The Municipality of Metropolitan Seattle (Metro) appeals from a judgment in favor of Marguerite Davidson for injuries she sustained in a fall while riding a Metro bus. Metro assigns error to the trial court's admission of expert testimony by Dieter Jahns and Andrew Tooke regarding Metro's alleged negligence. We reverse and remand for a new trial.

The accident in question occurred on the morning of August 13, 1980, near the intersection of Fourth Avenue and Marion Street in downtown Seattle. Mrs. Davidson, *570who was 73 years old at the time of the accident, was riding with a companion on a Metro bus traveling north on Fourth Avenue in the right-hand lane. When the driver announced that Fourth and Marion was the next stop, the two ladies arose while the bus was still moving and made their way to the back door, where Mrs. Davidson grasped a vertical pole near the door with both hands. The bus was still traveling at approximately 10 to 15 m.p.h. when a small car passed by on the left and swerved suddenly into its path. There was conflicting testimony at trial whether the car did or did not stop momentarily in front of the bus before proceeding up Fourth Avenue. In any event, the bus driver swerved and hit his brakes, bringing the bus to a sudden stop. Mrs. Davidson lost her grip on the pole, and was thrown into the aisle, causing her severe injury.

Mrs. Davidson advanced two theories regarding Metro's liability. First, she argued that Metro should have posted signs on the bus warning passengers about the unusual forces related to sudden stops. She retained a "human factors" expert, Dieter Jahns, to testify regarding that theory. Metro deposed Jahns approximately 3 weeks prior to trial, and, at that time, he proposed that

signs be placed in the back of the seats to indicate that if you are frail or weak, remain seated until the bus has come to a complete stop at your destination.

Jahns admitted at his deposition that he was unaware of any metropolitan transit system that had followed this suggestion. He also acknowledged that this was a concept that needed to be tested and evaluated for effectiveness and that it was "just a proposal, more or less." Nevertheless, he did state that

[t]he prevention of this type accident I can guarantee is possible through proper signing and instructions to the passengers.

Jahns was unable to testify at trial, however, and his deposition was read into the record over Metro's general objection. Afterward, Metro renewed its objection in a motion to strike, arguing that all of Jahns' testimony was *571"prospective", and therefore unrelated to the issue of negligence in this particular case. The trial judge denied Metro's motion, stating that it went to the weight rather than the admissibility of the testimony.

Second, Mrs. Davidson argued that Metro's bus driver was negligent. She called Andrew Tooke, an accident re-constructionist, to testify in that regard, again over Metro's objection. Tooke stated that the driver's response to the emergency situation—"violent steering" and "slamming on the brakes"—was inappropriate. He concluded that the driver was

subjecting [his passengers] to a more violent action than if he had possibly been involved in a minor side swiping type accident with [the] smaller vehicle.

The jury returned a general verdict for Mrs. Davidson of $175,000. Metro made alternative motions for judgment n.o.v. or a new trial, both of which were denied.

The deposition testimony of Dieter Jahns is problematic, and it would be instructive to begin our analysis by carefully outlining the proper basis for the admission of expert testimony.

Admissibility of Expert Testimony

The rule governing the admissibility of expert testimony is ER 702.1 Once the court is satisfied with a witness' expertise, the test for admissibility is whether the testimony "will assist the trier of fact to understand the evidence or to determine a fact in issue". ER 702; 5A K. Tegland, Wash. Prac. § 291 (1982); State v. Petrich, 101 Wn.2d 566, 575, 683 P.2d 173 (1984). The court should also consider whether the issue is of such a nature that an expert could express "a reasonable probability rather than mere conjecture or speculation." 5A K. Tegland, at 36. In *572addition, when ruling on somewhat speculative testimony, the court should keep in mind the danger that the jury may be overly impressed with a witness possessing the aura of an expert. United States v. Fosher, 590 F.2d 381 (1st Cir. 1979).

Nevertheless, the admissibility of such evidence is largely within the discretion of the trial court, and should not be disturbed on appeal absent a showing of abuse. Harris v. Groth, 31 Wn. App. 876, 879, 645 P.2d 1104 (1982), aff'd, 99 Wn.2d 438, 633 P.2d 113 (1983). Abuse occurs only where discretion is exercised on untenable grounds or for untenable reasons. State ex rel. Carroll v. Junker, 79 Wn.2d 12, 482 P.2d 775 (1971). Moreover, the trial court's decision is given particular deference where there are fair arguments to be made both for and against admission. In re Bennett, 24 Wn. App. 398, 404, 606 P.2d 1308 (1979). As this court has stated, " [i]f the reasons for admitting or excluding the opinion evidence are both fairly debatable, the trial court's exercise of discretion will not be reversed on appeal." (Italics ours.) Levea v. G.A. Gray Corp., 17 Wn. App. 214, 220-21, 562 P.2d 1276 (1977).

In the case at bar, it is difficult to see a clear nexus between Jahns' expert testimony and the accident that occurred. Indeed, much of Jahns1 testimony is so speculative that, standing alone, it would be inadmissible. For example, he made a statement to the effect that his ideas are merely concepts which need to be tested and evaluated. However, although this court does not wish to condone the admission of expert testimony which is shot through with speculation, Jahns' testimony must be looked at as a whole. When this is done, we cannot say that Jahns' entire testimony was inadmissible. In other portions of his testimony, he says quite unequivocally that, despite the need for research, in his expert opinion, warning signs would prevent this type of accident.

In summary, although portions of Jahns' testimony were inadmissible and subject to a motion to strike, Metro's objections addressed his entire testimony, and consequently *573preserved for appeal only the failure to strike the testimony in its entirety. 5 K. Tegland, Wash. Prac. § 10, at 24 (2d ed. 1982); also cf. Spinelli v. Economy Stations, Inc., 71 Wn.2d 503, 429 P.2d 240 (1967) (questions could not be raised on appeal where the defendant objected to the factual content of plaintiff's hypothetical question but did not advise the trial court of which facts the plaintiff had improperly included or omitted).

We must also determine, however, whether Jahns' testimony was relevant evidence. ER 402 (" [ejvidence which is not relevant is not admissible"). ER 401 defines "relevant evidence" as:

evidence having 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.

This definition encompasses two elements—probative value and materiality. 5 K. Tegland § 82. To be relevant, therefore, evidence must (1) tend to prove or disprove the existence of a fact, and (2) that fact must be of consequence to the outcome of the case.

Addressing the question of probative value first, ER 401 defines "relevant evidence” broadly as "evidence having any tendency to make the existence of any fact . . . more probable or less probable". (Italics ours.) Thus, it has been said that " [mjinimal logical relevancy is all that is required." 5 K. Tegland § 83, at 170. The relevancy of evidence will depend, therefore, upon the circumstances of each individual case and the relationship of such facts to the ultimate issue. Chase v. Beard, 55 Wn.2d 58, 346 P.2d 315 (1959).

With regard to materiality, ER 401 refers to relevant evidence as tending to prove or disprove "any fact that is of consequence to the determination of the action". (Italics ours.) This definition would include facts which offer direct or circumstantial evidence of any element of a claim or defense. 5 K. Tegland § 83.

Finally, the question of whether evidence is relevant is *574for the judge, not the jury, to decide. Hull v. Enger Constr. Co., 15 Wn. App. 511, 518, 550 P.2d 692 (1976). That determination is within the broad discretion of the trial court, and is reviewable only for manifest abuse. Roberts v. ARCO, 88 Wn.2d 887, 568 P.2d 764 (1977). In light of these standards, Jahns' testimony was both probative and material, and therefore relevant evidence.

Whether a sign on the Metro bus was necessary to acquaint Mrs. Davidson with the hazards involved in sudden stops was a question of fact. Cf. Grigsby v. Seattle, 12 Wn. App. 453, 457, 529 P.2d 1167 (1975) (question of fact whether a road sign would have reminded a driver of a hazardous curve in the road). Jahns testified that the average person does not appreciate the forces present in rapid deceleration. Therefore, his testimony would tend to prove the existence of the fact that such a sign was indeed necessary.

Whether a warning sign would have prevented this accident was also a question of fact. On this point, Jahns testified that proper warning signs would prevent accidents of this type. Although arguably "prospective" in nature, this testimony was probative in that it had a tendency to make more probable the existence of the fact that, had Mrs. Davidson been warned, this accident would not have occurred.

Thus, Jahns' testimony was also material in that it tended to establish the existence of a crucial element in Mrs. Davidson's proof of Metro's negligence—"standard of care". That is, his testimony allowed the jury to determine that Metro breached its standard of care by failing to post warning signs on the bus, if the jury found as a fact that (1) a warning sign would have apprised Mrs. Davidson of the danger, and that (2) she would have heeded that warning, remained in her seat and avoided injury when the bus stopped suddenly. Cf. Grigsby v. Seattle, supra (trial court abused its discretion by excluding expert testimony that City was negligent in failing to properly post highway warning signs).

*575Trial courts have broad discretion in determining the admissibility of expert testimony. The admission of Jahns' testimony was not an abuse of that discretion.

Addressing the testimony of Andrew Tooke next, we hold that there is inadequate evidence in the record upon which Tooke could base his expert opinion. Although Metro argues that Tooke's testimony was speculative, the main thrust of the appellant's argument is essentially an attack upon the sufficiency of the foundation facts which support his expert opinion.

Basis for Expert Opinion

The proper basis for expert opinions or inferences is defined by ER 703:

The facts or data in the particular case upon which an expert bases an opinion or inference may be those perceived by or made known to him at or before the hearing. If of a type reasonably relied upon by experts in the particular field in forming opinions or inferences upon the subject, the facts or data need not be admissible in evidence.

Although this rule is intended to broaden the acceptable basis for expert opinion, there is no value in an opinion where material supporting facts are not present. 5A K. Tegland, Wash. Prac. § 304 (1982). Likewise, an inference is a logical conclusion or deduction from established facts. Martin v. Insurance Co. of North Am., 1 Wn. App. 218, 221, 460 P.2d 682 (1969).

In the case at bar, Tooke reaches his expert opinion by drawing inferences from facts not in evidence or by assuming facts actually conflicting with eyewitness testimony. He uses these inferences in turn, piling one upon another, to bootstrap himself to the ultimate conclusion that the bus driver's actions were inappropriate and unnecessary. "Presumptions may not be pyramided upon presumptions, nor inference upon inference." Prentice Packing & Storage Co. v. United Pac. Ins. Co., 5 Wn.2d 144, 164, 106 P.2d 314 (1940).

Witnesses testified at trial that the driver slammed on *576his brakes for an emergency stop, causing rapid deceleration of the bus while Mrs. Davidson was standing. Tooke testified that the bus driver need not have slammed on the brakes, but rather could have avoided a collision with the automobile by applying his brakes moderately, risking at most a minor sideswiping accident. From this testimony, the jury could have concluded that the bus driver overreacted and negligently slammed on his brakes, causing Mrs. Davidson to fall.

In reaching his conclusion, Tooke attaches importance to the relative speed of the vehicles, testifying that even a moderate application of the bus' brakes would have increased the distance between it and the faster-moving automobile, hence avoiding a collision. We know from the trial testimony that the bus was traveling at approximately 10 to 15 m.p.h. To determine the degree of braking that was necessary to avoid the collision, it is also essential to know the speed of the other vehicle and how close it came when it swerved into the path of the bus. Obviously, if the passenger car passed by the bus at a substantially greater speed and at a relatively great distance, slamming on the brakes was an overreaction. However, the inverse is equally obvious. There is no evidence in the record of the automobile's actual speed. The only evidence in the record of how close the vehicles came to one another was the bus driver's interpretation that he missed the car "by an inch" and the passengers' testimony that the car came so close that the driver had to slam on his brakes to avoid colliding with it. Although he admitted that he had no physical evidence to the contrary, Tooke rejected the eyewitness accounts and assumed in supporting his opinion that the vehicles were never very close.

Tooke also assumed that the automobile converged on the bus at a relatively shallow angle, concluding that the angle of incidence between the line of travel of the passenger car and the line of travel of the bus was 12 degrees or *577less when the bus driver applied his brakes. Thus, any accident, Tooke testified, would have been of a minor sideswiping type with a less severe impact on the passengers. That conclusion is directly contrary, however, to the bus driver's accident report, indicating that at one point the car momentarily stopped in front of the bus at a 45-degree angle. Tooke conceded at trial that he did not know where the bus was located in or near the intersection when the driver applied his brakes. Yet, he was permitted to speculate that the bus was in the northern half of the intersection when that occurred, and based upon that assumption, he rejected the bus driver's account stating:

I don't feel that the car would have made a stop in front of an oncoming bus in a turning configuration, as shown [in the driver's report at], about a 45 degree angle, turning into an area that it couldn't have proceeded ahead in.

This would be a totally reasonable observation, but only if the accident occurred, as Tooke speculated, in the north half of the intersection, since the car would have been turning directly toward the northeast curb of Fourth Avenue. Again, the evidence in the record is to the contrary. The bus driver's report shows the bus south of the intersection and the automobile in front of it, as if turning on Marion. Under those circumstances, the automobile driver was in a position where he could have proceeded in an easterly direction on Marion Street, and the bus driver's description of the automobile's aborted right turn seems quite plausible. In fact, one passenger on the bus testified that the automobile actually completed the right turn in front of the bus and proceeded eastbound on Marion. Although that particular account is contradicted by other testimony, there is no evidence in the record to support Tooke's angle of incidence testimony. Thus, his opinion is bolstered only by his own unsupported speculation that the bus was in the north half of the intersection when the accident occurred. Consequently, Tooke had no evidence from which to conclude that the angle of potential impact was *578slight. Indeed, the facts in evidence indicate that a serious collision might have occurred.

In conclusion, despite the trial court's discretion in determining the admissibility of expert testimony, we hold that the expert opinion of Andrew Tooke lacked a factual basis. The admission of this testimony was error prejudicial to the defendant because the jury could have based its finding of negligence on Tooke's unsupported opinion. Therefore, we reverse the trial court and remand this case for a new trial.

Swanson, J., concurs.

(dissenting)—I am unable to join in the majority decision because I believe the testimony of Andrew Tooke was properly admitted under the rules. ER 702 provides:

If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise.

Specialized knowledge is of help in this case because the key witness, the driver of the bus, died before trial, leaving an improbable report of the accident filed with the company. In that report he stated:

I was N.B. on 4th Ave at Marion St with a green light. Just entering the intersection a blue Pinto started to make a right turn from the lane to my left I was in the curb lane. He came to a stop and changed his mind and cont. N. on 4th Ave. I had to make a very sudden stop missing the Pinto by about an inch.

How the Pinto, while overtaking the bus, could cut in front and stop, leaving the vehicles in the positions as shown on the diagram, is an important factual question. Specialized knowledge would assist the trier of fact to understand the evidence and determine what occurred. The accident reconstructionist, Tooke, presented impressive *579credentials of his qualifications as an expert in analyzing how accidents happen. It was clearly within the discretion of the trial court to permit him to testify in the form of an opinion or otherwise.

[[Image here]]

ER 703 provides:

The facts or data in the particular case upon which an expert bases an opinion or inference may be those perceived by or made known to him at or before the hearing. If of a type reasonably relied upon by experts in the particular field in forming opinions or inferences upon the subject, the facts or data need not be admissible in evidence.

The basic data Tooke used in support of his opinions were the bus driver's report, interviews with three of the six witnesses in the bus, the statements of all of them, the Fire Department Medic I's record and an examination of the unchanged Fourth and Marion intersection. The information was "of a type reasonably relied upon by experts in the particular field in forming opinions or inferences upon the *580subject". ER 703. Thus, the expert's testimony was admissible even though his conclusions may differ from those the majority draws from the same facts. That is of no moment, as the important consideration is that Tooke had ample and reliable information from which to reconstruct the event. He was cross-examined vigorously, skillfully, and with great care. The admission of his opinions was not error.

I would affirm.

Review denied by Supreme Court July 8, 1986.