HAMILTON, Circuit Judge.

Mark Bozovich was convicted of conspiracy to distribute heroin. He now appeals, seeking a new trial or at least a lower sentence. Bozovich argues that he is entitled to a new trial on the theory that the district court erred by allowing the government to cross-examine him well beyond the scope of his direct testimony. He also argues that his 235-month prison sentence was based on an erroneous drug quantity finding. We affirm both the conviction and the sentence.

I. Rule 611(b) and the Scope of Cross-Examination

Bozovich testified in his own defense at his trial about his criminal record and his heroin addiction. This direct testimony was intended to show that Bozovich was an addict, not a conspirator in heroin distribution. The government then cross-examined. After some preliminary questions about Bozovich's employment and earnings history, the cross-examination homed in on a statement Bozovich had made to a pair of DEA agents before he was arrested on the conspiracy charge being considered here. In that statement he had told DEA agents about who supplied him and his associates with heroin.

Bozovich's lawyer objected to the questioning about the statement, asserting that it was beyond the scope of direct examination and hence impermissible under Federal Rule of Evidence 611(b). The district judge overruled the objection and the government proceeded with its questioning. Bozovich admitted most of the contents of the statement, in which he identified a number of people who supplied heroin to him and others. On re-direct, defense counsel tried to establish that while Bozovich sometimes shared his heroin with friends who were sick from withdrawal, he did not participate in a conspiracy to distribute heroin.

Rule 611 governs the mode and order of examining witnesses, and it gives broad discretion to the district judge to manage the process to promote determination of the truth, to avoid wasting time, and to protect witnesses from harassment or undue embarrassment. Rule 611(b) provides more specifically: "Cross-examination should not go beyond the subject matter of the direct examination and matters affecting the witness's credibility." The standard under Rule 611(b) is whether the cross-examination was "reasonably related to the subject matter of direct examination." United States v. Harbour, 809 F.2d 384, 388 (7th Cir.1987). Determining the "subject matter" of the direct examination is not an exact science, and "both the United States Supreme Court and our court have liberally interpreted the extent of the defendant's direct examination for purposes of establishing the proper scope of the cross-examination," id. at 388-89 (brackets omitted), quoting United States v. Green, 757 F.2d 116, 120 (7th Cir.1985).

Our standard of review on appeal is the deferential "abuse of discretion" standard, United States v. Carter, 910 F.2d 1524, 1530 (7th Cir.1990), which requires us to keep in mind the trial judge's more immediate feel for the case and the fact that the judge ordinarily must rule on the question without full knowledge of what cross-examination is likely to show. The deferential standard of review under Rule 611(b) is consistent with our cases emphasizing that "management of cross-examination is peculiarly committed to the district court's discretion." United States v. Studley, 892 F.2d 518, 529 (7th Cir.1989), quoting United States v. Castro, 788 F.2d 1240, 1244 (7th Cir.1986) (internal quotation marks omitted). The district court here did not abuse its discretion under Rule 611(b).

The defense theory of this case was clear. Counsel for Bozovich began his opening statement: "I represent Mark Bozovich; Mark Bozovich, heroin addict." He quickly conceded that Bozovich bought heroin, used heroin, and sometimes even sold heroin. He just as quickly denied, however, that Bozovich conspired to distribute heroin. Defense counsel ended his opening statement by posing these rhetorical questions to the jury: "Was it a conspiracy? Was it really? Or was it just a bunch of people getting high together?"

The direct examination of Bozovich by his counsel, in particular the questioning about his drug use, advanced this theory. Bozovich testified about how long he had been a heroin addict — approximately five or six years — and how expensive his heroin addiction had been at its height — approximately $100 a day. (At sentencing the district judge took $100 to be the price of a gram of heroin.) Bozovich described his multiple attempts at recovery followed by relapse. The direct examination concluded:

On appeal, Bozovich frames the scope of direct testimony narrowly as his heroin addiction, so that "the only proper cross-examination would have been for the Government to try and prove that Bozovich was not addicted to heroin." In our view, though, it was not an abuse of discretion for the district judge to view the scope of the direct examination more broadly as Bozovich's heroin use, including his suppliers and his ability to pay for the heroin over the years in question. Those were the principal subjects of the cross-examination. By testifying on direct about his heroin purchasing habits and the motives for his purchases, Bozovich "opened himself up for cross-examination" as to those topics. See Harbour, 809 F.2d at 389.

According to the government, Bozovich had admitted in his statement to DEA agents to buying heroin from several suppliers, buying heroin in quantities much larger than $100 a day, and brokering drug deals among his associates. On cross-examination Bozovich accused the agents of lying about some aspects of his statement, but the accuracy of different versions of events is for the jury to decide. It is enough to withstand scrutiny under Rule 611(b) that the district judge could reasonably treat these subjects as "matters reasonably related to the subject matter of direct examination." Id. at 388.

II. Drug Quantity for Sentencing

Bozovich received a 235-month prison sentence, the low end of the 235- to 240-month guideline range the court calculated for the offense. The sentencing range in this case, as in most drug cases, was driven primarily by estimating the quantity of drugs for which the defendant should be held responsible. The district judge determined that Bozovich was responsible for conspiring to distribute between one and three kilograms of heroin, which produced a base offense level of 32. Bozovich argued instead that he was responsible for between 400 and 700 grams, which would have produced a base offense level of 28. If the right answer were somewhere in the middle — between 700 grams and one kilogram — the base offense level would have been 30.1

Bozovich's offense level was raised by four levels (two for possession of a weapon and two for obstruction of justice), and he was in criminal history category III. Lowering his base offense level by four levels would have reduced the low end of his range to 151 months, or about one-third. A two-level reduction would have reduced the low end of the range to 188 months, or exactly one-fifth.

A convicted defendant has a "due process right to be sentenced on the basis of accurate information." Ben-Yisrayl v. Buss, 540 F.3d 542, 554 (7th Cir.2008), citing United States v. Tucker, 404 U.S. 443, 447, 92 S.Ct. 589, 30 L.Ed.2d 592 (1972), and Townsend v. Burke, 334 U.S. 736, 741, 68 S.Ct. 1252, 92 L.Ed. 1690 (1948). In applying that general principle, however, it is "well-established that a preponderance of the evidence is all that is required for a factual finding of drug quantity under the Sentencing Guidelines, due process concerns notwithstanding." United States v. Medina, 728 F.3d 701, 705 (7th Cir.2013). Determining drug quantities under the Sentencing Guidelines is often difficult, and district courts may make reasonable though imprecise estimates based on information that has indicia of reliability. See, e.g., United States v. Hollins, 498 F.3d 622, 631 (7th Cir.2007) ("[T]he sentencing guidelines permit some amount of reasoned speculation and reasonable estimation by a sentencing court.") (internal quotation marks and citation omitted); United States v. Rodriguez, 67 F.3d 1312, 1325 (7th Cir.1995) ("Recognizing that drug dealers ordinarily do not use invoices and bills of lading, we have held that sentencing courts may make reasonable estimates as to drug quantities.").

Our standard of review on appeal is clear error, which is a "highly deferential standard of review." United States v. Hankton, 432 F.3d 779, 789 (7th Cir.2005). At the same time, while "a district court does not automatically commit clear error when it fails to use the most conservative calculation possible," United States v. Longstreet, 567 F.3d 911, 929 (7th Cir. 2009), a district court choosing among "plausible estimates of drug quantity" should normally "err on the side of caution," United States v. Beler, 20 F.3d 1428, 1436 (7th Cir.1994), quoting United States v. Walton, 908 F.2d 1289, 1302 (6th Cir. 1990).

For guideline purposes, the drug quantity attributable to Bozovich can be approached in terms of what he purchased or what he sold and what he used. Normally we would expect the first amount to equal the second. Lacking records that document those amounts, the district court had to do its best to estimate. Here the district judge drew on testimony about how much heroin Bozovich sold and testimony about how much heroin he purchased to arrive at two independent estimates of drug quantity. Both estimates are imperfect, as are the ways they were calculated. But in this endeavor, precision is not be expected or required. What matters is that both of these estimates were conservative and both were over one kilogram, the drug quantity the district court used to calculate the guideline range. This "extremely conservative finding" by the district court does not constitute clear error.

One estimate of the drug quantity attributable to Bozovich focused on his sales of heroin. This estimate would necessarily underestimate the drug quantity attributable to Bozovich, since under the guidelines participants in a drug distribution conspiracy are held responsible for amounts they use as well as amounts they sell. United States v. Wyss, 147 F.3d 631, 632 (7th Cir.1998).2 Using this estimate is thus one way in which the district judge used conservative factors in in his drug quantity finding.

The judge also used conservative assumptions to estimate how much heroin Bozovich sold. One witness estimated that Bozovich sold 40 grams of heroin per week. Two other witnesses also testified about his sales of heroin, one saying that he observed these sales and the other saying that he bought heroin from Bozovich. Bozovich admitted that he frequently acquired heroin in large quantities, which suggests an intent to resell. He also told the DEA agents that his acquisitions of heroin, and hence his potential resales, spanned a six-year period, or at least so testified one of the DEA agents. Bozovich testified that he told them three to six years, probably closer to three. Ultimately the district judge found that the conspiracy lasted six years, and we have no basis to upset that finding. Based on all of that testimony, the district court calculated that even the conservative assumptions that Bozovich sold just 20 grams of heroin per week for just 52 weeks of the six-year conspiracy would make Bozovich responsible for at least 1,040 grams of heroin.

The second estimate of the drug quantity attributable to Bozovich focused on his purchases of heroin. Bozovich told the DEA agents that sometimes he would buy up to $300 of heroin a day from just one of his suppliers. Another of his suppliers testified that Bozovich would buy approximately five grams of heroin every four or five days. And, again, the conspiracy persisted for six years. Based on all of that testimony, the district court conservatively assumed that Bozovich purchased an average of just four grams of heroin per week for five years of the six-year conspiracy. This estimate also holds Bozovich responsible for at least 1,040 grams of heroin.

Bozovich attacks these estimates in two main ways. First, he derides much of the testimony on which these estimates are based as "what junkies said about other junkies." That is one reason why estimates should err on the low side, and the district judge heeded that caution. Also the jury must have credited some of the government witnesses in finding Bozovich guilty beyond a reasonable doubt. The jury may well have done so for the same reason the judge did: "their testimony is generally consistent in that each establish[ed] that Bozovich was purchasing large quantities of heroin over an extended period of time and that some portion of that heroin was being sold."

Second, Bozovich maintains that these estimates must take account of the time he spent in drug treatment during the conspiracy. Even if the district judge had not taken account of this time in treatment, that would not constitute clear error. Bozovich admitted that he was not free of heroin during his attempts at rehabilitation. In any event, though, the assumptions the district judge used in both of his estimates were more than conservative enough to allow for this factor.

The case Bozovich cites that is most helpful to him is United States v. Beler, 20 F.3d 1428 (7th Cir.1994), but that case is still quite different from his. In Beler we vacated a sentence when the drug quantity finding underlying that sentence was supported only thinly if at all. The drug quantity in Beler was based on the testimony of one man. He was a "cocaine addict and government informant," id. at 1435, as Bozovich emphasizes. Central to our holding in Beler, however, was the fact that this witness testified at trial without providing a drug quantity. Instead, he submitted two affidavits after the trial. The affidavits were inconsistent with each other and with his trial testimony. In the face of such competing and shifting stories from the sole witness, we held that the district judge had not scrutinized the drug quantity evidence sufficiently before making a finding. Id. at 1433-35. Because the district judge here made a clear credibility finding and otherwise carefully scrutinized the drug quantity evidence, Beler does not control.

The judgment of the district court is AFFIRMED.