concurring
I join Chief Justice Frost’s opinion with the exception of section II.A. I write separately to address appellant’s second issue, in which he seeks a new trial on punishment. Appellant contends the trial court erroneously included a voluntary intoxication instruction in the punishment phase jury charge when the instruction is permissible only in the guilt/innocence phase charge. I conclude the challenged instruction was error under the circumstances of this case, However, based on my review of the record in light of the harm factors articulated in Almanza v. State, 686 S.W.2d 157, 171 (Tex. Crim. App. 1985) (op. on reh’g), I would hold that the erroneous instruction did not cause appellant actual harm under the present circumstances. Accordingly, I agree with the plurality that we should overrule appellant’s second issue, although for different reasons, and concur in affirming the judgment,
A. Charge Error
A jury convicted appellant of aggravated robbery with a deadly weapon. In the pum-ishment phase, the trial court included the following instruction in the jury charge, over appellant’s objection:
Voluntary intoxication does not constitute a defense to the commission of a crime. “Intoxication” means disturbance of mental or physical capacity resulting from the introduction of any substance into the body.1
The trial court should have sustained appellant’s objection. In Taylor v. State, 885 S.W.2d 154, 156 (Tex. Crim. App. 1994), the Court of Criminal Appeals addressed the propriety of including an identical instruction in a jury charge. Though the instruction in Taylor appeared in the guilt/innocence charge and not the punishment charge, the court addressed the purpose of Texas Penal Code section 8.04(a) and clearly stated it is not a mitigation provision but is “directed to the guilt/innocence phase of trial.” Id. I agree with, our dissenting colleague that Taylor speaks plainly enough on this point, though no published Texas case has squarely addressed whether inclusion of an instruction under section 8.Q4(a) in the punishment charge constitutes error.2
This is not to say I believe an instruction under section 8.04(a) would never be appropriate in a punishment phase jury charge. For example, language tailoring such an instruction to particular evidence or a defense may serve,- in certain cases, to “lead and to prevent confusion,” as jury charges are supposed to do. See Reeves v. State, 420 S.W.3d 812, 818 (Tex. Crim. App. 2013). Here, the instruction could *641have been modified to apply only to extraneous offenses.3
B. Harm Analysis
Because appellant preserved error by objecting to the voluntary intoxication instruction at trial, he is entitled to reversal if the record shows that the erroneous instruction was “calculated to injure [his] rights.” Tex. Code Crim. Proc. art. 36.19; Reeves, 420 S.W.3d at 816; Wooten v. State, 400 S.W.3d 601, 606 (Tex. Crim. App. 2013); Almanza, 686 S.W.2d at 171. The Court of Criminal Appeals has interpreted this requirement to mandate reversal only when the charge error caused “some” harm—meaning any harm, regardless of degree, is sufficient to require reversal so long as the defendant “suffered some actual, rather than merely theoretical, harm from the error.” Reeves, 420 S.W.3d at 816; Arline v. State, 721 S.W.2d 348, 351 (Tex. Crim. App. 1986).
To gauge “actual” harm, we consider the entire jury charge, the state of the evidence, including the contested issues and weight of probative evidence, the argument of counsel, and any other relevant information revealed by the trial record as a whole. See Reeves, 420 S.W.3d at 816; Wooten, 400 S.W.3d at 606; Bailey v. State, 867 S.W.2d 42, 43 (Tex. Crim. App. 1993); Almanza, 686 S.W.2d at 171. We conduct this analysis independently because neither appellant nor the State has the burden to prove harm. Reeves, 420 S.W.3d at 816.
Whether the outcome of the punishment trial could, have been affected by the erroneous instruction is not the question. See Wooten, 400 S.W.3d at 604, 606, 610. To support reversal, the degree of harm resulting from charge error must rise, above possibilities and courts must delve into “the extent to which the outcome of trial was actually affected.” See Saunders v. State, 817 S.W.2d 688, 690 (Tex. Crim. App. 1991); see also Wooten, 400 S.W.3d at 604, 606, 610 (court of appeals must address likelihood that jury would have made finding in question). Yet, as a practical matter, courts have not consistently described the point at which harm evolves from “theoretical” into “actual” harm in terms of. an articulable degree of likelihood.4 Sometimes, courts have considered whether actual harm appears “reasonably likely” or “likely;”5 in other cases, courts have looked for “fair assurance;”6 still other times, courts have reached a decision without specifically stating any particular degree of likelihood required.7 In my view, *642whether appellant is entitled to a new punishment trial depends on the substantive border between theoretical and actual harm under the present circumstances. I will draw the line at whether, in light of the Almanza factors, there exists a reasonable likelihood that the erroneous instruction actually affected the outcome. Ross, 133 S.W.3d at 624 (unable to conclude there was reasonable likelihood that charge error caused actual harm).81 invite the Court of Criminal Appeals to address and clarify these issues should it be so inclined.
Appellant argues the voluntary intoxication instruction actually harmed him because a reasonable juror would have interpreted it as an instruction to completely disregard his evidence of intoxication—Xa-nax addiction—for mitigation of punishment purposes. Under Almanza and its progeny, we must determine the likelihood that the jury would not have considered appellant’s intoxication evidence for mitigation purposes under the circumstances of this particular case. See Wooten, 400 S.W.3d at 608-09 (determining likelihood that jury would have believed appellant acted out of sudden passion had instruction been provided); see also Herrera, 513 S.W.3d at 229, 2016 WL 7480502, at *5. If the jury considered appellant’s voluntary intoxication evidence but afforded it little or no weight, then appellant suffered no actual harm from the challenged instruction. In that instance, the jury would have been within its right to assess the maximum punishment, and the law would not support dislodging the verdict.
In my view, based on this record, any likelihood that the jury would have construed the instruction as a directive to not consider the voluntary intoxication evidence is not sufficient to constitute actual harm. The Almanza factors support this conclusion.
1. The entire charge
I begin by evaluating the jury charge as a whole. The instruction at issue did not tell the jury to disregard appellant’s voluntary intoxication evidence in assessing punishment. Rather, it said, “[vjolun-tary intoxication does not constitute a defense to the commission of a crime.” Assuming the instruction, viewed in isolation, potentially created confusion as to whether the jury was permitted to consider appellant’s voluntary intoxication evidence, the risk of actual confusion was ameliorated by very explicit instructions appearing later in the charge. The trial court told the jury in no uncertain terms that it “may take into consideration all the facts shown by the evidence admitted before you in the full trial of this case”—which necessarily included all the voluntary intoxication evidence admitted during both the guilt/innocence and punishment phases. That instruction is clear and unambiguous on its face, and we are to presume the jury followed it. See Elizondo v. State, 487 S.W.3d 185, 208 (Tex. Crim. App. 2016); Miles v. State, 204 S.W.3d 822, 827-28 (Tex. Crim. App. 2006); Rose v. State, 752 S.W.2d 529, 554 (Tex. Crim. App. 1987) (op. on reh’g), *643abrogated on other grounds by Karenev v. State, 281 S.W.3d 428 (Tex. Crim. App. 2009).9
When a trial court instructs the jury properly elsewhere in the charge or during trial, an erroneous instruction is generally harmless. See, e.g., Ross, 133 S.W.3d at 623-24; Castaneda v. State, 852 S.W.2d 291, 295-96 (Tex. App.—San Antonio 1993, no pet.) (jury given proper instruction elsewhere during trial); see also Henriksen v. State, 500 S.W.2d 491, 496 (Tex. Crim. App. 1973) (instruction omitted from punishment charge was referenced in guilt/innocence charge and sufficient to fully protect defendant’s rights); cf. Ngo v. State, 175 S.W.3d 738, 752 (Tex. Crim. App. 2005) (noting that charge error was not ameliorated elsewhere in another portion of the charge).
Moreover, a reasonable juror would have likely cpnstrued the instruction as pertaining to the extraneous crimes and bad acts evidence, most of which the State introduced without objection. The dissent disputes this assertion because, it contends, a section 8.04(a) instruction is irrelevant to extraneous crimes or bad acts evidence, and because of the challenged instruction’s relatively segregated placement within the charge as a whole. I respectfully disagree. The dissent’s first contention is inconsistent with the record. The voluntary intoxication instruction pertained to the extraneous crimes issue because there was evidence appellant was taking Xanax at the time of at least some of the extraneous acts. See Taylor, 885 S.W.2d at 158 (when there is evidence from any source that might lead a jury to conclude the defendant’s intoxication somehow excused his actions, an instruction that voluntary intoxication is not a defense is appropriate). Appellant’s counsel notably appreciated this fact, as he urged the jury during closing argument to account for the intoxication evidence in weighing the alleged extraneous crimes and bad acts. The court told the jury it could consider extraneous “crimes”; the court also told the jury that voluntary intoxication is not a defense to a “crime.” If the jury understood the challenged instruction as applying to extraneous crimes or bad acts evidence, then all the instruction would have led the jury to believe is that appellant’s Xanax addiction did not prevent it from finding that appellant committed extraneous crimes beyond a reasonable doubt. But it does not follow that the jury could not consider the voluntary intoxication evidence in assessing punishment for the charged offense.
Second, the dissent is correct that an instruction’s placement in the charge may in some cases weigh in favor .of finding some harm. Reeves, 420 S.W.3d at 819. Here, however, I would hold that any potential confusion suggested by the instruction’s location within the body of the charge was ■ clarified by the subsequent explicit directive to consider all of the evidenced Again, to be sure, tailoring language would have provided greater clarity when reading the voluntary intoxication instruction .in conjunction with the later instruction to consider all the evidence from the entire trial. But I do not believe a reasonable jury would be so misled by this charge as a whole. It could have construed and applied both instructions without inconsistency based on the evidence and arguments presented to it.
*644It is not reasonably likely that the jury would have construed the instruction as appellant suggests for other reasons. For example, the State did not object to appellant’s voluntary intoxication evidence, and the court admitted all of it. At no time' during the trial did the court instruct the jury not to consider any part of the voluntary intoxication evidence. I do not believe a reasonable juror would have interpreted a section 8.04(a) charge instruction as a directive to disregard the totality of intoxication evidence without' the instruction explicitly saying so, considering the intoxication evidence was admitted without objection.
Finally, the jury did not send any notes during deliberations indicating confusion about whether it was free to consider appellant’s voluntary intoxication' evidence. See Ross, 133 S.W.3d at 624.
Considering the charge as a whole, I do not believe- the erroneous instruction resulted in a reasonable likelihood- that the jury was so misled as to believe it could not consider appellant’s voluntary intoxication evidence. Thus, I conclude the first Almanza factor does not indicate actual harm. See id. (erroneous instruction at punishment did not result, in reasonable likelihood of jury being so misled or applying instruction in way that prevented it from considering issue).
2. Argument of counsel
The record of closing argument shows that neither parties’ counsel construed the charge as precluding consideration of appellant’s voluntary intoxication evidence. Appellant’s counsel argued without objection that the intoxication evidence, and Dr. Rustin’s testimony in particular, was relevant both to mitigation of punishment and whether the jury could find that appellant committed the alleged extraneous crimes and bad acts. Appellant’s argument expressly distinguished between voluntary intoxication as a defense to a crime and voluntary intoxication as mitigation.
Additionally, the State did not argue that the jury was precluded from considering the intoxication evidence; it argued that the jury should not believe it. Those are different contentions entirely. The State argued expressly that Dr. Rustin’s testimony was not credible for several different reasons. The State never told the jury it could not consider Dr. Rustin’s testimony nor did the State object, to Dr. Rustin’s testimony. Like appellant, the State acknowledged that Dr. Rustin was called for mitigation purposes. On appeal, appellant does not complain that the, court excluded any evidence material to his voluntary intoxication theory, so the jury had before it all the defensive intoxication evidence appellant sought to introduce. The State’s argument tied appellant’s drug usage to the extraneous crimes and bad acts evidence. Lastly, the State argued that appellant’s voluntary "intoxication evidence was not credible because appellant committed crimes even when not taking Xa-nax.
I conclude this factor does not indicate a reasonable likelihood that the challenged instruction would have misled a jury into believing .that it was prohibited from considering appellant’s voluntary intoxication evidence. See Arline, 721 S.W.2d at 353 (finding charge error harmless; charge was not misleading in light of closing arguments).
3. The state of the evidence, including the contested issues and weight of probative evidence
In assessing the extent -to which the charge error “actually affected”10 the outcome—life imprisonment—I also look to *645the overall evidence, including its probative weight. Reeves, 420 S.W.3d at 816, 820; Wooten, 400 S.W.3d at 606. According to appellant, absent the erroneous instruction it is “entirely possible” that the jury would not have imposed the maximum sentence available. While courts consider the sentence as part of the harm analysis, a maximum sentence does not alone establish actual harm resulting from charge error. See White v. State, 779 S.W.2d 809, 828 (Tex. Crim. App. 1989) (no harm from omission- of “no-adverse-inference” in death penalty case); Castaneda, 852 S.W.2d at 296.
In my view, the overall state of the evidence does not give rise to a reasonable likelihood that the jury assessed the maximum sentence because it believed the voluntary intoxication evidence could not be considered in mitigation. It is far more likely that the jury’s verdict was motivated by the overwhelming weight of the nature of the charged offense and appellant’s extraneous offenses, together with the weakness of Dr. Rustin’s testimony and other intoxication proof.
For example, the jury heard damning evidence not only as to the charged offense but also a. litany of other extraneous crimes and bad acts committed by appellant. There was evidence that appellant participated in a murder where the victim was shot in the head while sitting in an automobile. That homicide occurred the day before appellant committed the charged offense of aggravated robbery with a deadly weapon, which unfolded with appellant brandishing a gun and approaching the victim in his car—much like the circumstances of the extraneous homicide. On multiple jail phone recordings, appellant is heard discussing how he tried to-“pull the trigger” twice while committing the charged offense, only to have his gun “jam.” On one recording, appellant also discussed the homicide. Notably, the only evidence the jury asked to re-hear during its punishment deliberations was one of the phone recording's. The jury heard evidence that appellant sold drugs and was “high on Xanax” a “lot of times.” There was also evidence appellant committed multiple assaults against family-members and others, including beating a jail inmate unconscious merely to obtain a private -cell. Cf. White, 779 S.W.2d at 828 (as part of its determination that appellant did not suffer some harm from punishment charge error, court notéd that the State introduced evidence of appellant’s commission of “remarkably similar murder” just days' before the charged offense, as well as evidence of other crimes and his general reputation for violence).
, Appellant called Dr. Rustin as a witness during the punishment phase. Dr. Rustin testified, without objection, that Xanax is a sedative that reduces inhibitions ánd one’s fear of punishment. Dr. Rustin said that appellant began using Xanax at a “young” age and increased his usage, according to appellant, up to eight two-milligram pills in a day. Appellant acquired the drug “off the street” or via the internet without a prescription. On cross-examination, Dr.- Rus-tin acknowledged that his testimony was based solely" on a single interview with appellant, who self-reported his own usage. Dr. Rustin reviewed no medical records regarding appellant’s drug use, nor did appellant offer any medical, records into evidence. Appellant successfully completed a drug treatment program in 2009 but continued committing offenses nonetheless. Dr.' Rustin acknowledged that appellant committed violent offenses in jail, while not under the influence of Xanax. The jury also heard a jail phone -recording in which one individual suggested to appellant that he should assert drug usage as a defense.
*646Relative to other evidence, the substance of Dr. Rustin’s testimony was weak and undermined substantially on cross-examination. In contrast, the State presented overwhelming evidence of extraneous crimes and bad acts.11 Even considering the totality of appellant’s voluntary intoxication evidence heard by the jury, it is not reasonably likely the jury afforded it more than nominal weight, if any, based on this record. Considering the state of proof, the jury was within its right to assess the maximum sentence for an offense of this nature.12 Accordingly, while I do not discount the gravity of the sentence, I cannot conclude that the jury was reasonably likely to assess less than the maximum punishment allowed but for the erroneous voluntary intoxication instruction.13
4. Other relevant factors.
Lastly, I consider any other relevant information revealed by the trial record as a whole. See Reeves, 420 S.W.3d at 816; Wooten, 400 S.W.3d at 606.
Appellant argues the instruction harmed him because it went to his sole defense for mitigation of punishment. See Reeves, 420 S.W.3d at 820-21. The intoxication evidence is undermined only if the jury believed it could not consider it as a justification for assessing punishment. But there is no indication from the record that the jury so believed. The trial court informed the jury it was free to consider all of the evidence admitted, and both parties’ counsel emphasized the nature and quality of the voluntary intoxication evidence without ever telling the jury it was precluded from considering the intoxication evidence. There was no jury note expressing confusion on that point. Appellant’s counsel argued that the jury should give lesser punishment based on Dr. Rustin’s testimony. True, the Xanax evidence was appellant’s principal defense on punishment, but it was not the central issue permeating the collective entirety of punishment evidence. While the erroneous instruction may have pertained to appellant’s sole mitigation defense, the record does not indicate a reasonable likelihood that the instruction actually affected the jury’s consideration of that defense.
The dissent criticizes the trial court’s stated justification for including the sec*647tion 8.04(a) instruction in the punishment phase charge. Assuming the dissent is correct, however, the trial court’s rationale could not have caused actual harm because the jury did not hear it. The court’s motivation for including the voluntary intoxication instruction, right or wrong, could not have actually affected the jury’s interpretation of the instruction.
Finally, in my view, Kresse, 2010 WL 1633383, at *2, also cited by the dissent, does not suggest a contrary outcome. Kresse held the inclusion of a section 8.04(a) voluntary intoxication instruction in the punishment phase charge was harmful. There, Kresse did not want the jury to consider ample intoxication evidence against him in assessing punishment for murder. Id. Kresse argued, and the Fort Worth court of appeals agreed, that placing the instruction in the punishment charge drew attention to his history of excessive drinking, which the prosecutor emphasized repeatedly in opening and closing. Id. The harm issue in the present case is much different. Here, the question is whether the voluntary intoxication instruction was likely to mislead the jury into thinking it could not consider appellant’s intoxication evidence. For the reasons expressed, I believe it was not.
Conclusion
For these reasons, I would hold that erroneously including a section 8.04(a) voluntary intoxication instruction in the punishment phase jury charge did not cause actual harm under the present record. Accordingly, I respectfully concur with the judgment of affirmance.
(Frost, C.J., plurality)
(Christopher, J., dissenting)
. The instruction tracked the statutory language verbatim from Texas Penal Code sections 8.04(a) and (d). Tex. Penal Code § 8.04(a), (d).
. In Kresse v. State, No. 02-09-00271-CR, 2010 WL 1633383 (Tex. App.—Fort Worth Apr. 22, 2010, no pet.) (mem. op., not designated for publication), the appellant, .chafi lenged the inclusion of a section 8.04(a) voluntary intoxication instruction in the punishment phase jury charge. Id. at - *2, But the state conceded error, and the court of appeals, did not reach the issue.
. In support of the conclusion that including the voluntary intoxication instruction in the punishment phase charge was not error, the plurality states that the absence of tailoring language does not render the instruction misleading. While I believe-it is not reasonably likely the jury was misled .by the voluntary intoxication instruction in this case, in my view that question is much more pertinent to the harm analysis father than whether the instruction was error in .thé first instance.
. The elusiveness of a consistently stated test may be in part by design. See Saunders, 817 S.W.2d at 689 ("Much of the impetus for our holding in Almanza was a perceived need to avoid the tyranny of hard and fast, rules.").
. See, e.g., Ross v. State, 133 S.W.3d 618, 624 (Tex. Crim. App. 2004); Herrera v. State, No. 04-16-00138-CR, 513 S.W.3d 223, 229, 2016 WL 7480502, at *5 (Tex. App.—San Antonio Dec. 30, 2016, no pet. h.); Mendoza v. State, 349 S.W.3d 273, 283 (Tex. App.—Dallas 2011, pet. ref’d).
. Trevino v. State, 100 S.W.3d 232, 243 (Tex. Crim. App. 2003) (per curiam).
. See, e.g., Reeves, 420 S.W.3d at 816; Dickey v. State, 22 S.W.3d 490, 492-93 (Tex. Crim. App. 1999); Abdnor v. State, 871 S.W.2d 726 (Tex. Crim. App. 1994); Arline, 721 S.W.2d at 353); Bravo v. State, 471 S.W.3d 860, 871 (Tex. App.—Houston [1st Dist.] 2015, pet. ref'd).
. I embark on the harm analysis in this case, like every case, without the benefit of information that would most clearly establish whether a particular instruction in reality had any effect on the jury’s verdict: juror testimony. Tex. R. Evid. 606(b). Courts may not consider direct evidence from the jury as to what factors motivated its deliberations and verdict. Tex. R. Evid. 606(b). Thus, in applying the Almanza factors, we will never know with certainty whether the error made a difference in the outcome. Rather, the Almanza exercise will yield merely an assessment of harm to the degree of likelihood necessary to show “some” harm based on the peculiarities of each case. Some harm occurs, and reversal is required, if we find the likelihood of harm exceeds the "theoretical.”
. I distinguish the character of the present instruction from that at issue in Reeves, in which the court stated that only “understandable” instructions warrant a presumption that juries followed them. Reeves, 420 S.W.3d at 818. The language here at issue is unlike the “impenetrable” "argle-bargle” in Reeves. Id. at 817.
. Saunders, 817 S.W.2d at 690.
. The trial court properly instructed the jury that it could consider evidence of extraneous crimes or bad acts in assessing punishment only if the extraneous crime or bad act was shown by the State beyond a reasonable doubt to have been committed by the appellant or was one for which the appellant could be held criminally responsible. Tex. Code Crim. Proc. art. 37.07, § 3(a)(1); see also Huizar v. State, 12 S.W.3d 479, 483-84 (Tex. Crim. App. 2000).
. Cf. Hopkins v. State, 487 S.W.3d 583, 584 (Tex. Crim. App. 2016); Bazan v. State, 403 S.W.3d 8, 10 (Tex. App.—Houston [1st Dist.] 2013, pet. ref’d); Parker v. State, 51 S.W.3d 719, 721 (Tex. App.—Texarkana 2001, no pet.); Phillips v. State, No. 07-15-00357-CR, 2016 WL 3696732, at *1 (Tex. App.—Amarillo July 11, 2016, no pet.) (mem. op., not designated for publication); Jackson v. State, No. 05-14-00985-CR, 2015 WL 3899573, at *1 (Tex. App.—Dallas June 25, 2015, pet. ref'd) (mem. op., not designated for publication); Patterson v. State, No. 05-13-00450-CR, 2015 WL 2400809, at *1 (Tex. App.—Dallas May 19, 2015, pet. ref’d) (not designated for publication); Carter v. State, No. 14-14-00061-CR, 2014 WL 5780691, at *1 (Tex. App.—Houston [14th Dist.] Jan. 14, 2015, pet. ref'd) (mem. op., not designated for publication); Gilstrap v. State, No. 14-01-011867-CR, 2002 WL 31718482, at *1 (Tex. App.—Houston [14th Dist.] Dec. 5, 2002, no pet.) (mem. op., not designated for publication).
.The dissent cites Erazo v. State, 167 S.W.3d 889, 891 (Tex. App.—Houston [14th Dist.] 2005, no pet), as analogous support for finding harmful error when a jury assesses life imprisonment. But Erazo involved the erroneous admission of evidence, not jury charge error, and applied a different test than the one at issue here. See id. at 890.