delivered the opinion of the court:
On January 17, 1990, Dennis Damron filed a complaint against Micor Distributing, Ltd. (Micor), for an injury caused by an allegedly defective pipe wrench that was distributed by Micor. A jury rendered a verdict in favor of Micor on April 15, 1994. On appeal, Damron contends that the jury verdict was against the manifest weight of the evidence because it was based on the speculative testimony of Micor’s expert witness.
BACKGROUND
On January 17, 1990, plaintiff filed a complaint against Micor Distributing, Ltd., for an injury caused by an allegedly defective pipe wrench that was distributed by Micor. The complaint alleged that the wrench was unreasonably dangerous when it left Micor’s control and the unreasonably dangerous condition of the wrench was the proximate cause of plaintiff’s injury.
At trial, plaintiff testified that A&R Transport (A&R) hired him as a wash boy in 1985. This duty consisted of cleaning the inside and outside of bulk semitrailers and trucks. Two weeks after hiring plaintiff, A&R requested that plaintiff bring his tools to work and perform the duties of a mechanic. Plaintiff had no formal training as a mechanic; however, his mechanical experience ranged from building a car out of parts from three cars, maintaining and rebuilding *903tractor engines, and maintaining a bulldozer and laser trencher. As a mechanic for A&R, plaintiff provided general service on trailers and later advanced to service work on trucks.
On November 17, 1988, plaintiff attempted to fix a flat tire on a truck. Plaintiff removed the 10 outer lug nuts that secured the outside tire of the dual tire set on which he was working. The outer lug nuts were removed with an air impact wrench while the tractor was jacked off the ground. Several of the inner lug nuts holding the inner tire broke when the outer lug nuts were removed. This prevented the broken inner lug nuts from being removed with the impact wrench. Plaintiff was able to remove all the inner lug nuts except for one or two of the broken ones. Consequently, to remove the remaining broken inner lug nuts, plaintiff went to the parts room and obtained an 18-inch aluminum-handled pipe wrench. The pipe wrench appeared to be new because it had paint all the way around.
When plaintiff returned to the truck, he put the wrench on the nut and pulled it to make sure the nut was tight. When plaintiff pulled, the movable jaw of the wrench broke, and plaintiff fell on his buttocks. Plaintiff immediately felt pain in his back, buttocks and leg. His crew leader, Gerald Curl, came over to him and asked whether he was okay. Plaintiff then took a 20-minute break and returned to work.
Plaintiff received medical attention the next day that ultimately led to surgery in December 1988 to remove disk material from his back. Plaintiff returned to work on April 18, 1989. However, because he could not perform duties that involved a lot of heavy lifting and bending, plaintiff was only able to answer the phone and perform computer work.
Gerald Curl testified for the plaintiff. Curl has been a mechanic at A&R for approximately 12 years and is crew leader. Curl testified that occasionally the inner lug nuts of a tire would break off when a tire was being changed. He stated the usual method for removing these inner lugs would be to pull them off with a pipe wrench or to use a welder to weld the outer lug nut back onto the inner lug nut. However, he stated that using a pipe wrench was the preferred method because it was faster than welding.
Curl stated that on November 17, 1988, he was working next to plaintiff on the bay. He heard the sound of a wrench fall, and he walked around the rear of the trailer to see what happened. He noticed plaintiff sitting on the ground with air tools and the broken pipe wrench. Plaintiff sat on the ground for a few minutes and resumed working after taking a break. However, Curl stated that after plaintiff’s shift was over, plaintiff did not return to work until six *904months later on light duty in the parts department. On cross-examination, Curl stated he did not actually see the wrench break or how much force plaintiff applied to the wrench. Curl also admitted that it was possible that the wrench slipped and broke when it hit the ground.
Plaintiff also introduced the testimony of Ed Oppenlander, the shop foreman at A&R. Oppenlander testified that mechanics supplied their own routine hand tools and A&R supplied specialty and larger tools such as the 18-inch pipe wrench. He also stated that it is normal to use a pipe wrench to take off a broken lug nut. He testified that the wrench was purchased on or about September 29, 1988, and was accessible to any of the 12 mechanics at the shop. He did not know when the wrench was used or the various ways in which the mechanics used the wrench. Oppenlander stated that after the wrench broke, he put it in his office for a couple of days and then turned it over to the safety department. He also stated that cheater bars were used on occasion at A&R; however, he was not aware of any mechanic using a cheater bar on that wrench. He explained that a cheater bar is a piece of a pipe or an extension to the handle of the wrench used to get increased leverage or power at the end of the wrench.
Charles Van Breemen gave expert testimony for the plaintiff. Van Breeman has a bachelor’s degree in mechanical engineering and has been a mechanic for 30 years. His experience includes work on aircraft, motorcycles, automobiles, trucks and tractors. He also received two weeks’ training in fatigue failure and fracture analysis. Van Breemen testified that he examined the wrench under a microscope and noticed signs of crack growth or fatigue failure. Van Breemen explained that fatigue failure is a failure in a part that occurs over time from loading, releasing the load and reloading. He characterized the crack in the wrench as “low-cycle” fatigue failure because “it did not have the scars of being around for a long time.” Van Breemen examined the wrench and concluded that a cheater bar had not been applied to the wrench since there were no circular marks at the end of the wrench.
Van Breemen also noticed marks or lines in the fixed jaw of the wrench. He characterized these lines as “beach marks.” These marks show the progressive growth of the crack. According to Van Breemen, the crack first grew 6/1000 of an inch deep and increased to 26/1000 of an inch, then to 46/1000, 61/1000 and finally to 91/1000. Van Breemen stated these marks do not exist when there is a single overload failure.
Van Breemen sent the wrench to Trace Laboratories for Rockwell hardness testing. A Rockwell hardness test is a method of *905determining the hardness of a metal. Van Breemen explained that the harder the steel, the stronger it is and the higher the load that it can carry. However, the stronger the material, the more brittle it is. Therefore, the designer of a wrench must strike a balance between strength and toughness or nonbrittleness. Van Breemen stated that the Ideal hardness of a wrench would be between 38 and 42 Rockwell.
Trace Laboratories performed 12 hardness tests. There were four hardness readings on each side of the break and four away from the break. The average of the readings on one side of the break was 50, while the average of the readings on the other side of the break was 40. Trace Laboratories’ interpretation of these readings was there was no great difference between the hardness at the fracture site and away from the fracture site. However, Van Breemen stated that, in Ms opinion, the hardness reading of 50 is very brittle and unsuitable for this type of wrench. Van Breemen believed that this variation in hardness was a result of faulty heat treatment during manufacturing, which made part of the wrench brittle. The manufacturing process introduced a crack in the wrench which grew to the point that it could not withstand the load which plaintiff applied to it. Furthermore, Van Breemen stated that the load which plaintiff applied to the wrench was well below the load that the wrench, if it had been properly manufactured, could have tolerated.
Mitchell Kaplan testified as the defendant’s expert witness. Kaplan has a bachelor’s degree in metallurgy and a master of science degree in material science. He also has had extensive experience in fracture analysis. Kaplan agreed with Trace Laboratories’ analysis that there was no great difference in the hardness at the area of the break and away from the break. He stated that this variation had nothing to do with the heat-treating process. Instead, it was the different properties of steel that made some parts harder than others.
Kaplan stated that the marks he viewed under a microscope were not beach marks but crack arrest lines. Kaplan explained that while beach marks are indicative of fatigue, crack arrest lines are indicative of a crock’s change in velocity due to an overload. Kaplan stated that according to the Metals Handbook, the wrench was probably 4140 steel. The book also discussed the hardness of this steel and listed the hardness in the 46 to 48 category. Kaplan opined that the hardness of the steel in the wrench as reported by Trace Laboratories was appropriate for a movable jaw in a pipe wrench. Kaplan disputed Van Breemen’s interpretation of the depths of the cracks, stating that the 6/1000-inch crack would have no effect on the wrench’s jaw. Although Kaplan did not do any calculations, he stated *906that a load of approximately 700 pounds would be necessary to make a crack from 6/1000 to 20/1000 of an inch. This load could not be applied by hand without either jumping on the wrench or using a cheater bar.
In examining the wrench, Kaplan stated he saw evidence of abuse. He noticed distress on the teeth of the wrench, as if the wrench was used on a threaded object, and this object cut gigantic gouges into the face of the teeth. He also noticed the wrench was missing a lot of paint. Kaplan concluded the wrench looked like it was well used and abused because of the condition of the teeth on the movable jaw. In his opinion, the pipe wrench was unsuitable for use on the lug nut.
On cross-examination, Kaplan stated he assumed that the steel in the jaw was properly heat-treated because there were not any geometric inhomogeneities that would cause the crack to change directions rather suddenly. Kaplan also stated that based on looking at the fracture surface, he assumed the steel was properly fabricated. Kaplan opined that the fracture was probably caused by a single overload failure; however, there also could have been a preexisting crack, which, due to crack arrest lines, could fail at a much lower load than 700 pounds.
Kaplan testified that he did not know how many overloads the wrench experienced, who overloaded the wrench, or when the overloads occurred. He also did not know whether the crack arrest lines occurred at the same time or after several overloads. However, Kaplan testified that a lot of room in the grips, which hold the pipe on, was evidence that there was a significant load applied to the wrench.
At the end of trial, the jury found for the defendant. Plaintiff filed a motion requesting the court to enter a judgment for the plaintiff notwithstanding the verdict or, in the alternative, to grant a new trial. The court denied his motion. Plaintiff appeals.
We affirm.
OPINION
Plaintiff contends that Kaplan’s expert testimony was speculative and based on assumptions. Therefore, plaintiff argues the jury’s verdict was against the manifest weight of the evidence. Defendant argues that plaintiff has waived this issue because he failed to object at trial that Kaplan’s testimony was speculative or lacked a proper evidentiary foundation. It is true that the failure to properly object to allegedly improper testimony waives the right to raise that matter on appeal. (Hall v. National Freight, Inc. (1994), 264 Ill. App. 3d 412, *907424, 636 N.E.2d 791.) However, we believe defendant misstates the issue. The issue is not whether Kaplan’s testimony was speculative. Bather, the issue is whether the jury’s verdict was against the manifest weight of the evidence based on Kaplan’s expert testimony.
We first note that a verdict is against the weight of the evidence where it is palpably erroneous and wholly unwarranted, is clearly the result of passion or prejudice, or appears to be arbitrary, unreasonable, and not based on the evidence. (Hulman v. Evanston Hospital Corp. (1994), 259 Ill. App. 3d 133, 150, 631 N.E.2d 322, citing Ford v. City of Chicago (1985), 132 Ill. App. 3d 408, 412, 476 N.E.2d 1232.) The court should grant judgment notwithstanding the verdict only if all the evidence, viewed in the most favorable light to the nonmoving party, so overwhelmingly favors the party seeking judgment that no contrary verdict could ever stand. (Hajian v. Holy Family Hospital (1995), 273 Ill. App. 3d 932, 936, 652 N.E.2d 1132, 1135-36; Glassman v. St. Joseph Hospital (1994), 259 Ill. App. 3d 730, 741, 631 N.E.2d 1186; Thacker v. UNR Industries, Inc. (1992), 151 Ill. 2d 343, 353, 603 N.E.2d 449.) Furthermore, it is an abuse of discretion for a trial court to grant a motion for a new trial when there is sufficient evidence to support the jury’s verdict. Krawczyk v. Polinski (1994), 267 Ill. App. 3d 258, 265, 642 N.E.2d 185.
Specifically, plaintiff argues that Kaplan’s opinion that the moveable jaw was not improperly manufactured was based on an “assumption” that the jaw was made of 4140-type steel, and his “assumption” that the metal jaw was properly fabricated and heat treated. Thus, plaintiff argues that Kaplan’s opinion has no basis in fact. It is well settled that an expert’s opinion is only as valid as the bases and reasons for the opinion. When there is no factual support for an expert’s conclusions, his conclusions alone do not create a question of fact. (Gyllin v. College Craft Enterprises, Ltd. (1994), 260 Ill. App. 3d 707, 715, 633 N.E.2d 111.) Put simply, an expert witness’ opinion cannot be based on mere conjecture and guess. Dyback v. Weber (1986), 114 Ill. 2d 232, 244-45, 500 N.E.2d 8.
Plaintiff relies on several cases to illustrate his point. First, plaintiff relies on Simers v. Bickers (1994), 260 Ill. App. 3d 406, 632 N.E.2d 219. In Simers, the plaintiff sued an optometrist when she received eye injuries from contact lenses fitted by the optometrist. The defense expert opined that plaintiffs eyes were injured from bacteria formed because plaintiff failed to clean her lenses. The court noted, however, that the expert failed to carefully examine the failure of defendant to follow the manufacturer’s guide in treating plaintiff. The court held that the notion that bacteria entered plaintiff’s eye was not supported by factual, scientific, or expert evi*908dence and the expert’s testimo^ relating to lack of proper cleaning should have been stricken.
The second case plaintiff relies on is Coffey v. Brodsky (1987), 165 Ill. App. 3d 14, 518 N.E.2d 638. In Coffey, plaintiff filed a complaint against a physician as a result of an injury allegedly sustained following an abdominal hysterectomy. Plaintiff contended that defendant negligently enclosed her ureter with a suture used to close her peritoneum. Defendant’s expert testified that an inflammatory process that thickened the peritoneum resulted in less pliability. Therefore, when defendant stitched the peritoneum, the ureter was included in the stitch. The trial court granted defendant’s motion for directed verdict. The appellate court reversed and remanded. The court stated that although there was some evidence that plaintiff may have had an old chronic inflammation, defendant’s report did not refer to thickening. Also, the court was unpersuaded by the expert’s statement that thickening was implied because adhesions were present. The court held that the expert speculated on the condition of the peritoneum and the testimony relating to a lack of pliability was unsupported by the evidence.
The third case plaintiff relies on is Gariti v. Karlin (1970), 127 Ill. App. 2d 166, 262 N.E.2d 179. In Gariti, defendant’s expert witness testified that defendant suffered from a diabetic attack, which caused his car to swerve across the center line and collide with plaintiff’s car. The court held that although there was evidence that defendant was given insulin for a diabetic condition after the accident, there was no evidence that he was diabetic before the accident. The court also held there was no evidence to support the assumption that defendant’s car swerved across the center line before the collision. The court concluded that the expert’s testimony was highly speculative and improper.
The fourth case on which plaintiff relies is Dyback v. Weber (1986), 114 Ill. 2d 232, 500 N.E.2d 8. In Dyback, the plaintiff sued reconstruction workmen for damages caused by fire to her home. Plaintiff contended that the workmen negligently left a fuel-oil heater on the premises where they were making repairs and the oil heater caused the fire in her home. At trial the expert’s testimony was based only in part on his own inspection, which referred to investigations performed by a private investigator and the fire department. The court concluded the expert’s testimony was filled with guesses of what he believed might have happened. The court held the trial court correctly ruled that plaintiff’s expert’s opinion was based on conjecture and guess.
*909Plaintiff also relies on Schultz v. Hennessy Industries, Inc. (1991), 222 Ill. App. 3d 532, 584 N.E.2d 235. In Schultz, the plaintiff brought an action against several defendants, seeking recovery for personal injuries sustained while attempting to use a remote starter switch to start an automobile. Plaintiff’s expert witness testified that he believed the design and manufacture of the switch raised the possibility that a foreign object or substance could enter the switch and short out the circuit. Yet, the expert stated he could find no physical evidence to support his belief. The court held that the expert’s testimony regarding the cause of the accident and the dangerous condition of the remote starter switch was based purely on speculation and conjecture.
Finally, plaintiff relies on Royal Elm Nursing & Convalescent Center, Inc. v. Northern Illinois Gas Co. (1988), 172 Ill. App. 3d 74, 526 N.E.2d 376. In Royal Elm, a gas customer filed a complaint against the gas company arising out of a bill issued for gas allegedly used but unmetered due to the meter’s having been tampered with. The gas company’s expert reviewed the customer’s billing history and opined that a significant drop in usage some years earlier evidenced the tampering. The appellate court noted that there could have been a number of reasons other than tampering which could have caused the drop in usage. The court held that although circumstantial evidence will generally suffice whenever an inference may reasonably be drawn from it, the existence of a fact may not be inferred when the existence of a fact inconsistent with the first can be inferred with equal certainty.
In examining the cases discussed above, it is clear that when experts fail to take into consideration their party’s actions (Simers), base their opinions on facts which are not in evidence (Gariti, Coffey, Schultz), ignore significant factors (Royal Elm), and base their opinions on what might have happened (Dyback), the court will hold that the experts’ opinions are based on mere speculation and conjecture. Plaintiff contends that the only arguably "scientific” basis for Kaplan’s opinions regarding the wrench’s manufacture lies in his interpretation of the hardness test and the crack arrest lines. Plaintiff refers to portions of Kaplan’s testimony where he states that the hardness testing is not an accurate method of testing tensile strength and Trace Laboratories’ grinding of the surface of the steel may very well have compromised the integrity of the test results. However, it was necessary for Kaplan to base his opinions on the evidence that was available to him. Although the Rockwell hardness test may not have been the best method of testing the metal’s hardness, Kaplan testified that the results gave an idea of the hardness of *910the metal and the test was an accurate representation of hardness when the readings were averaged. Although there was a big drop in hardness between one. side of the crack and the other, Kaplan stated that the average of the readings at the break and away from the break were sufficiently uniform. Also, Kaplan testified that in examining the fracture face of the wrench, the material was homogenous, there were no voids or imperfections in the material, and there was no appearance of grain growth that would be indicative of poor heat-treating techniques.
Kaplan also testified that the only way to know whether the grinding compromised the integrity of the steel was to grind it flat and then rerun the tests. Because Kaplan had to preserve the integrity of the evidence for trial, he could not cut into the wrench to perform his own tests. Furthermore, Kaplan testified that from an engineering perspective, it was important to interpret the results, taking into consideration any shortcomings of the tests.
Plaintiff also contends that Kaplan’s testimony regarding the crack arrest lines was speculative because Kaplan testified that he did not know what occurred to cause failure, how many cracks were caused, when any cracks were caused, the number of assumed overloads, when any overloads occurred, or who caused the overloads. However, Kaplan did testify that according to bending formulas, at least 700 pounds of force would be necessary to break the jaw. Kaplan stated that microscopy would be necessary to determine whether the cracks occurred at the same or different times. We note that plaintiff’s expert also did not know whether the wrench had a crack when it came from the distributor, did not know the load to be applied to cause the crack, did not know how many cycles were necessary to make the crack and did not know how much force was needed to deepen the crack to .020.
Defendant also argues that evidence of Kaplan’s "rank” speculation is illustrated in his deposition given only two weeks before the trial. At the deposition, Kaplan agreed with plaintiff that beach marks are cracks that change height as they progress through a material. However, on redirect examination, Kaplan explained that both beach marks and crack arrest lines are similar in that there is a change in height.
Also, Kaplan testified that his drawing of the crack was totally different from Van Breemen’s drawing of the same marks. In regard to the direction of the cracks, he explained that he could not categorically say that the crack was running in the direction in which Van Breemen said the crack was running, which would be indicative of a single overload of high force. Kaplan stated that greater magnifica*911tian would be necessary to determine the exact direction of the crack. He did not determine the precise direction of the cracks because to do so would again entail a destructive inspection.
Plaintiffs last contention is there was no evidence to support Kaplan’s theory that the wrench had been misused. Kaplan testified that the wrench probably failed because of a single overload. However, he accepted the fact that plaintiff did not jump on the wrench or use a cheater bar when the wrench failed. Thus, Kaplan stated "the only possibility [for the wrench’s failure] is that you have a preexisting crack of about a tenth of an inch,” which would allow the wrench to fail at a much lower load. Plaintiff argues that because there is no evidence that a crack existed before the wrench failed, Kaplan’s opinion was pure speculation. However, an expert’s opinion as to the cause of an injury is not improper or inadmissible because it is couched in terms of probabilities or possibilities based upon certain assumed facts. (McKenzie v. S.K. Hand Tool Corp. (1995), 272 Ill. App. 3d 1, 8, 650 N.E.2d 612.) An expert may testify as to "what might be” the cause of the injury despite any objection that his testimony is inconclusive or speculative. (McKenzie, 272 Ill. App. 3d at 8.) Use of the word "possibility” in conjunction with causation does not automatically render the expert’s answer inadmissible or improper. (McKenzie, 272 Ill. App. 3d at 9.) Because the expert’s testimony is but "the opinion of the witness given on facts assumed to be true,” the trier of fact still must determine the facts. (McKenzie, 272 Ill. App. 3d at 9, quoting Presswood v. Morris (1979), 70 Ill. App. 3d 513, 517, 388 N.E.2d 844.) The context in which Kaplan used the word "possibility” relates to causation, i.e., the reason for the wrench’s failure. Furthermore, Kaplan testified that upon examining the wrench, he noticed distress or gigantic gouges on the teeth of the movable and stationary jaw. He also noticed a lot of missing paint and the wrench appeared to be well used.
We note that the plaintiff’s expert relied upon the exact evidence in forming his opinion as was relied upon by the defendant’s expert witness. There only appear to be different interpretations of the evidence between Kaplan and Van Breemen. A reviewing court can neither substitute its judgment for that of the jury in determining the weight of conflicting evidence nor set aside a verdict simply because the jury could have drawn different conclusions from conflicting testimony. (Maple v. Gustafson (1992), 151 Ill. 2d 445, 452, 603 N.E.2d 508.) We hold that the jury’s verdict was not against the manifest weight of the evidence.
*912For the reasons cited herein, the judgment of the circuit court of Cook County is affirmed.
Affirmed.
GORDON and T. O’BRIEN, JJ., concur.