge. Because they do not corroborate Con-Way's theory that the gouge was preexisting, the records provide some evidence for Dortch's conclusion that the gouge was not preexisting.

Although we conclude that the trial court abused its discretion in precluding Dortch from inquiring about the absence of record evidence supporting Con-Way's theory by ruling that the evidence was irrelevant, we find that the error was harmless. We will reverse the district court only if we find that "the abuse of discretion caused more than harmless error." Tompkin v. Philip Morris USA, Inc., 362 F.3d 882, 897 (6th Cir. 2004). "Even if a mistake has been made regarding the admission or exclusion of evidence, a new trial will not be granted unless the evidence would have caused a different outcome at trial." Morales v. Am. Honda Motor Co., 151 F.3d 500, 514 (6th Cir. 1998).

In the present case, the fact that Con-Way could find no record of an incident creating the second gouge is of little probative value. The part of the highway where the accident occurred runs through the City of Louisville, a major metropolitan area, and is undoubtedly used by hundreds of thousands of vehicles a year, including large vehicles like tractor-trailers, snow plows, and construction equipment. Many, if not most, incidents that damage such well-used roadways likely go unreported.

Also, there was absolutely nothing preventing Dortch from presenting her own witness to speak to the record evidence. If the record evidence was really all that probative, Dortch could have put her own investigator on the stand to explain the lack of records supporting Con-Way's theory that the gouge was preexisting. That Dortch did not feel compelled or even think to take this approach speaks volumes about how unimportant she really believed this evidence to be. This was a multi-day trial with numerous experts and extensive testimony regarding the physical evidence. The absence of a record documenting a previous accident, while marginally relevant, has too little weight to raise any likelihood that it would have affected the jury's verdict. We therefore will not reverse the district court on this basis.

2. Admissibility of Officer Beauford's police report

Dortch next contends that the district court erred by admitting Officer Beauford's police report into evidence pursuant to the hearsay exception in Rule 803(8)(B) of the Federal Rules of Evidence. Rule 803(8) allows for the admission of public records and reports, including "matters observed pursuant to duty imposed by law as to which matters there was a duty to report . . . unless the sources of information or other circumstances indicate lack of trustworthiness."

Dortch claims that the police report was untrustworthy because Officer Beauford could have gone further in analyzing the scene of the accident, because the report mainly relies on Fowler's version of the events, and because both parties agree that some of the facts to which Officer Beauford testified contradict the report's conclusions. She also contends that the district court erred by allowing Officer Beauford to testify as an expert witness.

Fowler and Con-Way characterize the officer's report quite differently. Although both admit that the report is inconsistent with one part of Officer Beauford's testimony, they argue that the overall report is trustworthy and its facts and conclusions are accurate.

In analyzing whether Officer Beauford's report lacked trustworthiness, the district court turned to the nonexhaustive list of factors provided in the Advisory Committee Notes to Rule 803(8). These factors include: "(1) the timeliness of the investigation, (2) the special skill or experience of the official, (3) whether a hearing was held and the level at which [it was] conducted, [and] (4) possible motivation problems." Fed.R.Evid. 803(8) advisory committee's notes (internal citations omitted); see also Miller v. Field, 35 F.3d 1088, 1090-91 (6th Cir. 1994) (applying these factors to affirm the admission of police reports regarding an alleged rape); Baker v. Elcona Homes Corp., 588 F.2d 551, 558-59 (6th Cir. 1978) (applying these factors to affirm the admission of a police report regarding a traffic accident). The district court concluded that three of these four factors (all but whether a hearing was held) weighed in favor of the report's trustworthiness, and that any discrepancies could be addressed through the cross-examination of Officer Beauford.

We find no error in the district court's analysis of this issue. To begin with, Rule 803(8)(B) is directly applicable to this case. Officer Beauford clearly had a legal duty, as a police officer investigating the accident, to issue a report. The issue then becomes whether his report lacked trustworthiness. And the factors listed in the Advisory Committee Notes and used in Miller and Baker indicate that the report was trustworthy. See Baker, 588 F.2d at 558.

First, Office Beauford's team arrived at the accident scene roughly 35 minutes after the accident, making their investigation timely. Second, Officer Beauford had extensive experience and training in accident reconstruction, as even Dortch admits. But there was no hearing held regarding the report, so the third factor does not favor admission. See Miller, 35 F.3d at 1090. The fourth factor does, however, because there is no evidence that "possible motivation problems" were applicable to Officer Beauford or his team. Dortch notes, on the other hand, that Fowler provided information to Officer Beauford. But Officer Beauford was adamant that his report was largely based on his own observations at the scene. The fourth factor therefore weighs in favor of admissibility as well.

Dortch presents the additional argument that the report was untrustworthy because the conclusion in Officer Beauford's report is at odds with his testimony in his deposition and at trial. In his testimony on both occasions, Officer Beauford opined that a particular tire mark on the surface of the road was made by the right front tire of Fowler's tractor-trailer. Dortch insists that if this were the case, then Fowler was in Dortch's lane when the accident occurred, which would contradict Officer Beauford's conclusion that Dortch caused the accident. Fowler and Con-Way argue that Officer Beauford was simply mistaken on this point, that the tractor-trailer's left front tire actually made the mark, and that his police report, which does not discuss the tire mark at all, is otherwise accurate and thus admissible. But even assuming that Officer Beauford's deposition and trial testimony on this point is at odds with the report's conclusions, the district court's solution to the issue was sound: Dortch was free to attack these inconsistencies on cross-examination.

Dortch also contends that Officer Beauford should have gone further in analyzing the scene of the accident and conducted a complete accident reconstruction. But Dortch presents no support for the proposition that Officer Beauford's failure to conduct a complete accident reconstruction or a "total station survey" made his report untrustworthy. Moreover, she was permitted to cross-examine Officer Beauford about the limits of his investigation during the trial.

Dortch further asserts that the Advisory Committee Notes to Rule 803(8) state that "[p]olice reports have generally been excluded except to the extent to which they incorporate firsthand observations of the officer." According to Officer Beauford's testimony and the district court's decision, however, the portions of the police report admitted into evidence here were based on Officer Beauford's firsthand observations, not on Fowler's admittedly biased statements. Supporting this assertion is the fact that the report itself contains no statements from Fowler.

In sum, the district court's decision that the police report was trustworthy was reasonable or, at the very least, arguable, and thus not an abuse of discretion. Officer Beauford had an extensive background in accident reconstruction, was on the scene shortly after the accident, was unbiased, and his report was primarily based on his team's personal observations. Given these facts, the district court did not err in admitting Officer Beauford's report as evidence and permitting Dortch to raise any issues with the report or the officer's conduct via cross-examination. See Freitag v. Ayers, 468 F.3d 528, 540-41 n. 5 (9th Cir. 2006) (affirming the district court's decision to admit a report pursuant to Rule 803(8) and noting that the opposing party "had a fair opportunity to challenge the reliability of the report through their cross-examination"). Accordingly, admitting the report was not an abuse of discretion. See Pressman v. Franklin Nat'l Bank, 384 F.3d 182, 187 (6th Cir. 2004) (holding that the district court should be reversed for abusing its discretion "only if we are firmly convinced of a mistake that affects substantial rights" (citation and internal quotation marks omitted)); see also Merriweather v. Family Dollar Stores of Ind., Inc., 103 F.3d 576, 584 (7th Cir. 1996) ("If reasonable persons can disagree on a district court's actions, there is no abuse of discretion.").

This conclusion is reinforced by our own decision in Baker v. Elcona Homes Corp., 588 F.2d 551 (6th Cir. 1978). In that case, the parties disputed the admission, pursuant to Rule 803(8), of a police report about a traffic accident. Id. at 558-59. The police officer who conducted the investigation and authored the report arrived at the scene shortly after the accident, had investigated hundreds of previous accidents, and did not have any improper motives. Id. Furthermore, the report was based largely on the officer's own independent observations. Based on these facts, the Baker court concluded that the officer's report was admissible. Id. The facts in Baker closely track those in the present case.

In contrast, the cases relied on by Dortch are distinguishable. She cites, for example, Phillips v. Northwest Airlines Corp., 88 Fed.Appx. 862, 864 (6th Cir. 2004), which dealt with an accident where the plaintiff was injured by a baggage vehicle at an airport and sought to use a police report to prove that the driver worked for Northwest. Phillips is not instructive because the decision never makes a definitive holding on the admissibility of the police report, and the report was a collection of secondhand observations. Id. at 864-85. In the present case, Officer Beauford's report is based primarily on his team's own observations.

Dortch also cites Dallas Mavis Forwarding Co. v. Stegall, 659 F.2d 721 (6th Cir. 1981), and notes that it is "remarkably similar to this case." The Dallas Mavis court affirmed the exclusion of a police report detailing the investigation of a traffic accident. Id. at 721-22. But the report's factual findings and conclusions in that case were not based on any physical evidence, and instead related only the hearsay statements of biased witnesses. Id. at 722. Due to the total lack of "physical data or evidence," the Dallas Mavis court determined that the report failed to meet Rule 803(8)'s trustworthiness requirement. Id. This contrasts sharply with the present case where the report is founded on extensive physical evidence gathered by unbiased officers.

In addition to her argument regarding the admissibility of the police report, Dortch contends that Officer Beauford should not have been allowed to testify as an expert witness. Dortch, however, failed to preserve this argument for appeal. Before trial, Dortch filed a motion in limine to exclude Officer Beauford's opinion testimony, but the district court did not resolve the issue in its written order, instead deferring the matter until trial. During the trial, Officer Beauford testified about his extensive background in accident reconstruction and offered, without objection, an opinion about who caused the accident.

Dortch was free to renew her objection to the officer's opinion testimony by contesting Officer Beauford's qualifications at trial, but she failed to do so. "It is well settled that this court will not consider an error or issue which could have been raised below but was not." Raft v. Comm'r, 147 Fed.Appx. 458, 462 (6th Cir. 2005); United States v. Kimball, 194 Fed. Appx. 373, 376 (6th Cir. 2006) (holding that although a motion to exclude evidence "was made in limine[,] it is not properly before us because there was no contemporaneous objection made at trial to preserve it for appeal").

In sum, the district court did not abuse its discretion by admitting Officer Beauford's accident report. The court acted appropriately by admitting the report under Rule 803(8) and allowing Dortch to challenge the substance of the report via cross-examination. Dortch also waived any challenge to Office Beauford's opinion testimony.

C. Negligent-supervision-and-retention claim against Con-Way

Dortch's final two arguments affect only her negligent-supervision-and-retention claim. Specifically, she alleges that the district court wrongly limited discovery on this claim, preventing her from obtaining information about all of the previous accidents occasioned by Fowler while driving for Con-Way. She also contends that the district court incorrectly granted Con-Way summary judgment on the claim.

We have no need, however, to address these arguments. As Con-Way explains, and as Dortch concedes, a claim of negligent supervision and retention against an employer can succeed only if the conduct of the employee in question injured the plaintiff. See Airdrie Stud, Inc. v. Reed, No. 2001-CA-001397-MR, 2003 WL 22796469, at *2 (Ky.Ct.App. Nov. 26, 2003) ("An employer cannot be liable for negligent retention unless the employee committed a negligent or intentional act resulting in injury to the plaintiff."); see also Guthrie v. Conroy, 152 N.C.App. 15, 567 S.E.2d 403, 411 (2002) (holding that "[a]bsent a viable tort claim against" an employee, the plaintiff "cannot maintain an action against [the employer] for negligent retention and supervision"); Mulhern v. City of Scottsdale, 165 Ariz. 395, 799 P.2d 15, 18 (Ct.App. 1990) ("[T]he employer is not negligent in hiring or retaining the employee as a matter of law if the theory of the employee's underlying negligence fails.").

Put another way, Con-Way could be liable to Dortch for negligently supervising and retaining Fowler only if Fowler caused the traffic accident in question. But none of Dortch's arguments on appeal invalidate the jury's verdict in favor of Fowler and Con-Way on the underlying negligence claim. Because we have found no reversible error in the conduct of the trial, and because the jury found that Fowler was not negligent, the issue of whether the district court erred in addressing the negligent-supervision-and-retention claim is moot.

III. CONCLUSION

For all of the reasons set forth above, we AFFIRM the judgment of the district court.