OPINION
Opinion by
Chief Justice MORRISS.
In a Van Zandt County1 jury trial, William James Stewart was convicted of sexual assault of a child under seventeen years of age. See Tex. Penal Code AnN. § 22.011(a)(2) (Vernon Supp. 2008). During the punishment phase of trial, without objection, the jury was erroneously charged with an incorrect statutory alternative regarding parole law. See Tex.Code CRiM. PROC. Ann. art. 37.07, § 4 (Vernon Supp. 2008). The jury assessed punishment at twelve years’ imprisonment and a $5,000.00 fine, and Stewart was sentenced accordingly.
On appeal, Stewart does not contest his conviction. Instead, he attacks the charge error occurring at the punishment stage and also alleges ineffective assistance of counsel and a conflict of interest by his counsel. We affirm the judgment of the trial court because (1) the punishment-phase charge error was not egregiously harmful to Stewart, (2) Stewart was not harmed by his attorney’s failures regarding the erroneous punishment charge, and (3) Stewart’s attorney’s prior relationships did not color Stewart’s defense.
Stewart lived with A.N., A.N.’s mother, and Stewart’s son in Van Zandt County in 2004 and 2005. A.N., Stewart’s then-fifteen-year-old stepdaughter, described him as a good friend and listener, one who encouraged her to stop her drug usage and to stop hurting herself — she had been using drugs and having sex with boys since [855]*855she was twelve years old and described herself as a “cutter,” one who physically injured herself. She testified that she began writing poems and letters to him and that she was in love with him. She testified that she resented her mother for her relationship with Stewart and eventually considered him her boyfriend. She testified that she had sex with Stewart multiple times in 2004, that the sex was consensual, and that she stopped the sexual relationship after she had an argument with her mother.
A.N. testified that her mother took her to a detention center in January 2005, where A.N.’s statement was taken. In an interesting description of what followed, she testified the State’s representatives
tried to talk to me about some of the letters and poems that I had written Bill — or about Bill, and I tried to deny everything that was going on. Then the blonde took out some handcuffs and put them on my arms and then took me to the back where, I guess — I want to say it was a holding cell, I guess you could say, and then they took my jewelry off and my shoes off. And they put me in the holding cell and they had me do exercises like jumping jacks and different exercises.
She went on to testify that she finally told them something was going on because she wanted to go home and was afraid that she would be stuck at the detention center if she did not tell them something. A.N. added that she had severe asthma. She reiterated that even so, she was telling the truth about their relationship.
On cross-examination, she explained that her relationship with her mother was bad, that they had fights, and that her mother had hit her in the face and “busted” her lip. A.N. went to police and attempted to file charges against her mother. The month after A.N.’s attempt to file charges against her mother, her mother took her to the detention center. A.N. testified that, after being questioned at the detention center for about eight hours, she finally told them that she and Stewart had engaged in sex. A.N. also testified that, looking back on the sexual relationship Stewart had enjoyed with her during the three or four years before she testified at trial, she believes Stewart took advantage of her.
Two of Stewart’s daughters from a previous relationship testified about their good relationship with their father and their disbelief of any sexual impropriety. They also testified about the conversations they had with A.N. in which she had stated repeatedly that nothing had happened and that she had only said so in order to get out of the detention center.
(1) The Punishmentr-Phase Charge Error Was Not Egregiously Harmful to Stewart
Stewart first complains because the wrong parole-law charge was given to the jury at punishment. We agree. One paragraph dealing with parole contained erroneous language, and that paragraph was included in the jury charge on punishment along with four others dealing with parole.
The statute specifically sets out three lengthy, alternative jury charges concerning the parole law; and those are to be chosen based on a very exacting and at least potentially confusing set of conditions. See Tex.Code Crim. Proc. ANN. art. 37.07, § 4(a)-(c). Depending on the offense of which a defendant has been convicted, whether his or her sentence is to be enhanced, and whether a deadly-weapon finding has been made in connection with the conviction, the trial court is to select which one of the three alternatives will be [856]*856given to the jury. In Stewart’s punishment trial, the wrong one was chosen.
The jury was charged that Stewart would not become eligible for parole “until the actual time served plus any good conduct time earned equals one-fourth of the sentence imposed-” See Tex.Code Crim. ProC. AnN. art. 37.07, § 4(c) (emphasis added).2 The charge should have indicated that Stewart would “not become eligible for parole until the actual time served equals one-half of the sentence imposed ... without consideration of any good conduct time he may earn.” See Tex.Code Crim. Proo. Ann. art. 37.07, § 4(a) (emphasis added).3 It should have also added that, if he were sentenced to less than four years, he would be required to serve at least two years before, being eligible for parole. See id.
Neither the trial court, nor the attorneys, caught the problem. Thus, since no objection was made to the punishment charge, we look to see if egregious harm was caused by the error. Igo v. State, 210 S.W.3d 645 (Tex.Crim.App.2006); Almanza v. State, 686 S.W.2d 157, 171 (Tex.Crim.App.1984) (op. on reh’g). Jury-charge error is egregiously harmful if it affects the very basis of the case, deprives the defendant of a valuable right, or vitally affects a defensive theory. Allen v. State, 253 S.W.3d 260, 264 (Tex.Crim.App.2008).
The standard of harm is that for unob-jected-to error in the charge: whether “the error is so egregious and created such harm that he has not had a fair and impartial trial — in shorty egregious harm.... [T]he actual degree of harm must be assayed in light of 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 record of the trial as a whole.” Almanza, 686 S.W.2d at 171.
Trejo v. State, 280 S.W.3d 258, 261 (Tex.Crim.App.2009).
Courts generally agree that the statutory parole instructions were designed to favor the State and to increase sentences. See Arnold v. State, 786 S.W.2d 295, 300 (Tex.Crim.App.1990). But, the instructions can also help the defendant, because the jury could learn that the defendant [857]*857would serve longer than it expected and could be influenced to assess less time. Hooper v. State, 255 S.W.3d 262, 272 (Tex.App.-Waco 2008, pet. ref'd); Williams v. State, 975 S.W.2d 375, 378 (Tex.App.-Waco 1998, pet. ref'd).
First, we are to assess harm based, in part, on the entirety of the jury charge. The punishment charge consisted of sixteen paragraphs. Five of those dealt with parole. The paragraph containing the erroneous language is squarely in the middle of those five paragraphs. The two parole-related paragraphs following the erroneous paragraph are correct, are statutorily required, and essentially give the jury three main instructions: (1) one cannot predict how parole law and good conduct time might be applied to this defendant; (2) the jury may consider the existence of parole law and good conduct time, in general; but (3) the jury is not allowed to consider how good conduct time or the parole law might affect or be applied to this defendant, in particular. In other words, the jury was first erroneously instructed concerning when Stewart might become eligible to be considered for parole, but then it was told that, although it could consider, in general, the existence of parole and good conduct time, it could not apply that to Stewart, specifically.
The charge error was, in the context of much additional correct language, effaced by specific instructions not to apply the erroneous language to Stewart. This factor, alone, does not dictate a finding of egregious harm.
Second, we are also to assess harm based, in part, on the state of evidence, the contested issues, and the weight of probative evidence.
The court noted in Igo that the evidence relating to punishment was “exceptionally strong.” In this case, several additional witnesses were called during the punishment phase. Jan Brooks, a deputy director of the community supervision department for Van Zandt County, testified about her experience with Stewart’s pretrial supervision. She testified that he had always appeared for his weekly visits, but had gotten behind on payments to the community supervision department ($300.00 per month for a GPS anklet, plus a $40.00 per month “supervision” fee, and an additional monthly fee of $5.00 for urinalysis). She admitted that Stewart had caught up the shortfall and that he had appeared each week after driving from south of Houston to Van Zandt County, for two years. She complained, however, that his attitude was not the best.
During guilt/innocence, A.N. testified that she had vaginal sex with Stewart many times in Van Zandt County. The State called A.N. to the stand again during the punishment phase of trial. She then supplemented her prior testimony by recounting multiple instances of oral sex each performed on the other. She testified to one instance of anal penetration by Stewart, which hurt. She testified again that none of her actions were forced or coerced and that she engaged in the activity solely because she wanted to. She also indicated she had been in counseling as a result of the relationship with Stewart. She acknowledged, testifying with the perspective gained by having attained the age of eighteen years, that the relationship with Stewart when she was between the ages of thirteen and fifteen was wrong, that Stewart had taken advantage of her, and that, if she had it to do over again, she would stop him from the sexual relationship. She thinks it possible that Stewart might “do this” to someone else.
The defense also called several witnesses. Stewart’s employer, Nancy Luke, testified about his superior and reliable work. Stewart’s roommate found him en[858]*858tirely trustworthy and a good person. Stewart’s current girlfriend, Janice Boyd, who had lived with him for about fifteen months, explained that he had lost work for about a month after being bitten by thirteen spiders while mowing the yard and being placed on heavy antibiotics. She testified that, at this point, he could not walk or care for himself and that she had also provided some of the money to catch him up for the two months that he was behind on his payments.
Boyd testified at some length that she had come to trust Stewart, that his character was exceptional, and that his interactions with his own daughters was loving and careful. One of those daughters, with a special-needs daughter of her own, had moved in with Stewart and Boyd for a time. Boyd also explained that they had provided financial assistance for the daughter and granddaughter while they were living there.
Charlene O’Brien, one of Stewart’s three adult daughters, testified at punishment that Stewart was a very strict parent and that, when Stewart and her mother had divorced, Stewart was given custody of all three daughters. She stated that the relationships were strong and good and that there was no sexual impropriety with her. The State chose not to cross-examine O’Brien.
While the punishment evidence beyond the sexual relationship itself is not “exceptionally strong,” the extended and varied sexual relationship, the vulnerable position of A.N. essentially as Stewart’s stepdaughter, and the visceral nature of the offense itself offer sufficient support to explain the jury’s assessment of punishment without suggesting harm from the charge. Thus, nothing in the evidence suggests that the jury acted on the erroneous language.
Third, we are to assess harm based, in part, on the arguments made by counsel. The parole charge was the subject of argument by both the State and defense counsel, and both of them argued using the incorrect charge language.
The initial part of the State’s closing argument at punishment was a relatively brief ninety-nine lines of transcribed text in the reporter’s record. During this argument, the State sought to defuse the effect of Stewart’s character witnesses or his positive family or behavioral evidence, instead seeking to refocus the jury on the seriousness of the offense, in which a forty-five-year-old man had vaginal, oral, and anal sex with a fifteen-year-old girl. It also sought to deal with the evidence that the relationship between Stewart and A.N. was described as loving, gentle, and non-coercive, and to remind the jury of the relative ages of the “couple” and the responsibility under the law for the adult to refrain from such behavior with a child.
Near the beginning of this portion of the argument, the State mentioned that parole law and good conduct time exist, but added this discussion:
The judge has, in the charge, shown to you in paragraph 8, you are not to consider the manner which the parole law may be applied to this particular defendant, but you are allowed to consider that that parole law exists, that good conduct time exists, so I think that’s important when you’re back there trying to determine amongst yourselves what’s — what’s appropriate for this defendant.
At that point, in eleven lines of transcribed text, Stewart objected, making the point that the effect of the parole law is not to be considered by the jury in determining Stewart’s sentence. The State responded, in ten lines of transcribed text, by apologizing to the trial court “if that was confusing to the jury” and emphasized the in-[859]*859struetion in paragraph 8 that the jury is not to consider how parole law might be applied to Stewart. None of these comments used or referred to the erroneous language in the punishment charge.
Stewart’s closing argument at punishment covered a hefty 338 lines of transcribed text. During this argument, counsel addressed the range of punishment and the jurors’ need to consider, not only the maximum, but also the minimum end of the range. Counsel also attempted to defuse the fact that the offense was sexual assault of a child, noted how the grade of offense changes with various circumstances, and made a serious effort at giving the jury a rationale for satisfactorily assessing punishment in the lower end of the range. He made a foray into Biblical concepts about human weaknesses toward fleshly sin. He talked about the love between the parties. He painted the picture of Stewart as a valiant and loving family man. He touched on the lifetime obligation Stewart will have to register as a sex offender. He urged a lower punishment.
During that closing argument, Stewart’s counsel touched on parole law twice. Two or three pages into his argument, counsel discussed the parole law in twenty-seven lines of argument. At this time, counsel made no reference to the erroneous charge language and emphasized his main point that the jury was not to consider the effect of the parole law on Stewart.
Near the end of his argument, Stewart’s counsel again referenced parole law, this time in fourteen lines of argument. During that segment of his argument, counsel reiterated to the jury that it was not allowed to use the effect of parole law on Stewart to calculate his sentence. In the process, he made this argument:
[Parole law] is out there, but you cannot go back there and say, [“] Well, okay, if he’s really good and he paroles out after a quarter of his time, I want to make sure he does X number of years, so let’s give him the X to make sure he does that.[”] You cannot do that. That’s not your consideration. You say, [“] You did this, these are the number of years that you deserve, and it’s up to the TDC to decide if and when you get out and under what restrictions. [”]
It must be noted that counsel, here, did make a passing reference to the erroneous one-quarter measure as used in the charge to the jury, but it was all couched in terms urging the jury not to use any of that to calculate Stewart’s sentence.
The rebuttal portion of the State’s closing argument at punishment consumed ninety-nine lines of transcribed text. At no point during that argument did the State’s attorney return to the subject of the parole law.
"While both sides made reference to the parole law in closing arguments, both emphasized that how it applied to Stewart was not to be considered in assessing punishment. There was only one small, transitory reference to the erroneous charge language, and that was in the context of emphasizing that the jury was not to use the parole law to calculate Stewart’s sentence. The issue of the parole law was not central to the case regarding punishment.
We conclude that the jury arguments do not establish egregious harm.
Finally, we are to assess harm based, in part, on any other relevant information revealed by the record.
Nothing in the record suggests that the jury had any question concerning either the application or meaning of the parole law or that the parole law affected its assessment of punishment. There is no indication of how the jury read or applied the parole law given, or of any untoward [860]*860attention the jury might have given to the parole law. Thus, there is no active showing of any effect on the jury’s deliberation or its assessment of punishment.
At twelve years’ imprisonment, the prison term is not the maximum possible; in fact, it is close to the middle of the permissible range of two years to twenty years. In the same way, the fíne of $5,000.00 is precisely half of the maximum the jury could have assessed, that is, $10,000.00. Thus, one matter often used to show harm will not apply, as the sentence was not particularly harsh.
The middle-of-the-range punishment does not support a conclusion that the jury attempted to apply the parole law to Stewart, or to predict how it might be applied to him. Because the jury was instructed not to consider how the parole law might apply to Stewart, any harm to him here would be the result of a violation of the statute by the jury. We are to presume that the jury follows the instructions given to it by the trial court, unless there is indication to the contrary. Gamboa v. State, — S.W.3d -, - (Tex.Crim.App.2009); Luquis v. State, 72 S.W.3d 355, 366-67 (Tex.Crim.App.2002).
The State argues that we should be guided by Igo4 In Igo, the Texas Court of Criminal Appeals held that — in spite of an incorrect conversion of one-quarter to one-half like the one in this case — the error did not cause egregious harm because, (1) even though the maximum punishment was given, (2) the parole instruction contained the standard curative language admonishing the jury not to consider the extent to which the parole law might be applied to the defendant, (3) parole was mentioned by neither counsel during argument, and (4) the evidence relating to punishment was exceptionally strong. Igo, a teacher, had taken a fifteen-year-old student on a trip, had engaged in sex with her, had continued the relationship even after indicted, and had tried to bribe her to drop the charges. Igo, 210 S.W.3d at 647.
The State uses Igo as the basis for its argument- that we should distinguish the present case from our decision in Hill v. State, 30 S.W.3d 505 (Tex.App.-Texarkana 2000, no pet.). In Hill, we applied the Almanza egregious-harm standard to a similar, but arguably weaker, parole-charge error. We noted the importance of the fact that the charge erroneously indicated Hill was entitled to good conduct time. We found that fact to be important because the statutory charge states that the jury may consider the existence of the parole law and good conduct time in making its decision. We discounted the language that the jury was not to attempt to apply the parole law to Hill. We then concluded that the court’s “misstatement of the law in [that] case misled the jury and seriously affected how it viewed the existence of parole and good conduct time, which the instructions plainly told the jury it could consider.” Id. at 508.5
[861]*861As is explicitly stated by the dissent in Hill, the charge also directed the jury not to consider how parole might apply to the defendant. The majority, however, pointed out that the error concerned whether or how parole might apply to this particular defendant and misled the jury about the general existence of the parole law and good conduct time and that the misstatement about the parole system itself was not cured by that instruction. Id. at 509. Thus, we found egregious harm.
We revisited the same scenario four years later in Taylor v. State, 146 S.W.3d 801 (Tex.App.-Texarkana 2004, pet. ref'd). In Hill, the jury had been improperly instructed that the defendant would be eligible for parole when actual time served plus good conduct time equaled one half the sentence or thirty years, whichever was less. In Taylor, we placed considerable importance on the fact that the instruction, although similar to that in Hill, also contained a statement that parole would be calculated “without consideration of any good conduct time.” See also Villareal v. State, 205 S.W.3d 103, 110 (Tex.App.-Texarkana 2006, pet. dism’d) (focusing to some extent on jury notes concerning parole law); Loun v. State, 273 S.W.3d 406, 417-19 (Tex.App.-Texarkana 2008, no pet.) (error preserved, some harm found).
In Taylor, we also acknowledged that, in Ross v. State,6 the Texas Court of Criminal Appeals had found no harm, even though Ross had objected to an erroneous parole charge — thus requiring reversal if there was some harm. In Ross, a capital case in which the death penalty had been assessed, the erroneous jury instruction contained two paragraphs of improper instructions regarding “good conduct time.” Yet the high court found “no harm.”
With Ross in mind, we concluded in Taylor that the erroneous parole instruction did not affect the very basis of Billy’s sentence, deprive him of an essential right, vitally affect a defensive theory, or make the case for punishment “clearly and significantly more persuasive” — as is required to establish egregious harm. See Taylor, 146 S.W.3d at 810-13; see also Blumenstetter v. State, 135 S.W.3d 234, 240 (Tex.App.-Texarkana 2004, no pet.).
In Igo, like this case, the jury was erroneously instructed that the defendant would not become eligible for parole until the actual time served, plus good conduct time, equaled one-fourth of the sentence imposed, but the jury should have been instructed that he would not become eligible for parole until the actual time served, without considering good conduct time, equaled one-half of the sentence imposed. Igo, 210 S.W.3d at 646. The Texas Court of Criminal Appeals concluded that egregious harm was not shown, noting that the parole instruction contained the standard curative language admonishing the jury not to consider the extent to which the parole law might be applied to the defendant. (That matches the facts in this case.) The court also noted that parole was not mentioned by either counsel during argument on punishment. (That is not like this case: the parole law was discussed to some degree here by both counsel during final arguments.) In Igo, which is also a sexual assault case, the defendant continued his relationship with the underage girl after being indicted, and had attempted to use the same tactics with another girl, and the teacher/student [862]*862relationship involved was pointed out as being especially heinous.
In Hooper, involving a charge of assault on a public servant, the punishment charge contained erroneous language, the parole instructions were not mentioned during argument, the punishment evidence was considered strong (two prior felony convictions for possession of cocaine and possession of cocaine with intent to deliver), there were no jury notes sent out, and the punishment was not extreme. Despite the trial court’s two failed attempts to correct the charge error, thus bringing some attention to the erroneous language, the appellate court found no indication that the jury improperly considered the effect of the parole law on Hooper, thus concluding that harm would be only speculative or theoretical. See Hooper v. State, 255 S.W.Sd 262, 272-73 (Tex.App.-Waco 2008, pet. ref'd).
Here, evidence suggests that the sexual relationship between Stewart and A.N. had ended by the time A.N. was taken to the detention center by her mother. There is no evidence that Stewart either had or attempted to begin a similar relationship •with any other underage person while he was awaiting trial. The evidence shows no violence, no threat (explicit or implicit), and no history of any prior sexual misbehavior by Stewart. It suggests, instead, that Stewart had been the custodial parent for three young daughters of a prior marriage. Stewart’s oldest daughter, a nurse, testified at punishment that Stewart had been a good, although very strict, parent. His second daughter, April, testified similarly at guilt/innocence, but not at punishment. There was no indication of any sexual relationship with any of his natural daughters.
In Igo, the high court noted in its analysis that Igo’s sexual assault was by a teacher against his fifteen-year-old student and that such “conduct was abuse of [Igo’s] position as a teacher.” Igo, 210 S.W.3d at 647. Here, the field of potentially endangered children is much smaller, but the abuse of position is no less a factor. Igo tried to bribe his victim after the fact, while Stewart merely ingratiated himself to his victim. Igo’s sexual contact with his victim was isolated; Stewart’s was extensive, varied, and ongoing for some time. Igo received the maximum sentence of twenty years’ imprisonment; Stewart received a mid-range twelve years. In Igo’s trial, parole law was not mentioned in argument; in Stewart’s, it was mentioned, but in the context of the instruction, not to consider how it might apply to Stewart.
The question before this Court is whether, on these facts, Stewart suffered egregious harm from the erroneous charge language.
The charge error here is troubling, as is mention of parole during argument. On the other hand, appellate courts have been directed by the Texas Court of Criminal Appeals effectively to disregard such errors, under the presumption that the jury will follow the curative language directing the jury not to apply the information to the particular defendant.
Under the stringent standards necessary to show “egregious harm,” we conclude that this error did not deprive Stewart of a fair and impartial trial or affect the very basis of the case, deprive Stewart of a valuable right, or vitally affect a defensive theory. Thus, on review of the entirety of the record, we conclude that egregious harm has not been shown.
(2) Stewart Was Not Harmed by His Attorney’s Failures Regarding the Erroneous Punishment Charge
Stewart next contends that he did not receive constitutionally effective assis[863]*863tance of counsel at trial. That argument is based exclusively on the matters discussed above: counsel’s failure to object to the incorrect charge, and then compounding his error by arguing the incorrect law to the jury at punishment.
This matter was discussed at a hearing on Stewart’s motion for new trial. In relation to this matter, trial counsel focused his attention on a letter in his file in which he warned Stewart about the correct percentage of time that would have to be served before he would become eligible for parole.
The motion itself was not directed at a claim of ineffective assistance of counsel, and the hearing was not held for that purpose.
The standard of testing claims of ineffective assistance of counsel is set out in Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). To prevail on this claim, an appellant must prove by a preponderance of the evidence (1) that his or her counsel’s representation fell below an objective standard of reasonableness and (2) that the deficient performance prejudiced the defense. Id. at 689, 104 S.Ct. 2052; Rosales v. State, 4 S.W.3d 228, 281 (Tex.Crim.App.1999). To meet this burden, the appellant must prove that the attorney’s representation fell below the standard of prevailing professional norms and that there is a reasonable probability that, but for the attorney’s deficiency, the result of the trial would have been different. Tong v. State, 25 S.W.3d 707, 712 (Tex.Crim.App.2000). Under this standard, a claimant must prove that counsel’s representation so undermined the proper functioning of the adversarial process that the trial cannot be relied on as having produced a just result. Strickland, 466 U.S. at 686, 104 S.Ct. 2052.
Stewart complains that counsel was ineffective because he failed to object to an incorrect parole charge, and then compounded his failure by discussing the incorrect sections before the jury.
The ineffectiveness of counsel is a matter that must be firmly founded in the record, and the record must affirmatively demonstrate the alleged ineffectiveness. Smith v. State, 51 S.W.3d 806, 813 (Tex.App.-Texarkana 2001, no pet.). Normally, where an appellate record is silent as to why trial counsel failed to take certain actions, the appellant has failed to rebut the presumption that trial counsel’s decision was in some way (conceivable or not) reasonable. See Mata v. State, 226 S.W.3d 425, 431 (Tex.Crim.App.2007). That concept does not apply here, as counsel clearly was unaware that the charge was erroneous.
In the absence of direct evidence of counsel’s reasons for the challenged conduct, an appellate court will assume a strategic motivation if any can be imagined. Garcia v. State, 57 S.W.3d 436, 441 (Tex.Crim.App.2001). We will not conclude the challenged conduct constitutes deficient performance unless the conduct was so outrageous that no competent attorney would have engaged in it. Id.; see Thompson v. State, 9 S.W.3d 808, 814 (Tex.Crim.App.1999). We can imagine no way in which Stewart would benefit by allowing the jury to believe that he might be able to get out of jail much earlier than he actually could. In this case, we know both what counsel’s inaction was (failure to object) and what his action was (arguing the incorrect law). There is no strategic or tactical motivation that would justify this action or this inaction. It was, simply, a mistake in understanding the application of an overly complicated statute.
While the trial court’s instruction is clearly erroneous, there is nothing in the record that suggests the jury did not fol[864]*864low the trial court’s instruction to not consider how good conduct time might apply to Stewart. Accord Ross, 133 S.W.3d at 624. The jury sent no notes that might otherwise reveal either improper conduct or confusion among the jurors regarding the erroneous instruction. Cf. Shavers, 985 S.W.2d at 292. Under these facts, we do not believe that the record shows there is a reasonable probability that, but for the attorney’s deficiency, the result of the punishment proceeding would have been different, and we are not prepared to speculate to reach such a result. We overrule this contention of error.
(3) Stewart’s Attorney’s Prior Relationships Did Not Color Stewart’s Defense
Stewart finally argues that his trial counsel was ineffective due to an impermissible conflict of interest because counsel was also, at the same time, working as a special prosecutor in a mentor capacity with the prosecutor who tried this case.
In order to prevail, Stewart must show an actual conflict of interest and that the conflict “actually colored” counsel’s actions. Acosta v. State, 233 S.W.3d 349, 356 (Tex.Crim.App.2007).
The evidence at the hearing on motion for new trial was that counsel had acted as a special prosecutor for two cases and that the trials on those eases ended before this trial began.7
Neither of these cases had any apparent connection with this prosecution. Stewart’s argument is that his retained trial counsel had failed to pursue a plea agreement up to the time of trial because he did not want to negatively interact with his former, and possibly his future employer, the district attorney.
The record does not support this argument. Counsel was working on both sides of the docket, but not concerning this case. Even though he had been acting as a mentor for the assistant district attorney who tried this case,8 there is nothing to suggest that defense counsel’s actions were improperly colored by the relationship.
Further, although counsel did state that, in hindsight, he probably should have continued right up to the day of trial to obtain a satisfactory plea bargain, that alone does not suggest that his actions were colored by the relationship.
We affirm the judgment.
Concurring Opinion by Justice CARTER.