Carol Layfield as guardian of Michael Layfield appeals the grant of summary judgment to the Department of Transportation (DOT) and to Everett Dykes Grassing Company (Dykes) on her complaint alleging negligence in the construction and resurfacing of a road where Michael lost control of his vehicle in a rainstorm and crashed into a tree. Because Michael could not remember the events associated with the accident, and because his expert’s testimony as to causation was based on speculation, no competent evidence supported the allegation that the water depth on the road caused the accident. Absent evidence of proximate cause, we must affirm.

“Summary judgment is proper when there is no genuine issue of material fact and the movant is entitled to judgment as a matter of law. OCGA § 9-11-56 (c).” Matjoulis v. Integon Gen. Ins. Corp.1 We review the grant of summary judgment de novo, construing the evidence in favor of the nonmovant. Id.

So construed, the evidence shows that on his way home from work in August 2001, Michael was driving through a heavy rainstorm that, according to his boss who was a few minutes behind him on the road, made it very difficult to see. After a mile or so on the road, the boss found an unconscious Michael in his vehicle off the left side of the road and crashed into a tree. Michael has no recollection of driving his *807vehicle or of how the crash occurred, and no witnesses observed the accident. The officer investigating the accident determined that the tread on three of Michael’s tires was illegally low.

An attorney investigating the accident testified that a few days after the wreck, he observed the road near the wreck during a rainstorm and determined that an inch and a half of water was running along the top of the road for some distance before draining off to the side. He took pictures. Evidence also showed that two weeks prior to the accident, another driver during a heavy rainstorm had lost control of his trailer near the same stretch of road and had ended up crashing into the same set of trees. The driver and the emergency medical technician responding to this prior accident both testified that the road near the prior wreck was retaining water that was puddling up and running parallel with the road and then off to the side.

Alleging that the DOT had failed to construct the road with sufficient slope to drain water and that Dykes’s recent uneven microsurfacing of this road had exacerbated the road’s retention of water, Carol Layfield as guardian for Michael sued the DOT and Dykes for negligence. Layfield presented the testimony of an expert who opined that the DOT had not originally constructed the road to proper sloping specifications in 1949 and that Dykes’s recent micro-surfacing of the road in June 2001 contained irregularities that exacerbated the water accumulation problem. The expert further opined that Michael’s loss of control of his vehicle was caused by encountering the accumulated water on the road.

The expert’s deposition revealed, however, that the only bases for his opinion that the accumulated water caused Michael’s accident were (1) the presence of accumulated water on the road at the time of the accident, and (2) the prior accident occurring during a rainstorm on the same stretch of road that ended up with that vehicle losing control due to accumulated water and hitting the same set of trees. Although capable of doing so, the expert did not do a reconstruction of Michael’s accident. Thus, he made no effort to determine (a) the depth of the water at the time of the accident, (b) how long it had been raining, (c) how hard it was raining, (d) whether the water was moving, (e) Michael’s speed at the time he lost control (even though the expert could have inspected the damage to the vehicle to calculate this), (f) whether Michael’s speed along this road was uniform, (g) what activities Michael was engaging in at the time he lost control, (h) whether Michael had driven this road before in the rain, (i) where along a 300-foot stretch of road Michael lost control of his vehicle, (j) whether an animal darted from the woods, (k) whether the illegally low tread on three of Michael’s tires played a role, or (1) whether the accumulated water caused Michael’s vehicle to hydroplane, to have *808differential friction, to have dynamic drag, or to simply impact the water. Indeed, he conceded the possibilities that Michael may have lost control of his vehicle because an animal such as a deer darted into the road, because Michael was distracted by the radio or other activities in the vehicle, because Michael’s three tires were so worn down that they may have mounted the water, because Michael fell asleep, or because Michael was driving too fast.

The expert also failed to make any such determinations with regard to the prior accident (on which he repeatedly relied in concluding that the accumulated water must have caused Michael’s accident also) to determine similarities, such as determining as to that prior accident the depth of the water, the amount of rain, the speed of the vehicle, the role of the trailer in causing the accident, the effect of deeper tire treads on the vehicle, and whether the accumulated water resulted in hydroplaning or differential friction. Indeed, the expert never even spoke to the driver involved in the prior accident and could only identify two similarities between the two accidents: it was raining and both vehicles went off the left side of the roadway into the same set of trees.

Because of the speculative basis of the expert’s conclusions, the trial court granted summary judgment to the DOT and Dykes on the ground that no competent evidence showed a causal link between Michael’s losing control of his vehicle and the presence of the accumulated water on the roadway. We agree and affirm.

“Causation is an essential element in any negligence action.” (Footnote omitted.) Hobday v. Galardi.2 The plaintiff has the burden of proof on this element, and when challenged on summary judgment, the plaintiff must point to specific evidence giving rise to a reasonable conclusion that it is more likely than not that the conduct of the defendant was a cause in fact of the result. Head v. Sears Roebuck & Co.3 “Amere possibility of such causation is not enough; and when the matter remains one of pure speculation or conjecture, or the probabilities are at best evenly balanced, it becomes the duty of the court to grant summary judgment for the defendant.” (Punctuation omitted.) Id.

Here, the tenuous threads on which the expert hangs his conclusion that the accumulated water on the road caused Michael’s accident are that accumulated water was present on the road and that another accident occurred on this same stretch of road two weeks earlier that resulted in a collision with the same set of trees. Since the expert made no effort to quantify the amount of water on the road or *809to compare the circumstances of the two accidents, this was nothing more than an impermissible “res ipsa loquitur” approach: a short-circuited analysis requiring quantum leaps of faith and logic and smacking of rank speculation to link the two accidents together.

Indeed, the expert’s approach ignored stark, undisputed differences between the two accidents. First, in the prior accident, the vehicle was pulling a fully-loaded 1,500 pound trailer that swerved around and hit the pulling vehicle (which was not sliding), causing the vehicle to go “haywire” and into the ditch. Michael was pulling no trailer. Second, the vehicle in the prior accident had good tread on its tires, whereas Michael’s vehicle had illegally low tread on three of its tires. Indeed, in the week prior to the accident, Michael had told his parents he needed new tires. Third, the trailer in the prior accident began swerving before that area of highway where the water was accumulating.

The expert ignored these undisputed differences with no attempt to explain them away; he failed to investigate any details of Michael’s accident or of the prior accident so as to compare the two accidents; and his only explanation as to why the other admittedly possible causes for Michael’s accident (going too fast, distracted, tread too low, could not see, animal darted out, etc.) were less probable was that this prior accident — the details of which he had not investigated — had occurred there two weeks earlier. Such specious reasoning severed the tenuous threads on which the expert’s opinion hung and left the expert with rank speculation as the basis for his opinion on proximate cause.

We have long rejected such expert testimony as competent. Bankers Health &c. Ins. Co. v. Fryhofer4 held:

While it is true that an expert may give his opinion without stating the facts upon which it is based, yet when the basis of his opinion is given and it appears that it is wholly speculative or conjectural, it must follow that his opinion is without foundation and has no probative value. Speculation and conjecture by an expert is still speculation and conjecture, and will not support a verdict.

(Emphasis supplied.) Thus, where the only evidence of the cause of an accident is expert testimony based on conjecture and speculation, summary judgment in favor of the defendant is mandated. Denson Heating &c. Co. v. Oglesby.5

*810The present case closely approximates the facts and circumstances set forth in Johnson v. Dept. of Transp.,6 in which we affirmed summary judgment in favor of the DOT. The plaintiffs alleged that ruts and water on the roadway caused their car to hydroplane, resulting in a one-car collision. Id. But neither plaintiff could recall anything about the incident. Id. at 840. The only evidence concerning causation was testimony that other accidents had occurred in the same general location, at least one of which involved hydroplaning. Id. We held that summary judgment to DOT was mandated, explaining:

None of these witnesses stated the necessary facts or had the required expertise to determine the manner in which the incident occurred, its cause, or whether that cause was attributable to a breach of duty on the part of the DOT. As the trial court observed:

While there may be evidence showing that there were ruts in the roadway of State route 32, there is no evidence showing that the vehicle in fact hydroplaned, that it hydroplaned at the place in the road where these ruts were, or that, at the time the vehicle went out of control, there was water standing on the road in these ruts. Consequently, there is simply no evidence upon which a jury could draw an inference that the vehicle went out of control as a result of hydroplaning due to standing water at a point some distance from where the vehicle came to rest.

The only testimony regarding the incident itself is that several witnesses observed ruts and water, some time after the incident, near where the vehicle left the roadway. As the trial court also observed, “whether there were ruts in the road at the site where the vehicle left the road is not material. The material inquiry is whether there were ruts in the road at the point where the vehicle first went out of control.” But no witness has ever established a precise location for any element of this incident or the position of the vehicle at any time, whether by photographs, diagrams, maps, measurements, or distance from mile markers.

Id.

*811Michael’s attempts to fill in similar gaps in his evidence by submitting the testimony of an expert must fail, since that expert failed to bother gathering any of the essential information necessary to support his conclusion that the water in the roadway caused Michael’s vehicle to lose control. See Drawdy v. Dept. of Transp.7 (summary judgment affirmed where expert witness’s opinion that water in roadway caused plaintiffs vehicle to hydroplane was based on speculation). Absent competent evidence of proximate cause, we must affirm the trial court’s grant of summary judgment to the DOT and Dykes here.

Judgment affirmed.

dissenting.

Because I cannot agree that no evidence shows that the negligence of DOT and Dykes caused Michael Layfield’s injuries and because the trial court and this court can decide this case on summary judgment only by assuming the role reserved for juries in this State, I must respectfully dissent.

Our consideration of this appeal must start from the proposition that the routine issues of negligence cases are generally not susceptible of summary adjudication, see, e.g., Robinson v. Kroger Co., 268 Ga. 735, 748 (2) (b) (493 SE2d 403) (1997), and summary judgment should not be granted in negligence cases unless the nonexistence of liability is plain, palpable, and indisputable. Ellington v. Tolar Constr. Co., 237 Ga. 235, 237 (227 SE2d 336) (1976). This is not such a case.

Further, we must not forget that when ruling on a motion for summary judgment, Layfield is entitled to the benefit of all reasonable doubt, and courts must construe the evidence and all inferences and conclusions therefrom in Layfield’s favor. Moore v. Goldome Credit Corp., 187 Ga. App. 594, 595-596 (370 SE2d 843) (1988). Any doubts about the existence of a genuine issue of material fact must be resolved against DOT and Dykes as the movants for summary judgment. Kelly v. Vargo, 261 Ga. 422, 423 (1) (405 SE2d 36) (1991). Most importantly, courts cannot resolve the facts or reconcile the issues when deciding a case on motion for summary judgment. Fletcher v. Amax, Inc., 160 Ga. App. 692, 695 (288 SE2d 49) (1981).

Here the evidence shows that even though the road was improperly constructed without sufficient superelevation for drainage, no accidents caused by accumulation of water were reported on this stretch of the road from the time it was constructed in 1949 until DOT *812and Dykes started the repairs to the road in 2001. Then, within the span of two weeks, two incidents occurred in heavy rainstorms that resulted in cars leaving the roadway at the same low point in a curve, proceeding on the same path, and hitting the same tree. After the first wreck, the driver and an EMT stated that the roadway was holding water. During the period between the wrecks, an attorney visited the site of the collisions during another heavy rain and noted that the rain was not draining from the road and that the rain was accumulating on the road surface to a depth of one and a half inches.

This evidence, coupled with the testimony of Layfield’s expert that the defective condition of the road caused Layfield’s car to hydroplane and leave the road, is sufficient to create a jury issue on whether DOT and Dykes caused Layfield’s injuries. Although the majority raises several reasons why the expert’s opinion should not be credited, judging the credibility of a witness is a role for the jury. In addition, we cannot disregard the highly persuasive evidence that whatever caused both vehicles to leave the road did so at the same place so that both vehicles left the road and struck the exact same tree.

Although Dykes argues that it should not be held liable because DOT ultimately accepted its work, the record shows that the work was not accepted until after the road in this area was resurfaced. Therefore, at the time of the accident it cannot be concluded reasonably that this particular work was acceptable to the DOT.

Also, whether the condition of Layfield’s tires or his possible inattention played some role in the accident is not properly considered on causation. Even if Layfield’s tires were defective and he was inattentive, such issues raise the defense of contributory negligence. Contributory negligence is an affirmative defense on which the defendants, not Layfield, would have the burden of proof, and, in any event, under the facts in this appeal, are not capable of summary adjudication. Ellington v. Tolar Constr. Co., supra, 237 Ga. at 238.

Although, as the majority points out, Layfield’s expert perhaps could have done a more thorough job and answered the many questions the majority poses, one must ask what the purpose of all this analysis would be.8 Ultimately, the burden on Layfield will be to prove her case by a preponderance of the evidence, not to prove it beyond a reasonable doubt.9 Thus, the question before us is whether *813the evidence presented was sufficient to show that it was more likely than not that the negligence of DOT and Dykes caused the wreck. Given the absence of accidents from 1949, the witnesses’ testimony that water was accumulating on the roadway, the expert’s testimony, the fact that both cars left the road at the same low point where water was accumulating, and the fact that both cars proceeded on the same trajectory until they hit the same tree, I am satisfied that Layfield met her burden. This evidence establishes more than the mere possibility that the water on the road caused the wreck. Accordingly, I must respectfully dissent.

I am authorized to state that Judge Adams joins in this dissent.