On February 21, 2002, after a five-day trial, a jury convicted Juan Santiago-Vazquez (“Santiago”) on one count of conspiracy to distribute controlled substances, 21 U.S.C. § 846 (2000). He was thereafter sentenced to life imprisonment. He now appeals, claiming ineffective assistance of counsel, trial errors and errors at sentenc*509ing. He has supplemented his counsel’s brief with his own.
From the evidence presented at trial, the jury could have found the following. Between 1990 and 1993, Santiago ran a drug point in the Manuel A. Perez housing projects in Puerto Rico. Santiago’s organization primarily sold crack — 1.5 to 2 kilograms per month, and over five kilograms in 1992 alone — but also dabbled in cocaine, heroin and marijuana. Santiago supervised at least ten other individuals in the operation of his drug point.
Starting in 1991, Santiago’s organization found itself at war with rival drug dealers in the area. Over the course of the conflict — which lasted through 1993 — Santiago allied himself with another drug dealer named Victor Negrón. Negrón and Santiago loaned firearms to one another for protection, sold drugs to one another, and coordinated manhunts to kill the supporters of rival gangs.
Santiago planned and sanctioned the participation of his subordinates in these manhunts, which resulted in the deaths of at least five rivals. The firearms used in these killings, as well as in the day-to-day operations of Santiago’s drug enterprise, included pistols, submachine guns, shotguns and rifles. Santiago was arrested with one such weapon in 1993, and a stock of similar weapons was seized from an apartment at which Santiago’s main lieutenant was present.
At trial much of the government’s testimony came from three cooperating witnesses: Alexander Cruz-Rojas, who had run a drug point near Santiago’s and had fought against Santiago in the drug war; Ramon Cesareo-Soto, who worked in Negron’s organization, was closely allied with Santiago in the drug war, and frequented Santiago’s drug point; and Negrón himself. Given their professed roles in drug dealings and drug wars, their testimony was very harmful to Santiago.
We deal first with Santiago’s myriad claims of trial error, all of which are fact-based and none of which entails novel or complex issues of law. The claims are, among others, that defense counsel failed to object to leading questions, permitted introduction of prejudicial evidence relating to events outside the relevant time frame or not sufficiently connected to the defendant, permitted the jury to learn that other defendants had pled guilty, allowed answers from witnesses without personal knowledge of the events, and permitted expert testimony from a witness not qualified as an expert.
Virtually none of these claims, many made in a sentence or two or at most a paragraph, is adequately developed on appeal, which is alone fatal. See United States v. Zannino, 895 F.2d 1, 17 (1st Cir.), cert. denied, 494 U.S. 1082, 110 S.Ct. 1814, 108 L.Ed.2d 944 (1990). In most cases, it appears that no objection was made in the district court, so review would in any event be only for plain error, but no effort is made on appeal to show that any such error likely altered the outcome of the trial given the other evidence against the defendant. United States v. Olano, 507 U.S. 725, 732, 113 S.Ct. 1770, 123 L.Ed.2d 508 (1993).
Santiago says that various trial errors insufficient in themselves may cumulate so as to undermine the fairness of a trial and make out a due process violation. See United States v. Sepulveda, 15 F.3d 1161, 1195-96 (1st Cir.1993), cert. denied, 512 U.S. 1223, 114 S.Ct. 2714, 129 L.Ed.2d 840 (1994); see also United States v. Meserve, 271 F.3d 314, 332 (1st Cir.2001). However, the case against Santiago was a strong one, uncontradicted by any witnesses for the defense; and in any event the brief on appeal fañs to show that there were serious errors, let alone a prejudicial accumulation.
*510The government has taken the trouble to address the individual claims of error one by one. If any one of them was properly preserved in the district court and appeared to have a clear potential to alter the outcome of the trial, we would discuss it. But none falls into this category. Given their sketchy development on appeal we think it is enough to say this and no more about the individual trial error claims.
Santiago’s next challenge (made in his pro se brief) is to the court’s special verdict form. The form, a modified version of which had been requested by defense counsel, asked jurors three questions: first, whether Santiago was guilty of the conspiracy to distribute drugs charged in the indictment; second, whether more than a specific quantity of each of four named drugs was involved; and third, whether Santiago had conspired to commit murder over the course of the conspiracy. The drug quantities contained in the second question had been charged in the indictment against Santiago and were included in response to a defense argument invoking Apprendi v. New Jersey, 530 U.S. 466, 120 S.Ct. 2348, 147 L.Ed.2d 435 (2000). The murders mentioned in the third question were not charged in the indictment, but could have resulted in a sentencing enhancement under U.S.S.G. § 2D1.1(d)(1); defense counsel therefore sought a separate jury determination regarding the murders on the theory (loosely stated by counsel) that Apprendi applied to that particular enhancement under the guidelines.
Santiago — -this is his own argument, not appellate counsel’s — says that the second and third questions both pointed the jury in the direction of a life sentence, so that either way he lost. That either alternative sufficed to raise the sentencing range is true, but that is a function of the guideline in question. See U.S.S.G. § 2D1.1. Santiago would not have “lost” if the jury had answered in the negative on both questions; unfortunately for Santiago, the jury found that the government had proven its drug quantities beyond a reasonable doubt.
Alternatively, Santiago complains that the second question did not allow the jury to find lesser drug quantities than those specified. This misunderstands the function of the question, which was simply to determine in accordance with Apprendi whether facts charged in the indictment (and raising the statutory maximum) had been proven to a jury. If the jury concluded that only lesser amounts had been proven, the answer would have been negative and the precise lesser quantity irrelevant to the issue that Apprendi reserved for the jury.
Santiago’s final challenge to the special verdict form claims that the mention of murder in the form unfairly prejudiced the jury against him — particularly because no murders were charged in his indictment. Leaving aside the fact that the special verdict form — in particular the question about murders — was requested by defense counsel and opposed by the government, we note that there had already been ample (and appropriate) testimony about Santiago’s involvement in the murders over the course of the trial. The jury would have already known of Santiago’s alleged involvement in the killings, and that knowledge would not have disappeared once deliberations began. If the additional mention of drug murders in the special verdict form affected the jury’s deliberations at all, it could not have tainted them so effectively as to amount to a denial of due process to Santiago.
Santiago also contests two of his supervised release conditions: (1) that he submit to drug tests at his probation officer’s discretion; and (2) that he participate in a substance abuse program at the officer’s *511discretion if those tests are positive. Santiago did not object to either condition at sentencing, but the government concedes that both conditions are plain error under United States v. Melendez-Santana, 853 F.3d 93 (1st Cir.2003). Melendez said that both conditions were error but, for reasons not relevant here, the court was required to decide only whether one of them was “plain error” and made no ruling as to whether the other would be so classified.
Two judges of this court have recently questioned whether Melendez ’ plain error analysis is correct, United States v. Padilla, 393 F.3d 256, 259-61 (1st Cir.2004) (Campbell and Selya, JJ., concurring in the judgment), and we have recently granted rehearing en banc in that case to assess the continued viability of the Melendez rationale. But unless and until the en banc court holds to the contrary, Melendez’ analysis controls. The government concedes that both delegations in this case are inconsistent with Melendez and does not seek to distinguish between them as to “plainness.” Under these circumstances, we accept the government’s concession without further inquiry.
Santiago’s remaining challenges to his sentence do not fare as well. Santiago claims that both of his sentencing enhancements — two levels for possession of a firearm in connection with the offense, U.S.S.G. § 2D1.1(b)(1), and four levels for a leadership role in the offense, U.S.S.G. § 3B1.1(a)—violate Blakely v. Washington,—U.S.-, 124 S.Ct. 2531, 159 L.Ed.2d 403 (2004), as do the guidelines as a whole. Although Santiago relied upon Apprendi in seeking a special verdict form, he got such a form (even if not exactly the one he sought) and he thereafter made no further attack on the guidelines at sentencing. We thus review for plain error.2
Following the Supreme Court’s decision in United States v. Booker,—U.S.-, 125 S.Ct. 738, 160 L.Ed.2d 621 (2005), and this circuit’s decision in United States v. Antonakopoulos, 399 F.3d 68 (1st Cir.2005), the district court’s use of mandatory guidelines at sentencing constitutes “error” that is “plain.” Nonetheless, Santiago must still demonstrate prejudice and fundamental unfairness. See Olano, 507 U.S. at 732-36, 113 S.Ct. 1770; Antonakopoulos, 399 F.3d at 77. In particular, Santiago must show that there is a reasonable probability that he would have been sentenced differently but for the error. “The burden is on the defendant to convince us on specific facts” that this is so. Id. at 80.
Santiago has been afforded the opportunity to point to such facts in supplemental briefing, but has failed to do so. He notes that the district court stated that its sentence and enhancements were “pursuant to the guidelines,” and that it relied primarily (or perhaps exclusively) on the presentence report to support its enhancements. But the former is not sufficient to suggest that the district court would have sentenced him differently under a discretionary post-Booker regime; and the latter remains a valid exercise of the district court’s authority even after Booker. Santiago has pointed to nothing exceptional or sympathetic about his case that would give us reason to suspect that the district court would have sentenced him more leniently had it been empowered to do so, and without more he has failed to meet the requirements of Antonakopoulos for remand and resentencing.
There remains the claim that Santiago’s trial counsel was ineffective. Such claims cannot be raised on direct appeal, save in the rare case where the error claimed can *512be considered without further development of the record — a step usually necessary to determine why counsel took the actions challenged as incompetent. See United States v. Martinez-Vargas, 321 F.3d 245, 251 (1st Cir.2003). This case falls squarely within the general rule and we decline to reach the ineffective assistance claim, without prejudice to a section 2255 motion.
We affirm Santiago’s conviction and sentence save that we vacate the drug treatment and drug testing portions of Santiago’s supervised release conditions, and remand for resentencing as to those conditions.
It is so ordered.
. Santiago claims that his allocution at sentencing, in which he denied having dealt drugs with his co-conspirators, is enough to preserve his Blakely claim on the leadership enhancement. We disagree.