filed a concurring opinion, joined by KEASLER and HERVEY, JJ.
In the classic injury-to-a-ehild case, the child is either dead or too young to testify, the defendant was the only person with the child at the time the injuries were received, and the defendant claims the injuries were caused in a manner that the State’s experts say is impossible. This is just such a case. I join the Court’s opinion, but I write separately to detail the flaws in the Court of Appeals’ opinion.
We explained in Cain v. State that a Court of Appeals’s opinion must take into account all of the evidence, defer to the jury’s findings, and adequately explain why the evidence in favor of the verdict is greatly outweighed by the contrary evidence.1 If it fails to do any one of these three things, we will reverse and remand the case to that court for further review.2 The Court of Appeals in this case failed to do all three things. To fully explain the Court of Appeals’ failure to abide by these requirements, it is necessary first to detail the relevant evidence.
I. THE EVIDENCE
A. Prosecution Evidence
Two young sisters were taken out of their parents’ home because of physical and sexual abuse. They were placed in foster care with Barbara Rawlings until, in August 1995, they were taken in by appellant and her husband, who at that time began adoption proceedings. The younger girl, who is the complainant in this case, was then about two and a half years old. Appellant and her husband also assumed care of the girls’ baby brother, who was born after the girls had been placed in foster care. The children were placed in the Goodmans’ home for a six-month trial period, after which the children could be legally adopted. Around December of 1995 or January of 1996, the older sister was removed from the home for allegedly sexually abusing the younger sister.3
*289On February 7, 1996, appellant brought the complainant to the hospital. Appellant appeared upset and concerned and told the hospital personnel that the complainant had fallen backwards off her tricycle and hit her head. The complainant’s heart rate had dropped significantly and her breathing was too shallow for her to maintain consciousness. Her condition, if left untreated, was life-threatening.
After the complainant was given a breathing tube and cardiac medication, the emergency room nurse (Beverly Allen) conducted a head-to-toe inspection. She found bruising on the right and left sides of the complainant’s chest, one of her shoulders, both sides of her forehead, her face (including the cheeks, an eyebrow, an eye, and under her nose), her back, her ears, the back of her head, her neck, her collarbone, and her chin. A CT scan revealed that the complainant had a small subdural hematoma — bleeding on the brain. The hematoma was in the front of the head. Medical personnel also discovered that the complainant had recent vaginal and rectal tears. Several medical experts testified that the bruises were of different ages — that at least three different stages of bruising were apparent, and in some places, there were new braises on top of old braises. According to Nurse Allen, many injuries to complainant’s face, chin, neck, and shoulder appeared to be hand marks, with visible finger and/or thumb marks present. Allen also remarked that it was difficult to tell from the photographs that these were hand bruises, but they looked like hand braises at the time the complainant was brought to the hospital. Most of the injuries suffered by the complainant were on soft areas of the body; injuries to such areas were not common for toddlers. By contrast, there were very few injuries on the bony prominences of the complainant’s body — where toddlers are ordinarily injured. There was only one small bruise on the shins, there were no bruises on the bottom, and there were no bruises or scrapes on the arms or hands.
When questioned by medical personnel, appellant claimed several incidents had occurred that might have caused the complainant’s injuries: (1) Immediately before being brought to the hospital, the complainant fell backwards off a tricycle. (2) The night before, the complainant fell forward into the bathtub and hit her forehead. Later that night, the complainant experienced nausea and vomited on her pillow. (3) The previous Sunday, the complainant had fallen down along the side of a Blazer vehicle, striking her back. (4) The complainant had also fallen off of her bed. (5) After the fall from the tricycle, appellant tried to pry the complainant’s jaw open to give her mouth-to-mouth resuscitation. (6) The vaginal and anal injuries were self-inflicted.
The attending nurse (Beverly Allen), the treating physician (Dr. Michael Mar-quardt), and the complainant’s regular physician (Dr. Arlene Meyer) all testified that the pattern of braising was inconsistent with appellant’s explanations.4 No accident described by appellant could explain the marks on the complainant’s ears or on the side of her face. Nurse Allen testified that accidental bruises do not look like the ones seen inside the complainant’s ears. While some of the back braises could have been caused by falling out of bed, in a bathtub, or from a tricycle, many of the braises on the complainant’s back were soft tissue bruises that could not be explained by ordinary falls. The described fall against the Blazer also could not ex*290plain all of the back bruises because the bruises were of varying ages. Dr. Mar-quardt and Dr. Meyer both testified that, although one could not pinpoint the exact age of a bruise from its color, one could safely determine that various bruises were in fact of different ages due to their coloration. And Dr. Meyer testified that she would expect the described fall from a Blazer to cause abrasions, but no abrasions were found on the complainant’s back.
Dr. Marquardt testified that the hema-toma could not be explained by the complainant falling backwards from the tricycle because the hematoma was in the front of the head. Dr. Meyer testified that a blow to the back of the head could not have caused a hematoma in the front of the head in a toddler, although such a result might occur in an infant. Dr. Mar-quardt also testified that he did not believe that a fall from a bed or in a bathtub would have caused the subdural hemato-ma. Dr. Meyer opined that the injury that caused the subdural hematoma occurred within a few hours of the complainant’s arrival at the emergency room. She gave two reasons for that conclusion. First, the CT scan revealed acute bleeding — showing that the hematoma occurred within the previous twenty-four hours. Although this time frame would place the hematoma within range of the bathtub incident, the hematoma was also the only explanation for the complainant’s breathing stoppage— a fact indicating that the hematoma was of a more recent origin.5
Moreover, according to Nurse Allen, some of the bruises and abrasions on the face appeared to be fairly new and worsened during the complainant’s stay in the emergency room. On the other hand, according to Dr. Marquardt, the bruises on the chin were too old to have occurred within the previous two to four hours. And Dr. Marquardt testified that the chin area would be difficult to grab and bruise (i.e. during resuscitation efforts) due to the lack of anything to grab onto.
In addition, the vaginal and anal injuries were determined to have occurred within the previous twenty-four hours. Dr. Mar-quardt testified that self-infliction of these injuries was unlikely because the injuries would be very painful. Dr. Meyer also did not believe that the vaginal and anal injuries were self-inflicted. Moreover, despite the fact that Dr. Meyer was the complainant’s regular doctor, appellant had not told Dr. Meyer in the past that the complainant was injuring herself.
Although Dr. Marquardt testified that it was difficult to determine whether injuries in general were accidental or non-accidental, he concluded that the injuries suffered by the complainant were the result of physical abuse and sexual abuse. According to Dr. Meyer, it was “very straightforward” that the complainant’s injuries were the result of abuse, and she did not think a single physician had a question about it.
Dr. Meyer testified that she asked the complainant if anyone had hurt her, and the complainant nodded yes. When Dr. Meyer asked who had hurt her, the complainant mumbled incomprehensibly. Dr. Meyer then asked whether the person who hurt her was a man or a lady, and the *291complainant replied “a lady.” During cross-examination, the defense brought up the possibility that the complainant could have been hurt during a rape exam conducted by Nurse Allen the day before. However, Dr. Meyer testified that she occasionally conducted such exams and that “most of the time it does not hurt.” The defense had earlier cross-examined Nurse Allen concerning the possibility that she was the “lady” being referred to by the complainant. Nurse Allen laughed but did acknowledge that it was possible. Nurse Allen also acknowledged that the complainant turned away during parts of the exam — which probably indicated that she was suffering some degree of pain.
Finally, Dr. Meyer testified that the complainant’s demeanor changed markedly when her former foster mother — Barbara Rawlings — arrived at the hospital. Before Rawling’s arrival, the child was “very withdrawn,” but after Rawlings’s arrival, the complainant became “much more talkative and very cuddly.”
Furthermore, in rebuttal the State called Barbara Rawlings. Rawlings had taken care of the complainant and her older sister the eleven months before they were sent to five with appellant and her husband, and the complainant had been returned to Rawlings’s care after being removed from appellant’s home. Rawlings testified that, in her care, while the complainant did occasionally receive scrapes, she did not bruise more than any other child.
B. Defense Evidence
Appellant and Dr. Gary Newsom presented controverting testimony. Dr. New-som testified that, in children the complainant’s age, subdural hematomas were caused by minor falls fifty percent of the time. Dr. Newsom opined that abnormalities in PPT and fiver tests indicated that the complainant had a bleeding disorder. Dr. Meyer had previously testified, however, that the complainant’s PPT test was only slightly higher than normal, that this variation was not significant, and that it would not explain the bruising on the complainant.
Dr. Newsom further testified that the complainant’s hematoma was not a big he-matoma, that children often fall forward due to the size of the head, and that a hematoma can occur in an area opposite of the area of trauma. An injury in the area opposite of the trauma occurred more often in small children due to lax membranes. He stated that a subdural hema-toma could be caused by a child falling in a bathtub.
Dr. Newsom testified that all of the injuries were consistent with the history given by appellant. He stated that a scrape on the chin was a common injury for children the complainant’s age. Bruising on the right clavicle could have been caused by someone lifting the complainant while she was unconscious. Bruising on the back was consistent with a fall from a Blazer. Dr. Newsom further testified that a fall from a Blazer would not cause abrasions if the child was wearing a heavy garment. He concluded that the bruises on the complainant’s back could have come from a source other than child abuse.
Dr. Newsom also talked about the general inability of experts to determine the age of an injury. He said that it was almost impossible to determine the age of the bruises on the complainant’s back. He also stated that children from abusive homes have a two to three times greater likelihood of significant and repeated injuries, due to repetitive self-injurious behavior. Dr. Newsom conceded that he could not rule out the possibility that the injuries were the result of child abuse. Nevertheless, it was his opinion that the complain*292ant was not a victim of child abuse. However, Dr. Newsom’s conclusions were based solely upon the medical records and the history given by appellant; he never personally examined the complainant.
Appellant testified that she had never been in trouble before for anything other than a speeding ticket. She said that she and her husband attended two and half months of a CPS M.A.P.P. model parenting skills class in preparation for adopting the complainant and her siblings. The class taught prospective parents how to handle children from abusive homes. Upon completing the course, parents agreed to refrain from using physical force to discipline children. “Time out” was a method of discipline favored by the class. Appellant denied using physical violence against any of her adoptive children.6 She maintained that she used “time out” as the primary method of discipline.
According to appellant, Dr. Meyer had performed checkups on the children and raved about the progress appellant was making with them. However, in her own testimony, Dr. Meyer did not recall one way or the other whether she had made such statements.
Appellant also testified that the complainant had delayed motor skills and was accidentprone. Appellant stated that the complainant would bruise immediately after any injury.
Appellant gave the following detailed history of the complainant’s injuries: On Saturday, February 3rd, the complainant fell off her bed, bumped her head on the dresser, and fell to the floor. The fall resulted in a bruise on the complainant’s forehead. On Sunday, February 4th, the complainant fell from the Blazer. She was wearing a jacket. On Tuesday, February 6th, the complainant fell in the bathtub and held her forehead afterwards. Later that night, the complainant experienced nausea and vomited on her pillow. On Wednesday, February 7th, the complainant fell backwards off her tricycle. After that fall, she began jerking as if she were suffering a seizure. Appellant tried to pry open the complainant’s jaw to give her mouth-to-mouth resuscitation and held the complainant by the shoulder. Appellant testified that she believed the bruises to the chin and shoulder were caused during this attempted resuscitation.
Appellant testified that her husband took care of the complainant on Monday, February 5th, from 8:00 a.m. to 4:00 p.m. while appellant was at work. On Tuesday and Wednesday, appellant worked from 7:45 a.m. to about 2:00 or 2:30 p.m., during which times her mother-in-law took care of the child. The bathtub fall took place on Tuesday night at 6:00 p.m. — while appellant was home. The fall off the tricycle on Wednesday also occurred after appellant had come home from work.
Appellant further testified that the complainant was happy to see her at the hospital — calling her “mom” and saying “I’m at Mommy’s work,”7 “I love you,” and “Don’t go.”
On cross-examination, appellant admitted that she was the one with the primary responsibility for dressing and bathing the complainant. Appellant asked her husband to leave this responsibility to her because she felt the girls would be more comfortable with a woman doing those *293tasks. Appellant acknowledged that, as a result, her husband was less likely to see bruises covered by the complainant’s clothing than appellant would be.8
Appellant also said that she was the only adult at home with the complainant when she fell off the tricycle. Appellant was the one to take the complainant to the hospital; appellant’s husband arrived at the hospital an hour later. Appellant opined that the fall off the tricycle was the injury that rendered the complainant unconscious. However, appellant further testified that she believed the fall in the bathtub caused the subdural hematoma, resulting in vomiting and nausea, and that falling backwards off the tricycle jarred the complainant’s brain into the front of the head, aggravating the injury. Although stating that the complainant always had a bruise on her, appellant acknowledged that it was unusual for the complainant to accumulate the number of bruises she had at the time she was taken to the hospital.
Appellant supplied an explanation for all of the bruises except the bruises on the ears and perhaps on the collarbone.9 She stated that she did not know what happened to the ears and that she did not recall “seeing any redness or anything” on the complainant’s ears. Appellant also stated that she did not pay attention to how much bruising was on the complainant’s back after the Blazer incident; so, new bruises could have occurred without her realizing it. But appellant could not state that any new bruises had appeared while the complainant was in the custody of appellant’s husband or mother-in-law.
The assistant manager at the apartment complex in which appellant lived testified that she had seen a tricycle at appellant’s apartment. The testimony at trial also indicated that no signs of abuse were found on the complainant’s younger brother.
II. ANALYSIS
A. Failure to Consider All the Evidence
The Court of Appeals’s first mistake was its failure to consider all of the evidence. An appellate court fails to conduct a proper factual sufficiency review when it fails to consider some of the evidence supporting the guilty verdict.10
1. Evidence Regarding the Subdural Hematoma
While the Court of Appeals summarized Dr. Meyer’s conclusion that the subdural hematoma occurred within a few hours before the complainant’s arrival at the emergency room, the court did not consider the supporting reasons Dr. Meyer gave for coming to that conclusion — the acuteness of the bleeding and the difficulty with breathing. These reasons were never rebutted by the defense expert and may have persuaded the jury to agree with Dr. Meyer’s conclusion. If the injury that caused the hematoma was only a few hours old, then the hematoma could not have been caused by the complainant falling out of bed (four days ago) or falling in a bathtub (one day ago). That would make the fall off the tricycle the only explanation that would fit the time frame for this particular injury.
*294And while the Court of Appeals cited Dr. Marquardt’s conclusion that the alleged fall from the tricycle would not have caused the subdural hematoma, the court left out some of his explanation. The court did not recite Dr. Marquardt’s explanation that the alleged fall from the tricycle could not have caused the hematoma because the hematoma was in the front of the head and the fall from the tricycle was to the back of the head. The Court of Appeals also failed to take into account Dr. Meyer’s testimony that a fall to the back of the head might cause a hematoma to the front of the head in an infant but not in a toddler.
2.Mechanism for the Injuries
The Court of Appeals claimed that the State never showed a mechanism for the injuries. But that court failed to recite the evidence from Nurse Allen that there were hand, finger, and thumb marks on the complainant’s face, chin, neck and shoulder. The court also did not mention Nurse Allen’s testimony that the bruises looked like hand bruises from her personal observation of the complainant but were not clearly depicted as such in the photographs. That the photographs failed to tell the whole story would be a reason for the jury to place less credence on Dr. Newsom’s testimony because Dr. Newsom never examined the complainant.
3.Common v. Uncommon Injuries
While the Court of Appeals did mention Nurse Allen’s conclusion that the complainant’s injuries were uncommon for a toddler, the court failed to mention her testimony about the absence of common injuries. The relative lack of injuries common to toddlers cuts against appellant’s testimony that the complainant was accident-prone.
4.Different Stages of Bruising
The Court of Appeals made much of the defensive evidence that it is difficult to determine the age of particular bruises. But the court made no mention of the testimony from Drs. Marquardt and Meyer that one could accurately determine that various bruises were of different ages due to their differences in coloration, even if one could not determine the precise age of an individual bruise.
5.Chin Bruising
The Court of Appeals also failed to mention Dr. Marquardt’s testimony that the chin area would be difficult to grab and bruise. If the jury believed this testimony, they could decide that appellant contrived the story about attempting to pry open the complainant’s jaw, and the jury could believe that no explanation other than abuse could explain the bruising in that location.
6.References to “Ladg” Abuser
The Court of Appeals acknowledged Dr. Meyer’s testimony that the complainant had said a “lady” hurt her. But the court minimized such testimony by citing Nurse Allen’s testimony that the conversation could have been referring to her conducting a rape exam. The Court of Appeals omitted from its opinion that Nurse Allen laughed when asked the question, and the Court of Appeals omitted Dr. Meyer’s testimony that the exams usually do not hurt.
7.Change in the Child’s Demeanor
The Court of Appeals’s opinion also failed to mention the change in the complainant’s demeanor after her original foster parent, Barbara Rawlings, arrived on the scene. The evidence of a change from “withdrawn” to “talkative and cuddly” may have helped convince the jury that the *295complainant did not feel safe with appellant because appellant was a child abuser.
8. Appellant was the only Caregiver Present at the Time of the Most Recent Injury — which Injury Resulted in Hospitalization
The Court of Appeals failed to mention that appellant was the only adult present when the complainant allegedly fell off the tricycle. It is after this fall supposedly occurred that the complainant became unconscious, suffered heartbeat and breathing problems, and was taken to the hospital.
B. Failure to Give Deference to the Juiy
Clewis also requires that the appellate court give deference to the jury’s determination of witness credibility.11 But the Court of Appeals gave uncritical credence to witnesses the jury was free to disbelieve. The court observed that appellant “flatly denied ever injuring or abusing the complainant,” that appellant had taken model parenting classes, and that appellant had testified that she used the “timeout” method of discipline favored by such classes. But this testimony from appellant means nothing if the jury believed that appellant was lying. Appellant’s credibility was a central part of the trial. The jury was called upon to determine whether the history of injuries given by appellant reflected what actually occurred or was simply a cover for child abuse. That appellant took a parenting skills class does not mean she followed its teachings.12 That .appellant denied engaging in abuse does not mean the jury had to believe her denials.
Likewise, the jury was free to disregard Dr. Newsom’s testimony where it conflicted with the State’s witnesses. Because Dr. Newsom had never examined the complainant, the jury could have reasonably believed that Drs. Meyer and Marquardt were in a better position to determine the likely causes of the complainant’s injuries. But the Court of Appeals uncritically accepted Dr. Newsom’s opinion that the complainant’s injuries were consistent with the history related by appellant, despite the testimony from three other witnesses who saw the child (Nurse Allen, Dr. Meyer, and Dr. Marquardt) that the injuries were not consistent with the history.
C. Failure to Adequately Explain Why the Evidence is Factually Insufficient
Clewis also requires an adequate explanation for why the evidence is found to be factually insufficient. A Court of Appeals must provide “a detailed explanation” of its finding, “so that this Court can ensure that the appellate court accorded the proper deference to the jury findings.”13 When an appellate court reverses on factual insufficiency grounds, it should “detail the evidence relevant to the issue in consideration and clearly state why the jury’s finding is factually insufficient ... as to be manifestly unjust; why it shocks the conscience; or clearly demonstrates bias. Further, those courts, in their opinions, should state in what regard the contrary evidence greatly outweighs the evidence in support of the verdict.”14
*296The Court of Appeals failed to fulfill the explanation requirement of Clewis in the present case. The court’s opinion simply listed evidence favorable to appellant and asserted that such evidence “greatly out-weighted] any evidence in support of the verdict.” The Court of Appeals made no attempt to explain why or how the favorable evidence outweighed the unfavorable evidence, and in fact, the court neglected even to list the unfavorable evidence in its analysis. An examination of the Court of Appeals’s purported reasons for finding the conviction to be “manifestly unjust” illustrates the lack of any attempt to undertake a critical analysis of the evidence.
1. Theories of Insufficiency
The Court of Appeals found the evidence to be factually insufficient in two respects: (1) to prove that appellant (as opposed to someone else) was the cause of the complainant’s injuries, and (2) to prove that the injuries were caused by the intentional and knowing act of another (as opposed to a series of accidents). While some of the evidence cited by the Court of Appeals is relevant to both reasons, much of the evidence is relevant to only one of the theories. The Court of Appeals sometimes cites, in support of the first reason, evidence that is relevant only to the second. To this extent, the Court of Appeals obscures the discussion by confusing these two sufficiency questions.
2. Cause of the Injuries
The Court of Appeals advanced five reasons for finding the evidence to be factually insufficient to show that appellant caused the injuries. The first reason listed was “there is no direct evidence appellant injured the complainant.” But the absence of direct evidence does not mean the State’s case is weak. Circumstantial evidence often has equal or even greater probative value than direct evidence.15 The State introduced circumstantial evidence of appellant’s guilt by showing: (1) the complainant suffered injuries indicative of abuse, (2) appellant gave explanations for the injuries that were inconsistent with the injuries observed, (3) the complainant’s condition showed that she suffered prior injuries for which appellant did not report or seek medical attention, (4) the complainant told Dr. Meyer that “a lady” hurt her, (5) the complainant was withdrawn when appellant took her to the hospital but became talkative and cuddly when her former foster mother arrived, and (6) appellant was the only adult present when the complainant suffered the injuries that occurred immediately before hospitalization.
The Court of Appeals found it significant that the State’s expert witnesses had no information on whether appellant was the perpetrator or how exactly the injuries were sustained. But that statement is not entirely correct. For example, when asked that question on cross-examination, Dr. Marquardt testified, “The information I have is that the history provided by her is inconsistent with the patterns of injury to the child.” Defense counsel then characterized that answer as relating to “circumstantial evidence,” and asked a series of questions designed to elicit whether Dr. Marquardt had any direct evidence of abuse.
It is true that the complainant did not specifically identify appellant as the perpetrator, that no one saw appellant abuse the complainant, and that there was no physical evidence that specifically linked appellant to the crime. But there was circumstantial evidence linking appellant to the abuse (items (2) through (6) above). That the evidence might have been stronger *297does not mean the evidence was insufficient.
And the fact that medical personnel could not state with certainty how injuries were sustained does not mean that they could not reasonably conclude that the injuries were a result of abuse. In fact, all three of the State’s expert witnesses concluded that abuse had occurred. The number of injuries, the placement of those injuries on the complainant’s body, the severity of her injuries, the different ages of the injuries, and the fact that the injuries were not consistent with appellant’s explanations all constituted bases for concluding that the complainant’s injuries were a result of abuse. Dr. Meyer testified that it was very straightforward that abuse had occurred, and she did not think that any physician had any question about that fact. And hand marks found by Nurse Allen on the complainant’s face, neck, chin, and shoulder indicated that at least some of the injuries were caused by a human hand using force hard enough to produce marks.
The Court of Appeals also stated that there was no evidence of prior abuse of the complainant committed by the appellant. But the court did not explain why the absence of evidence of past abuse, combined with other information, rendered the jury’s verdict manifestly unjust. Given that appellant had the child for less than six months, one could hardly fault the State for not having a record of prior abuse. Moreover, the jury could have concluded that prior abuse had in fact occurred from the proof that different stages of bruising were present on the complainant’s body.
The second reason given by the Court of Appeals was “there is evidence that appellant did not injure the complainant.” The only evidence mentioned by the court in this regard is appellant’s testimony that she took parenting classes and that she did not abuse or physically discipline the child. The Court of Appeals articulated no reason why the jury should have believed appellant’s testimony. And appellant admitted causing some of the injuries (bruising on the shoulder and underneath the chin) but claimed she did so unintentionally-
The third reason given by the Court of Appeals was that appellant provided a history of how the complainant sustained the injuries and Dr. Newsom found this history to be consistent with the injuries. The court did not explain why Dr. Newsom’s testimony outweighed the testimony of three other expert witnesses who actually treated the child. They testified that the injuries were inconsistent with appellant’s account of how they happened. This is where the difference between the theories of insufficiency (listed in part 1) becomes apparent. If appellant’s account were consistent with the injuries, one might infer that no abuse occurred. But one cannot infer from her account that someone else might have been the perpetrator. The account appellant gave explained the injuries as being accidents, or at most, self-injurious behavior on the part of the complainant.
On the other hand, if the factfinder disbelieved appellant — because her explanation was inconsistent with the physical evidence — then the factfinder could infer both the occurrence of abuse and appellant’s status as the perpetrator. The Court of Appeals held that mere disbelief of appellant could not provide substantive proof that appellant caused the injuries. But the jury had more: expert testimony that the child had been abused, as well as physical evidence indicative of intentional abuse and an explanation that appeared to be false because it was inconsistent with physical evidence. Under those circumstances, the false explanation became evi*298dence that incriminated the storyteller.16 In addition, as discussed with regard to the Court of Appeals’s first reason, the jury heard other testimony connecting appellant to the abuse including testimony that the complainant said a “lady” hurt her, that the complainant’s demeanor changed when the former foster mother arrived, and that appellant was the only adult present when complainant suffered the injury leading to hospitalization.
The Court of Appeals’s fourth reason was “there is undisputed evidence that the complainant was outside appellant’s presence during several time periods, most notably when appellant was at work.” The court found it significant that appellant’s husband and mother-in-law spent time with the child but did not testify that they did not injure the complainant. The fact is, they did not testify at all. Moreover, the only evidence arguably connecting the husband and the mother-in-law to the complainant’s injuries was the simple fact that they took care of the complainant during the day on February 4th through 7th. Appellant never claimed in her initial explanation or on the witness stand that any injuries occurred to the complainant during those time periods. And no evidence was presented at trial to indicate that either of these persons perpetrated abuse upon the complainant. On the other hand, there was circumstantial evidence (listed above) to indicate that appellant was the perpetrator. Most of the complainant’s injuries were explained by appellant as accidents that occurred when appellant was around, and in at least one instance, when appellant was the only adult present. The Court of Appeals failed to explain how the rather slender evidence regarding the husband and mother-in-law, namely their care for the child on a few occasions, “greatly outweighs” the circumstantial evidence against appellant, which included far greater periods of care for the child.
The final reason given by the Court of Appeals for concluding that appellant did not cause the complainant’s injuries was that the injuries could have been self-inflicted. In this regard, the Court of Appeals mainly discussed the complainant’s past history of sexual abuse and the likelihood that a sexually abused child will engage in repetitive, self-injurious sexual stimulation. But Drs. Marquardt and Meyer both believed that the sexual injuries were not self-inflicted. The Court of Appeals did not explain why the evidence that the injuries could have been self-inflicted greatly outweighs the evidence that they were not.
3. Intentional or Knowing Conduct
The Court of Appeals advanced six reasons for finding the evidence to be factual*299ly insufficient to show that the injuries were caused by the intentional or knowing act of another. First, the Court of Appeals stated that “there is no direct evidence of an intentional or knowing injury to the complainant.” But there was circumstantial evidence: the number and severity of bruises, the fact that many bruises were on soft tissues and other areas that toddlers do not commonly injure, the fact that the bruises were of varying ages, the demeanor of the complainant, the complainant’s statement that a “lady” hurt her, and the fact that appellant gave explanations for most of the injuries and those explanations were, according to three experts, inconsistent with the physical evidence
The Court of Appeals’s second reason was “Allen testified the injuries could have been the result of an accident.” But that statement is not entirely accurate. The Court of Appeals’s statement appears to be its interpretation of the following testimony:
Q. And once again you’re not telling this jury you know how the injuries were sustained?
A. No, sir.
Q. Whether or not they were done by a person or by accident, you don’t know how they happened; is that correct?
A. No, sir.
Aside from the ambiguity created by these negatively-phrased questions, Nurse Allen’s answers simply signify that she did not know how the injuries occurred. Contrary to the Court of Appeals’s statement, Nurse Allen did not testify that the injuries could have been the result of an accident. In fact, Nurse Allen specifically testified on direct examination that accidental bruises did not look like the marks found on the complainant’s ears. Nurse Allen also testified that the pattern of injuries was inconsistent with appellant’s explanations and that it was her opinion that the injuries were the result of child abuse.
The third reason given by the Court of Appeals was that “Marquardt testified that it was difficult to determine if the injuries were accidental or non-accidental.” Again, the court’s statement is misleading. Dr. Marquardt testified that, in general, it was difficult to determine the accidental or non-accidental nature of a child’s injury, since medical personnel would not have observed the occurrence of the injury. As a result, Dr. Marquardt explained, medical personnel would compare the injuries with the history given by the caregiver. Mar-quardt testified that he did arrive at a determination of the cause of the injuries found on the complainant. He determined that the injuries were the result of “child abuse, physical abuse, sexual abuse.”
The Court of Appeals’s fourth reason was, “Meyer testified she did not see any evidence of a mechanism, such as a hand, fist, or electrical cord that could have caused the injuries to the complainant.” The court found this to be significant “because an intentional or knowing injury could not have been inflicted without the use of some mechanism.” The court’s argument is flawed in several respects. To begin with, being unable to determine the mechanism of abuse does not mean there was no mechanism, or no abuse. A particular method of abuse might not leave a telltale sign of its origin, or signs of the origin of abuse may be erased with the passage of time. The absence of evidence of a mechanism did not come as some sort of revelation in Dr. Meyer’s testimony. She simply noted that absent seeing the injury-causing instrument or finding a telltale sign — such as a hand or cord imprint — a doctor would not be able to state what exactly caused a particular bruise. But Dr. Meyer nevertheless stated with *300confidence that the complainant’s injuries were a result of abuse due to the large number of bruises, the fact that most of the bruises were in locations that were unusual for toddlers, and that appellant’s history was inconsistent with the physical evidence. In addition, Nurse Allen did observe a mechanism for some of the bruises — many bruises on the face, chin, neck, and shoulder appeared to be hand bruises.
The Court of Appeals’s fifth reason was that Dr. Newsom formed the opinion that the complainant was not a victim of child abuse, that appellant’s history was consistent with the injuries, and that the injuries could have been accidental. But, as discussed above, the Court of Appeals gave no reason why the jury was required to accept Dr. Newsom’s testimony over that of the State’s three medical witnesses directly to the contrary. And even Dr. Newsom could not rule out the possibility that the injuries were the result of child abuse. Moreover, the State’s witnesses all examined the complainant; Dr. Newsom did not.
The Court of Appeals’s final reason was that the complainant might have inflicted the injuries upon herself, either through self-abuse or accident-prone behavior. The only injuries that appellant claimed the complainant intentionally inflicted upon herself were the sexual injuries. Dr. Marquardt and Dr. Meyer disputed the claim that the sexual injuries were self-inflicted. Regardless, there were numerous other injuries to the head, neck, and back that no one at trial claimed to be a result of self-abuse. To these injuries, the claim was that the complainant was accident-prone. But Rawlings, the original foster mother, testified that while in her care the complainant did not bruise more than any other child. Moreover, the evidence showed that the complainant had very few of the typical toddler injuries. If the complainant were accident-prone, it would be logical for the jury to expect that she would have far more bruises in the typical places for toddler injuries. And even if it were believed that the complainant was accident-prone, there was still the evidence that appellant’s explanations were inconsistent with the injuries received. The Court of Appeals did not even attempt to explain away all of this evidence adverse to the defense’s position.
In most of the six reasons given, the Court of Appeals distorted the evidence, viewed the evidence in the light most favorable to the defense, relied upon speculation, and disregarded contrary prosecution evidence. The only evidence cited by the Court of Appeals that truly favored the defense was Dr. Newsom’s testimony and appellant’s testimony. But the Court of Appeals made no attempt to explain why or how this testimony outweighed the prosecution evidence. The Court of Appeals just asserted that it did. The Court of Appeals’s discussion fails to explain adequately why the evidence is factually insufficient to prove that the child’s injuries were caused by intentional acts.
In sum, then, the Court of Appeals failed in numerous respects to conduct a proper factual sufficiency review as required by Clewis and Cain. The remedy is to remand the case to the Court of Appeals in order that it might conduct a proper review. I join the opinion of the Court.
. Cain v. State, 958 S.W.2d 404, 407, 408 (Tex.Crim.App.1997).
. Id.
. The record shows that the older sister had been removed from the Goodman home because she allegedly sexually abused this complainant. But an adoption worker from a state agency testified that the older girl was removed from the household solely on the basis of reports made by the Goodmans.
. Dr. Meyer was not on duty when the complainant was brought to the hospital on February 7th but examined the complainant the next day.
. Dr. Meyer testified:
[Meyer]: Well, I’m not certain that on the CAT scan there would be a difference in appearance of the blood in less than 24 hours. In fact, I doubt that there would be. Why you would have a subdural hematoma a day before and what would cause you to quit breathing the next day — I mean there aren’t very many reasons a child would stop breathing.
[Prosecutor]: Okay. And the recent injury causing the subdural hematoma would be one of them?
[Meyer]: Right.
. When asked if she ever popped the complainant on the bottom or slapped her, appellant acknowledged that she might have done so twice during the six-month period the complainant lived with her. Appellant also testified that her husband had sometimes spanked the complainant’s older sister on the bottom, causing bruises.
. Appellant worked at the hospital.
. Appellant testified that she showed her husband the bruises on the complainant’s back after the complainant fell from the Blazer.
. The record is unclear, but at one point, apparently referring to an unnamed bruise in a picture — that may have been the collarbone bruise — appellant stated, "I don’t know where that one came from.”
.Cain, 958 S.W.2d at 409-410.
. Cain, 958 S.W.2d at 408-409.
. Appellant acknowledged violating the parenting class teachings on two occasions. She also acknowledged that her husband violated the teachings by spanking the complainant's older sister.
. Cain, 958 S.W.2d at 407.
. Id. (quoting Clewis; ellipsis in Cain).
. Brown v. State, 911 S.W.2d 744, 745-746 (Tex.Crim.App.1995).
. The Court of Appeals relied upon our prior case of Johnson v. State, 673 S.W.2d 190, 196 (Tex.Crim.App.1984). But that case relied upon the now defunct outstanding reasonable hypothesis test for determining the sufficiency of the evidence. Id. at 195-197; see also Geesa v. State, 820 S.W.2d 154, 160-161 (Tex.Crim.App.1991). While alternative hypotheses may be a factor in conducting a factual sufficiency review, we have never held that the old outstanding reasonable hypothesis test must be satisfied for the evidence to be factually sufficient. In addition, there are factual distinctions between the Johnson and the present case. In Johnson, there was little, if any, evidence that appellant had any relationship with the child. 673 S.W.2d at 194. And there was no evidence that the injury in Johnson was the result of intentional conduct. Id. at 197. The present case is more like Willis v. State, 785 S.W.2d 378, 382 (Tex.Crim.App.1989), in which a defendant charged with arson gave a story that was inconsistent with the physical evidence. We upheld the conviction against a sufficiency challenge where the evidence showed that the fire was intentionally set and the defendant's story of his actions during the fire was inconsistent with the physical evidence. Id.