OPINION OF THE COURT
JAMES, Judge:Sergeant Williams was convicted by a general court-martial of three rapes, a robbery, sodomy, and an aggravated assault.1 On appeal he contests the propriety of identifications resulting from a photographic line-up, identification of his automobile, admissibility of testimony about blood type analysis by the investigators, the sufficiency of the evidence to sustain the convictions, “future dangerousness” testimony in sentencing, the severity of the resulting sentence, and the convening authority’s qualifications to act on the sentence.
After we first decided this appeal, appellate government counsel moved for reconsideration, bringing to our attention United States v. Stinson, 34 M.J. 233 (C.M.A. 1992).2 Recognizing our responsibility to consider Stinson and to address it, we granted the government’s motion and withdrew our prior decision. On reconsideration we find no merit in the evidentiary issues on the merits, we find the evidence to be sufficient, and we affirm the convictions. We dispose of those issues first. Then we turn to the sentencing evidence, which was the more troubling issue in our earlier decision and which remains troubling. We find error but no prejudice. Finally, we find the sentence not to be inappropriate, and we find no error in the convening authority’s post-trial action.
I. Issues of Evidence on the Merits
Sergeant Williams contested the admissibility and the reliability of the identifications of him by the several victims. To summarize the facts, several women were given rides by an assailant who then forced them at knife-point to submit to the assailant’s sexual demands, took their money, and left them. The assailant deliberately struck one of them with his car. As the investigation came to focus on Williams, Air Force security police investigators had two photographers take black-and-white photographs of the torsos of several men, all dressed the same, and all posed before the same concrete block wall. Regrettably, one photographer shot only the suspect, *814Williams, and the other shot all the other models, who were not suspects. Inevitably the resulting sets of photographs were slightly different in focus, image size, and in shadowing. The victims were separately shown the photographs by police investigators. Each victim picked Williams as the assailant. Each also identified Williams at trial, and each said in various ways that she was certain of her identification. Williams sought to show that the victims (some of whom were prostitutes) had consulted while in jail and on the streets, thus perhaps pooling information. Nonetheless, the military judge overruled his timely objection to admission of the identification testimony and the in-court identifications that followed.3
Line-ups have had a significant constitutional history, but the law on such evidence, including photographic line-ups, now seems settled.4 It is summarized by Mil.R.Evid. 321, which draws no distinctions pertinent in this case between “[a] lineup or other identification process.” In brief, a military accused may suppress an “unlawful” identification. Mil.R.Evid. 321(a)(2). For purposes of this case, an identification is “unlawful” if it results from “a lineup or other identification process” which is “so suggestive as to create a substantial likelihood of misidentification.” Mil.R.Evid. 321(b)(1); see Manson v. Braithwaite, 432 U.S. 98, 97 S.Ct. 2243, 53 L.Ed.2d 140 (1977) (reliability under totality of circumstances, no per se rule); Neil v. Biggers, 409 U.S. 188, 93 S.Ct. 375, 34 L.Ed.2d 401 (1972); Simmons v. United States, 390 U.S. 377, 88 S.Ct. 967, 19 L.Ed.2d 1247 (1968). Biggers and Braithwaite supply factors for that decision. When the military judge at trial applied those factors to her findings of fact and stated her “essential findings,” see Mil. R.Evid. 321(f), she concluded that “the photographic lineup was not suggestive.” Addressing the effect of the process, she further concluded that “there did not exist a substantial likelihood of irreparable misidentification.” Her findings of fact are supported by the record, and we adopt them.
We agree with her result and her conclusion that there was no substantial likelihood of misidentification. The defense argument that differences in the photographs were at least subconsciously suggestive does not carry the point because suggestiveness alone is not fatal. Mil.R.Evid. 321(b)(1). Instead, one must examine the extent and effect of the suggestiveness, as the military judge did. Id. We note that Mil.R.Evid. 321(b)(1) does not test for an “irreparable” misidentification, and it is possible that the judge’s test was less favorable to the defense than the test in Mil.R.Evid. 321(b)(1).5 However, when we *815apply the seemingly more favorable test of Mil.R.Evid. 321(b)(1), we remain confident that the evidence was correctly admitted. See United States v. Quick, 3 M.J. 70 (C.M.A.1977).
Williams also contends that the methods used to identify his vehicle as the one involved in the offenses were suggestive and that the resulting testimony should have been excluded. The police investigators, having narrowed the investigation to Williams, photographed his vehicle and later showed those photographs alone to the victims. Of course, trucks and cars have no constitutional rights, and no special line-up rights are extended to them by Mil.R.Evid. 321. However, the idea that such methods could be suggestive and thereby detract from the reliability of the identification is plausible. Such a result might even be so likely that the resulting testimony is without probative value, or that its probative value is offset by the hazard that it might present. Accordingly, a military judge might exclude it under MiLR.Evid. 402 or 403. The military judge in this case overruled timely objections and admitted the testimony. Her ruling is reviewable only for abuse of discretion, see, e.g., United States v. Jenkins, 27 M.J. 209, 211 (C.M.A.1988), and we find no such abuse. The assignment of error is without merit.6
We reach the same result on Williams’ invitation to review admission of the blood type testimony. The rules give the military judge the discretion to determine whether to admit expert opinions. Mil.R.Evid. 702. We review that decision only for abuse of discretion. United States v. Stark, 30 M.J. 328, 330 (C.M.A. 1990); United States v. Mustafa, 22 M.J. 165, 168 (C.M.A.1986), cert. denied, 479 U.S. 953, 107 S.Ct. 444, 93 L.Ed.2d 392 (1986); United States v. Sawyer, 32 M.J. 917 (A.F.C.M.R.1991). See generally United States v. Gipson, 24 M.J. 246 (C.M.A. 1987) (“the judge has considerable room to exercise ‘judgment’ ”). We find no abuse here, either. Even if we had found error, we note that the defense counsel’s commendable skills were applied here as throughout the trial to limit the value of the evidence to exactly that which it deserved. This evidence was so limited by his efforts that its admission, in the context of the other evidence in the case, was not prejudicial. Article 59(a), UCMJ, 10 U.S.C. § 859(a) (1988).
Having disposed of the appellate issues affecting the evidence on the merits,7 we may now examine the sufficiency of the evidence to convict and to sustain the convictions on appeal. “The test for [legal sufficiency] is whether, considering the evidence in the light most favorable to the prosecution, a reasonable fact-finder could *816have found all the essential elements beyond a reasonable doubt.” United States v. Turner, 25 M.J. 324, 324 (C.M.A.1987) (citing Jackson v. Virginia, 443 U.S. 307, 99 S.Ct. 2781, 61 L.Ed.2d 560 (1979)). The test for factual sufficiency is “whether, [we] are [ourselves] convinced of the accused’s guilt beyond a reasonable doubt.” Turner, 25 M.J. at 325. We apply both in this case, as always. We find no lack of evidence, and we are ourselves convinced of the appellant’s guilt of the charges and specifications of which he was found guilty. Having found no error affecting the findings,8 we now turn to Sergeant Williams’ assignment of error affecting the evidence during sentencing.
II. Sentencing Evidence
During sentencing the prosecution produced a forensic psychiatrist, Colonel William Grant. The military judge found him to be an expert without objection. Colonel Grant testified briefly about the extent of the impact of the offenses upon the victims, and then turned to a lengthy briefing 9 on recidivism. As anyone acquainted with criminal law would expect, the substance was not encouraging. To make matters worse, Colonel Grant described his view that past violent behavior is likely a predictor of futuré violent behavior. With that background, trial counsel asked—
Based on your knowledge, your training, your experience in the field of forensic psychiatry together with the knowledge, information and understanding that you have concerning this case, concerning the evidence, do you have an opinion as to how dangerous this accused is?
The defense objected, “he’s being asked to predict the future and speculate.” The military judge overruled the objection.
Colonel Grant then continued, first pointing out attributes of the offenses of which Sergeant Williams had been convicted that help establish how dangerous Williams might have been at the time, then, and might be in the future. Trial counsel closed by asking the recidivism rates among persons who had participated in rehabilitative programs, and, once again, the reply was not encouraging. Before closing his direct examination, trial counsel offered as exhibits copies of the slides used by Colonel Grant during his briefing on recidivism. The defense asked the military judge to defer the ruling and then cross-examined.
Having been unable to persuade the military judge to exclude the testimony as speculative, the defense sought to diminish its weight. Trial counsel had only one question on redirect:
TC: Once again, Doctor Grant, you would agree that the best predictor of future violence is his past violence?
A: I would.
Trial counsel renewed the offer of the slides about recidivism, the defense objected to them on the basis that the predictions were unreliable and not accepted by the scientific community, but the military judge overruled the objection and admitted the slides.10
*817When we are confronted with new prosecution strategies on sentencing we look first to R.C.M. 1001(b) to determine whether there is a procedural rule governing them, about which a body of interpretative decisions has already developed. Only two possibilities exist for admission of the “future dangerousness” evidence: R.C.M. 1001(b)(4) (aggravation) and (5) (rehabilitative potential).
The “rehabilitation potential” theory is plausible at first glance, and some might say that it gains persuasiveness from United States v. Stinson, 34 M.J. 233 (C.M.A. 1992), to which the government brought our attention upon reconsideration, but we still find it unpersuasive.
A. R.C.M. 1001(b)(5), “Evidence of rehabilitation potential”
The rule seems facially clear: “The trial counsel may present ... evidence, in the form of opinions concerning the accused’s ... potential for rehabilitation.” However, the rule has been amplified by the decisions of the Court of Military Appeals about testimony of commanders and similar authorities in euphemisms that an accused should not be retained in the service, that the court-martial should include a punitive discharge in the sentence.11
In the decisions on “euphemism testimony” there are several themes: that such testimony sometimes too closely resembles command influence, that it often should be excluded because the opinion lacks a rational basis, and that such testimony is excludable as irrelevant. Today’s parallel with the relevance theme becomes immediately apparent in this excerpt from a footnote:
Having rehabilitative potential is a mitigating factor. Lacking rehabilitative potential is not an aggravating factor. [A]n accused should not receive a more severe sentence than otherwise generally warranted by the offense, the circumstances surrounding the offense, his acceptance or lack of acceptance of responsibility for his offense, and his prior record.
United States v. Aurich, 31 M.J. 95, 96-97 n. * (C.M.A.1990) (Cox, J. concurring) (emphasis added). The same thought echoes anew in United States v. Claxton, 32 M.J. 159, 164-65 (C.M.A.1991) (Everett, S.J., concurring), and we emphasized it in United States v. St. Romain, 33 M J. 689, 692 (A.F.C.M.R.1991). We conclude through the same approach that evidence relating to an accused’s likelihood to commit future acts is not relevant unless it shows the accused unlikely to do so. To paraphrase Aurich, a low likelihood of future dangerousness is mitigating, but a high likelihood is not an aggravating factor. Otherwise the risk is that the accused will be sentenced not for his actual crimes but for his possible future badness. While psychiatric assessments of a prisoner’s future dangerousness might be vital to those in corrections who consider parole and similar questions of early release, we hold that “future dangerousness” testimony is not relevant to rehabilitative potential under R.C.M. 1001(b)(5).
In its motion for reconsideration, the government accurately points out that the quoted passage from Aurich is dicta.12 Yet, the quoted passage seemed to us to synthesize the relevance aspect from the decisions about “evidence of rehabilitative potential” and its relationship to sentencing. The reappearance of the theme in Claxton reassures us that the synthesis was intended—at least by the two judges involved—as we have taken it.
*818We must consider Stinson against that background. The facts in Stinson are functionally equivalent to those in Sergeant Williams’ case. However, the issue addressed in Stinson is not admissibility based on relevance—the Aurich idea that we have followed—but is instead the admissibility of the opinion as that of an expert under Mil.R.Evid. 702. Stinson is about a different problem. It is apparently only coincidental that the introductory sentence of the discussion of the Stinson opinions states, seemingly without qualifications, “In a sentencing hearing, an accused’s potential for rehabilitation is a proper subject of testimony by qualified experts.” 34 M.J. at 238.
We are not prepared to read into that discussion any abandonment of the relevance theme from the Aurich note. It is true that the Court of Military Appeals did not choose to write about the introduction of evidence of a lack of rehabilitative potential, but we cannot find law in silence. This is not a simple matter: Reasonable persons can easily disagree on whether the ambiguous “fairly poor prognosis” and “high risk of reoffense” in Stinson and the generalized discussion of recidivism and violence in Sergeant Williams’ case contribute to an understanding of the offender’s present capacity to be rehabilitated, in contrast to that of the hypothetical average offender and in contrast to the need or lack of need to disable the offender. Those critical aspects are not considered in Stinson. Thus, we do not find Stinson to permit introduction of the evidence that was admitted against Sergeant Williams, over objection.
B. R.C.M. 1001(b)(4), “Evidence in aggravation”
Now we turn to the remaining possibility, that “future dangerousness” testimony is admissible as aggravation under R.C.M. 1001(b)(4). It would seem very unlikely to be so, for the rule and resulting decisions clearly limit aggravation to matters “directly relating to or resulting from the offenses of which the accused has been found guilty.” R.C.M. 1001(b)(4) (emphasis added); United States v. Bartoletti, 32 M.J. 419 (C.M.A.1991); United States v. Gordon, 31 M.J. 30 (C.M.A.1990).
However, there is some precedent that might seem to permit an enlargement of that limitation to include “state-of-mind/ depth-of-problem” evidence. See, e.g., United States v. Ciulla, 32 M.J. 186 (C.M.A.1991), cert. denied, — U.S. -, 112 S.Ct. 172, 116 L.Ed.2d 135 (1991); United States v. Mullens, 29 M.J. 398 (C.M.A.1990); United States v. Silva, 21 M.J. 336 (C.M.A.1986). But cf. United States v. Martin, 20 M.J. 227 (C.M.A.1985), cert. denied, 479 U.S. 917, 107 S.Ct. 323, 93 L.Ed.2d 295 (1986). It might be argued from a glance at those cases that an expert opinion establishing a propensity for violence, based on attributes of the established crimes, shows the depth of the accused’s recent propensity for violence and his willingness to inflict fear and pain. However, Ciulla, Mullens, and Silva all involved evidence that was somehow factually related by events in history to the established crimes. In Silva, it was statements during the crimes that the accused had committed prior such crimes, and in Ciulla it was a statement that the accused wanted to do similar crimes to a different victim, apparently made in both cases to induce the child victims to acquiesce. In Mullens, the acts were uncharged parts of the same course of behavior. Thus, in all three cases there is some demonstrable, concrete linkage between the convictions and the evidence showing the uncharged acts to be “directly relating” to the established crimes.
Despite the expert’s reliance on concrete attributes of the established crimes to analyze Sergeant Williams’ future dangerousness, the “future dangerousness” testimony lacks that directness required by R.C.M. 1001(b)(4) because there is no act yet which can be shown “directly relating” to the past crimes.13 That was apparently *819the point of the defense objection that the testimony was speculative: The future acts that would show continuing violent behavior have not yet occurred. There is no directness, and therefore the “future dangerousness” testimony has no connection with the crimes of which Sergeant Williams was convicted. Similarly, the witness’ general discussion of recidivism had no direct connection with Sergeant Williams. Colonel Grant’s testimony about recidivism and future dangerousness was not admissible under R.C.M. 1001(b)(4) or (5).
C. Is R.C.M. 1001(b) Exclusive?
We have exhausted the possibilities under R.C.M. 1001(b). Is it exclusive? Future dangerousness certainly has obvious logical appeal on the sentencing question, despite the analysis above. It is not unconstitutional, impossible, or even unusual for a state to provide that it will be or may be considered in sentencing, including capital sentencing. Barefoot v. Estelle, 463 U.S. 880, 896-903, 103 S.Ct. 3383, 3396-3400, 77 L.Ed.2d 1090 (1982); Estelle v. Smith, 451 U.S. 454, 472-73, 101 S.Ct. 1866, 1878, 68 L.Ed.2d 359 (1981); Jurek v. Texas, 428 U.S. 262, 274-76, 96 S.Ct. 2950, 2957-58, 49 L.Ed.2d 929 (1976). The maximum imposable confinement is set by statute and Executive prescription (subject to precedent on multiplicity). Article 56, UCMJ, 10 U.S.C. § 856 (1988); R.C.M. 1003(c). It makes sense that, within that period, one who is likely to be dangerous again should be disabled from public misconduct for a longer term than one who isn’t.14 The answer is simple: Rule-making in this area is an Executive prerogative, Articles 36(a) & 56, UCMJ, 10 U.S.C. § 836(a) & 856 (1988), into which we have no warrant to intrude in this case. Though the President provided for admission of some facts from which one might infer amenability to discipline and the likelihood of future misconduct, e.g. R.C.M. 1001(b)(l)-(3), he has not provided for consideration of future dangerousness testimony like that in this case. It should have been excluded.
D. Prejudice?
We must now determine whether the error resulted in any prejudice to Sergeant Williams. Once again, the defense did a commendable job at limiting the effectiveness of the evidence. During cross-examination, the defense probed the science underlying the witness’ testimony, the completeness of his inquiry into Sergeant Williams’ case, and other factors to diminish the impact of Colonel Grant’s briefing on recidivism. Having considered this carefully, we are persuaded that the resulting sentence in this case shows the military judge limited her consideration of this testimony to that which its diminished weight deserved. If Colonel Grant’s testimony is reflected in the sentence adjudged, it is because he pointed out the aspects of the criminal acts actually committed, and their actual impacts on the victims, not because of his statistical analysis and predictions from them.
E. Sentence Appropriateness
Finally, Sergeant Williams urges that the sentence is excessive. He was sentenced to be discharged from the service with a dishonorable discharge, to be confined for 20 years, forfeiture of all pay and allowances, and to be reduced to E-l. The convening authority approved the sentence as adjudged. We have considered the record and testimony relating to his good service and to his domestic life, and the other information in his behalf. However, considering the violence actually committed and that threatened in these offenses, we have no difficulty concluding that the sentence is not inappropriate.
*820III. Post-Trial Procedure
The results of a court-martial are examined by the commander who convened the court-martial, and that commander has the responsibility for acting on the sentence, approving it as adjudged, reducing it, or disapproving it altogether. Article 60, UCMJ, 10 U.S.C. § 860 (1988); R.C.M. 1107. The accused may submit “matters for consideration” to the convening authority, typically asking clemency. Article 60(b), UCMJ, 10 U.S.C. § 860(b) (1988); R.C.M. 1105. Sergeant Williams complains that the convening authority who acted on his case was himself suspected of sexual misconduct and was therefore disqualified. But see R.C.M. 1107(a) Discussion (examples of disqualifications). In an unrelated case involving the same commander, we set aside the convening authority’s action in an abundance of caution and remanded the case for a new action by a different convening authority. See generally United States v. Kroop, 34 M.J. 628 (A.F.C.M.R.1992). Sergeant Williams asks the same relief.
We fashioned relief in Kroop to recognize the effect of any “psychological baggage” that might have been borne by a convening authority whose own misconduct was similar to that involved in a case under his review. We have never been shown that the misconduct of which the commander was suspected included violent behavior like that of which Sergeant Williams is convicted, and so the cases are qualitatively different. We find no need for such relief in this case.15
The findings of guilty and the sentence are correct in law and fact and, on the basis of the entire record, are
AFFIRMED.
Senior Judge LEONARD and Judge RIYES concur.
. The convictions are for violations of Articles 120, 122, 125 and 128, UCMJ, 10 U.S.C. § 920, 922, 925, and 928 (1988), respectively.
. We commend appellate counsel’s diligence in recognizing that Stinson might apply in this case and in recognizing that it was too recently decided to have been considered in our prior opinion. This is a classic example of correct use of a motion for reconsideration. The importance of counsel’s decision is not diminished by the fact that we have found Stinson unhelpful in the resolution of this appeal. As counsel recognized, the result might have been otherwise.
. The issue was litigated thoroughly at trial. We note for the benefit of other practitioners this military judge’s unhappy experience. Required by Mil.R.Evid. 321(f) to resolve the matter before pleas, the military judge began to take testimony and evidence in what would have been an Article 39(a) session without members had the accused not already elected trial by the military judge alone. Many of the witnesses who testified on the objection also were needed on the merits. Accordingly, counsel asked the military judge to consider the testimony taken on the merits when it became appropriate, without the need to recall a witness simply to give the same testimony again, to accommodate the parties and the witnesses. When the military judge agreed to this and similar requests to hear testimony out of the usual sequence, she was rewarded with repeated controversies about which party was a proponent of the witness at any given moment, who had the witness on direct or cross-examination, and whether the party had the privilege to lead. It is likely that the military judge came to regret having diminished her control and the orderliness of the proceeding for what turned out to be an ephemeral promise of convenience to others.
. See generally annotation, “Admissibility of Evidence of Photographic Identification as Affected by Allegedly Suggestive Identification Procedures,” 39 A.L.R.3d 1000 (1971). This case requires only application of the reliability analysis because there is no issue about a right to counsel at the “line-ups.” There is no right to have defense counsel present when a victim or witness is shown photographs. United States v. Ash, 413 U.S. 300, 93 S.Ct. 2568, 37 L.Ed.2d 619 (1973) (pre-indictment). Accord Mil.R.Evid. 321(b)(2); United States v. Quick, 3 M.J. 70, 71 (C.M.A.1977).
. Cf. Braithwaite, 432 U.S. at 116, 97 S.Ct. at 2254; Biggers, 409 U.S. at 198, 93 S.Ct. at 381; Simmons, 390 U.S. at 384, 88 S.Ct. at 971; Quick, 3 MJ. 70; Manual For Courts-Martial, United States, 1984, p. A22-30 (drafters’ analysis accompanying Mil.R.Evid. 321(b)(1)).
. We note that this assignment was stated jointly with the assignment relating to the line-up. That was perhaps a forensic tactic by which appellate defense counsel sought to improve the odds, but it violates Courts of Military Review Rules of Practice and Procedure, rule 15.a. ("setting forth separately each error asserted"), 22 MJ. CXXVII, CXXXII; accord, Int. Rule 5-5.a. This case illustrates one reason for that requirement: Different errors require differing analysis and may be measured by differing standards of review based on differing scopes of review. To join them simply confuses the matter unnecessarily, and that is impermissible, despite any forensic advantage.
. We have not overlooked trial counsel’s interference at trial with the defense counsel’s efforts to do his job. However we might characterize trial counsel’s behavior, it was exacerbated by the confusion that followed when the military judge agreed to consider testimony taken on the line-up also on the merits, and trial counsel was no doubt frustrated by the defense counsel’s continuation of discovery at trial. The civilian defense counsel did the kind of probing that we would expect of a practitioner whose home jurisdiction lacks our generous discovery, but in most instances it was also the kind of detailed cross-examination that the defense must sometimes do to expose inconsistencies and other flaws. Trial counsel was clearly impatient with the length of the defense examinations, but the prosecution objections were more frequent than sustainable. Much too often counsel stated "we don't see the relevance of that.” The trial judge usually did, and we easily do, with the benefit of the orderliness of a transcript. Fortunately, the military judge usually ruled in favor of the defense, and in those instances adverse to the defense, her rulings were on matters within her discretion. We have seen no abuse. This is not the kind of “outrageous” record that was seen in United States v. Stroup, 24 M.J. 760 (A.F.C.M.R. 1987), rev’d, 29 M.J. 224 (C.M.A.1989).
. However, we note that the military judge omitted to state findings as to some of the charges, in contrast to the specifications. The affected charges included specifications as to which appellant was convicted without exceptions or substitutions, and so findings of guilty are necessarily implied as to the associated charges. The error is minor, United States v. Logan, 15 M.J. 1084 (A.F.C.M.R.1983), pet. denied, 17 M.J. 36 (C.M.A.1983) (citing cases); United States v. Hathaway, 1 C.M.R. 776 (A.F.B.R.1951), and no prejudice resulted, United States v. Timmerman, 28 M.J. 531 (A.F.C.M.R. 1989), pet. denied, 28 M.J. 356 (C.M.A.1989). The promulgating order anticipated our analysis and would need no correction except that, as appellate defense counsel properly note, the "constructive finding” to be attributed to Charge IV was guilty. The order is incorrect in that respect, but the error is minor and has been corrected by hand.
. At one point the witness was allowed to narrate for over four pages of single-space transcript, uninterrupted by anyone or anything. Permitting a witness to narrate during direct testimony may deprive the opponent’s lawyer of the opportunity to anticipate remarks the admissibility of which ought to be contested.
. Our disposition does not turn on the expertise involved in the testimony. A military judge has discretion when determining whether to admit expert opinions, and we review such decisions only for abuse of that discretion. See, e.g., Stinson, 34 M.J. at 238.
. See generally United States v. Pompey, 33 M.J. 266 (C.M.A.1991); United States v. Claxton, 32 M.J. 159 (C.M.A.1991); United States v. Corraine, 31 M.J. 102 (C.M.A.1990); United States v. Aurich, 31 M.J. 95 (C.M.A.1990); United States v. Wilson, 31 M.J. 91 (C.M.A.1990); United States v. Kirk, 31 M.J. 84 (C.M.A.1990); United States v. Cherry, 31 M.J. 1 (C.M.A.1990); United States v. Gunter, 29 M.J. 140 (C.M.A.1989); United States v. Ohrt, 28 M.J. 301 (C.M.A.1989); United States v. Homer, 22 M.J. 294 (C.M.A. 1986). We would be remiss if we did not recognize that scattered statements in the euphemism cases seem to support admission of testimony like that we address here.
. Indeed, it is the separate opinion of two judges (one now inactive), added to a per curiam decision in which all three judges wrote or joined in separate opinions.
. We emphasize that Colonel Grant’s identification and discussion of attributes of the crimes showing a then-existing willingness to do violent acts do not constitute the error. Indeed, we found those remarks helpful when considering *819the quality of the behavior for which Sergeant Williams was convicted.
. The civilian Federal Sentencing Guidelines declare, ‘To protect the public from further crimes of the particular defendant, the likelihood of recidivism and future criminal behavior must be considered.” United States Sentencing Commission, Guidelines Manual ch. 4, pt. A, intro, comment (Nov. 1990). However, the implementation of that idea is strictly objective and empirical, using the point system that characterizes the Guidelines to score past criminal history of the offender. We found nothing in the Guidelines providing for expert projections about future dangerousness.
. When deciding what action to take on a case, a convening authority has the recommendations of his staff judge advocate. R.C.M. 1106, 1107(b)(3)(A)(ii). The staff judge advocate’s advice in this case included some mistaken advice. Using a standard form designed for the purpose, the staff judge advocate completed a block entitled “maximum imposable sentence for offenses convicted upon" (Air Force Form 242 (January 1985), item 34) by entering “as adjudged.” The sentence adjudged included a dishonorable discharge and 20 years, but the maximum which might have been adjudged was (because of the violations of Article 120, UCMJ, 10 U.S.C. § 920 (1988)) arguably death and assuredly imprisonment for life. This is the second time that we have noted that the legend for the block is ambiguous. See generally United States v. Wilson, 33 M.J. 512, 513 n. 2 (A.F.C.M.R.1991) (“imposable” probably means after consideration of multiplicity). We assume that "maximum imposable sentence” refers to the maximum which might be adjudged at trial, not to the maximum which could be approved by the convening authority on review. Were it not so, the entry would always be “as adjudged,” for no more than was adjudged may ever be approved. R.C.M. 1107(d). This error was waived by the omission of the defense to note it when replying to the staff judge advocate’s recommendations, R.C.M. 1106(f)(6). We note that it could not have had any effect adverse to appellant. However, the recurrence of such an error suggests that practitioners need some help with the form.