OPINION OF THE COURT
Catterson, J.In this personal injury action arising out of an accident on the subway tracks at the Union Square station, we examine whether expert testimony as to stopping distances is sufficient to establish negligence. The jury found the defendant liable on the basis of a mathematical formula that used a purported average reaction time as a factor in calculating whether the defendant’s train operator could have stopped the train to avoid running over an intoxicated 22 year old.
The issue before this Court, therefore, is whether such a unit of time-distance measurement may be the sole basis for establishing what amounts to a standard of care in these types of cases. We find that a reaction time that is seconds or fractions of a second longer than the purported average cannot, as a matter of law, constitute the difference between reasonable and unreasonable conduct, or proof of negligence.
The plaintiff, Dustin Dibble, commenced this action after he was hit by a subway train at the Union Square station on April 23, 2006 at about 1:30 a.m. The accident resulted in, among other things, the amputation of the lower half of his right leg. It is uncontested that he was intoxicated at the time of the accident and remembers nothing of the accident itself, only that he woke up in the hospital.
The train operator, Michael Moore, died before trial, and his deposition transcript was read into evidence. At the time of the accident, Moore had worked on the N subway line for eight years. He testified that, on the night in question, as he was coming into the Union Square station, he saw a dark object at *274the beginning of the station. He stated, “It looked like garbage. . . Maybe some material left by some of the track workers.” It was dark in color and just looked like a “mass” or a “lump.” The object was to the left of the rails, almost under the platform, about a foot and a half above the road bed. He testified that he was about three car lengths away at that point, and that he slowed up. He did not stop the train, and did not want to slow up too much. Then, when he was one car length away, he “saw the debris move,” and he put the train into emergency.
When asked about what he was trained to do regarding debris on the roadbed, he responded that he was supposed to stop if it was something that would interfere with the train, like a tree or a piece of a fallen wall. He testified that there was a trip cock underneath each car of the train, and if it was hit by debris, the train was put into emergency, and automatically stopped. The operator could also put the train into emergency manually to stop it quickly.
In Moore’s career, he had put a train into emergency two or three times, including once so as to avoid hitting a dog. He had once stopped a train when he saw a man lying on the tracks bleeding, but had time on that occasion to stop normally, and did not put the train into emergency.
When asked if there was a reason he did not stop the train when he first saw the debris, he responded that, if he stopped whenever he saw debris on the tracks, he would have to stop the train every five minutes. He estimated that the time that elapsed between when he first saw the “mass” and when he stopped the train was about four seconds. He was not sure how far the train traveled after he stopped it. He could not tell if he had run over the object, but knew that he had stopped at a point past where he had first seen the debris.
After the train stopped, Moore called the control center to have the power turned off. He saw the plaintiff lying partially on the left running rail between the first and second cars. When asked if plaintiff was in the same location as he had been in before the train hit him, Moore responded that he definitely was not, that he was about a car length further into the station than when Moore had first observed the object he described variously as a mass, a lump or debris.
Moore was questioned at his deposition about a statement he made about three hours after the incident to a Transit Authority motor instructor, in which he said, “I thought I saw garbage *275on the track. Subsequently, I saw movement between the running rails and placed the train into emergency.” When asked if that statement refreshed his memory as to whether plaintiff was between the rails or to the left of the left rail, Moore responded that perhaps the instructor had not understood what he was saying. “I told him . . . that he was on the left, he was almost under the platform when I first saw him” (emphasis added).
The plaintiff called Nicholas Bellizzi, a licensed professional engineer with a transportation background. He was recognized as an expert in subway accident analysis and safety. He testified that, in his opinion, if Moore had put the train into emergency when he first saw what he described as the mass or the debris, the train could have stopped before striking plaintiff. He based his opinion on a series of calculations that produced different stopping distances utilizing Moore’s estimated speed of between 20 and 24 miles per hour at the time of the incident and Moore’s approximation of the distance between the train and the object he described as debris when he first saw it (three car lengths away). Each stopping distance comprised a braking distance and the distance traveled before the brakes were applied, that is during the reaction time which in each calculation was fixed at what Bellizzi claimed was the “average” of one second.
On cross-examination, Bellizzi conceded that Moore had not testified that he had seen a person at three car lengths away, but only that he saw something, and that he did not discern that it was a person until it moved, at which time he was only one car length away. Bellizzi acknowledged that Moore would not have been able to stop the train from that distance without hitting plaintiff. Bellizzi also acknowledged that he had never driven a train, and that his opinion relied heavily on measurements that were only estimates.
Alphus Robb, a New York City Transit Authority (NYCTA) employee since 1957, also testified on the plaintiffs behalf. He was recognized as an expert in the field of train operations. He testified that operators are taught never to pass anything unless you can identify what it is, because, “[y]ou may injury [szc] it, harm it or damage the train. [But] [i]f brakes are applied in an emergency, you may have people thrown about the train.”
On cross-examination, Robb acknowledged that he had no training as an engineer, and had two incidents of hitting people on the tracks when he was a motorman. He acknowledged that Moore had testified that the debris he had seen was in the roadbed, not on the tracks.
*276James Harris, who worked for the NYCTA and had been a train operator for 11 years, and a trainer of train operators for 14 years, from 1993 to 2007, testified for the defense. He stated that there is always a lot of debris on the roadbed especially in the station areas. He testified that operators are told that, if they see anything that would come into contact with the train or that would trip the train or impede the train from moving safely, they are to stop the train, but that if the debris is off to the side where they do not believe that it would hit the trip cock, they could operate normally through the area.
The defense also preferred the testimony of Dr. Mark Marpet, an engineer who disagreed with the plaintiffs expert’s testimony ascribing the average one-second reaction time to the train operator in this case. Marpet explained that reaction time involved three phases during which (1) an object is perceived and identified, (2) an analysis is conducted as to what should be done about it, and (3) the decision is acted upon. He opined that, in this case, Moore’s analysis could have been slowed by the fact that the plaintiff was wearing dark clothing on a dark subway roadbed. Marpet also testified that reaction time not only varies from individual to individual but that it can vary for any one individual at different times.
At the close of the evidence, both sides moved for directed verdicts. The court denied the motions. After less than one day of deliberations, the jury reached a verdict finding that both plaintiff and defendant were negligent, attributing fault 65% to defendant and 35% to plaintiff. It awarded the plaintiff $1 million for past pain and suffering, $1 million for. future pain and suffering over 50 years, and $1.5 million for future medical payments. The defendant moved to set aside the verdict as inconsistent with the weight of the evidence, and contended that the percentages were against the weight of the evidence and the amounts awarded excessive. Despite counsel’s request for time to make a written motion, the court denied the motion, and stated that there was no basis in law or fact to reduce or set aside the verdict. This was error.
Initially, we note that “[a] jury’s verdict should not be lightly overturned.” (See Pavlou v City of New York, 21 AD3d 74, 76 [1st Dept 2005], lv dismissed 5 NY3d 878 [2005].) If there is a question of fact and “it would not be utterly irrational for a jury to reach the result it has determined upon . . . the court may not conclude that the verdict is as a matter of law not supported by the evidence.” (Soto v New York City Tr. Auth., 6 *277NY3d 487, 492 [2006] [internal quotation marks and citations omitted].) In this case, the question of fact was whether Moore could have avoided hitting the plaintiff. For the reasons that follow, it is clear that the jury’s determination that the accident could have been avoided was based on nothing more than a series of estimated stopping distances that incorporated purported average reaction time. We agree with the defendant’s assertion that the plaintiff’s case was based entirely on impermissible speculation, and that the verdict was thus based on insufficient evidence, as a matter of law.
The Court of Appeals has held that “a train operator may be found negligent if he or she sees a person on the tracks ‘from such a distance and under such other circumstances as to permit him [or her], in the exercise of reasonable care, to stop before striking the person.’ ” (Soto, 6 NY3d at 493, quoting Coleman v New York City Tr. Auth., 37 NY2d 137, 140 [1975].)
In this case, as in others of its type, an expert witness testified about calculations based on speed and distance so that the jury could determine whether, under the circumstances, it was physically possible for Moore to stop the train without striking the plaintiff. As a threshold matter, however, it should be noted that none of the variables utilized by the plaintiffs expert to calculate possible stopping distances were established conclusively at trial. All were estimates or approximations. It was Moore at his deposition who estimated his speed to be between 20 and 24 miles per hour as he approached the station. His conductor stated that the train might have been traveling at 25 miles per hour. Further, it was solely Moore’s estimate that he was about three car lengths away when he first saw debris to the left, almost under the platform at the beginning of the station. Moore further stated that the cars were 75 feet in length; in fact, as Bellizzi subsequently acknowledged, the cars on the subject train were just 60 feet long. Moore was the sole witness as to what exactly was visible as the train approached the station; he was also the sole witness as to how far away he was when he saw what he described as the debris moving.
The one undisputed fact is that Dibble was found with his severed foot beside him 40 feet into the station, that is, 40 feet from the location, the beginning of the station, where Moore testified he first saw the debris. There was no evidence presented to indicate that the plaintiff was struck at the beginning of the station and then dragged for 40 feet. Indeed that scenario was roundly rejected. There was no blood evidence except in the *278location where Dibble was found, and he had no injuries consistent with being dragged or pushed by the train from the beginning of the station. This strongly suggests that the debris that Moore first saw was not, in fact, the plaintiff whom he struck 40 feet further along.*
Nevertheless, the trial proceeded with the plaintiffs expert, Bellizzi, presenting a number of scenarios, all of which showed that Moore would have avoided striking the plaintiff had he put the train into emergency when he first saw the debris, “about three car lengths away.” In the first scenario, Bellizzi posited that Moore was 265 feet away from plaintiff (three car lengths at 75 feet per car plus 40 feet in from the beginning of the station). Then, in order to calculate the stopping distance at 24 miles per hour, he testified that, the first factor to consider was the distance traveled during the “reaction” time, or as he stated, without foundation, “during that one second” before the brake is applied. At 24 miles per hour, this would have resulted in the train traveling a distance of 35.2 feet. Then, using the Transit Authority’s chart for emergency brake stopping distances, he showed that at 24 miles per hour, the train would travel another 167 feet after braking bringing the total stopping distance to 202.2 feet, and thus well within the 265 feet available before reaching the plaintiff.
The record reflects that Bellizzi next applied the same formula to the same speed, but substituted car lengths of 60 feet to conclude that Moore would have had 220 feet available before reaching the plaintiff, so he still could have stopped without hitting him. Again, in this scenario, Bellizzi used the purported average reaction time of one second.
He then applied the formula to a speed of 20 miles per hour and found that one second of reaction time would add 29.3 feet to the braking distance of 121 feet for a total stopping distance of 150.3 feet. Hence, Bellizzi testified, with 265 feet available, Moore would have stopped with 112 (sic) feet to spare. Moreover, Bellizzi opined that at this speed, the train operator could *279have stopped before hitting the plaintiff even if he had needed four seconds of reaction time (4 x 29.3). On the other hand, with 220 feet available, Bellizzi opined that Moore could have taken two seconds in reaction time and still stopped before striking the plaintiff.
Bellizzi, however, was not asked to apply, and did not apply, a four-second reaction time to his original scenario where the train was traveling at 24 miles per hour. In such scenario, Moore would have traveled approximately 141 feet (4 x 35.2) before he applied the brake, and a further 167 feet of braking distance for a total stopping distance of approximately 308 feet, whereupon he would have unavoidably hit the plaintiff.
Such a scenario, of course, makes perfectly clear that Moore’s failure to exercise reasonable care could be established only by arbitrarily imposing upon Moore the purported average reaction time of one second. In other words, in determining that the defendant’s train operator failed to exercise reasonable care because he could have stopped, the jury improperly equated negligence with possession of a motor skill that is essentially a reflex action. Moreover, in this case, the motor skill that determines the reaction time in any individual, and which is measured in seconds and fractions of a second, was assumed to be the purported average of just one second with no variability for identification, analysis and decision.
In Mirtah v New York City Tr. Auth. (48 AD3d 764 [2d Dept 2008]), the Second Department rejected the utilization of an average reaction time as a constant in a similar case. The Court rejected a report by the same expert witness, Bellizzi, which concluded that the train operator in that case could have stopped the train before striking the plaintiffs decedent. The Second Department observed that
“given the close tolerances described, a contrary inference is clearly warranted. For example, even accepting all of Bellizzi’s data, an increase in [the train operator’s] reaction time of just over one third of a second, or an increase in the speed of the train of just over one mile per hour, would result in the train still striking the decedent.” (48 AD3d at 766.)
In our view, the court simply did not go far enough. As the defendant in this case asserts, the use of an average reaction time of one second implicitly renders negligent any train operator with a longer than average reaction time.
*280More egregiously, the record does not reflect that the plaintiffs expert provided any foundation or evidentiary support for his observation that the average reaction time of a train operator is one second. Much less was it established as the average reaction time for nonnegligent train operators.
Bellizzi acknowledged that, in this case as in the cases of hundreds of other plaintiffs for whom he has testified, he uses one second for a train operator’s reaction time even though he has never seen or conducted a study of reaction times of train operators. Indeed, when asked on direct how he arrived at the one-second reaction time, Bellizzi replied:
“Well, there are many, many, many studies for automobile drivers. I myself have never seen a reaction time study for a train operator, I know of none . . . [But reaction times for automobile drivers] [t]hey’ve pretty much all come to the conclusion it’s about a second for an auto driver under normal circumstances.”
The paucity of research on train operator reaction times notwithstanding, on cross-examination, Bellizzi testified to choosing one second because “that’s a reasonable average reaction time” of train operators. He defended the choice by stating that this was not a “complex situation,” that there was only one reaction required, that is throwing the brake, and that “there [was] no reason to think that Moore had a reaction time slower than average.”
Even were we to accept arguendo that an average reaction time for a train operator is indeed one second, the necessary corollary to Bellizzi’s speculation is that there is no reason to assume that Moore’s reaction time was the purported average. On the contrary, it is self-evident that if the average reaction time is deemed to be one second for train operators, then a number of all train operators will have a reaction time of less than one second, and correspondingly a number of all train operators a reaction time of more than one second. Moreover, as Dr. Marpet testified, those in the 85th percentile will have a reaction time of 2¼ seconds.
Nothing in the record indicates where Moore might be found along that spectrum. But if, for example, Moore had been in the 85th percentile, 2¼ seconds of reaction time and car lengths of 60 feet would have resulted in the plaintiff being struck even if Moore had put the train into emergency when he first saw the debris. Further, as Marpet testified, and Bellizzi conceded, reac*281tion time also may be affected on any particular occasion by factors such as age and vision and other variables such as lighting or weather or time of day.
It is troubling that, aside from one suggestion made by Dr. Marpet that the plaintiffs dark clothing could have hampered Moore’s analysis of the situation and thus increased his reaction time, no other attempt was made to apply any of the above-mentioned factors or ranges to the train operator in this case. Had the effort been made, it would have become apparent to the jury that there was insufficient evidence to determine whether Moore could have stopped without striking the plaintiff.
For the foregoing reasons, we also reject the plaintiffs contention that expert Bellizzi merely provided scientific corroboration for Moore’s concession that he could have stopped the train before hitting the plaintiff had he put the train into emergency when he first saw the debris. Moore’s own speculation, in any event, was not an acknowledgment of negligence since it was made in the context of testimony as to Moore’s belief that what he first saw was debris and not a person.
Accordingly, the judgment of the Supreme Court, New York County (Michael D. Stallman, J.), entered April 13, 2009, after a jury trial, in favor of plaintiff in which defendant was found to be 65% liable, should be reversed, on the law, and the complaint dismissed, without costs.
Friedman, J.E, Acosta, DeGrasse and Abdus-Salaam, JJ., concur.
Judgment, Supreme Court, New York County, entered April 13, 2009, reversed, on the law, and the complaint dismissed, without costs.
Moore’s testimony that he “saw the debris move” when he was a car length away does not appear to be factually possible. Upon review of the record, it would appear more likely that Moore saw the debris to the side at the beginning of the station, and then subsequently saw another object, the plaintiff, between the running rails. This scenario would explain the apparent inconsistencies between Moore’s testimony, the undisputed fact that Dibble was found 40 feet into the station, and Moore’s statement just hours after the accident that he saw the plaintiff between the running rails. Unfortunately, there was no real inquiry as to this scenario, either at deposition or at trial.