United States v. Vixamar

679 F.3d 22, 2012 WL 1649702

• Court: Court of Appeals for the First Circuit
• Decided: May 11, 2012
• Docket: 11-1217, 11-1251
• Status: Published

Opinion

THOMPSON, Circuit Judge.

Overview

Imartha Vixamar and Mary Saintfleur think the district judge got it all wrong when he revoked their probation and re-sentenced them to prison terms above the ranges recommended by the Sentencing Commission. 1 We think the opposite and affirm.

Probation

Back in June 2009, Vixamar and Saintfleur pled guilty to two counts each of passport fraud. See 18 U.S.C. § 1542. According to their testimony at the sentencing hearing held a little later, this is how that crime went down: Following Vixamar’s lead, Saintfleur applied for a new passport sometime in 2007, falsely claiming that someone had stolen her old one. Vixamar gave Saintfleur a photo to hand in with the application, promising her $1,000 for her troubles. Vixamar later picked up the passport, hightailed it to Haiti, and passed it on to another woman there.

Getting off with probation, Vixamar and Saintfleur pledged not to consort with convicted felons without probation’s okay and, commonsensically enough, not to commit future crimes either. They also promised to file truthful and complete monthly supervision reports with probation, to notify probation within 72 hours of their being questioned or arrested by police, and to remain regularly employed and to notify probation of employment changes at least 10 days in advance. On top of that, Vixamar agreed not to work at any healthcare job that gave her access to “patient valuables” unless she first told her employer or the patients about her criminal past and let probation confirm that she had done that — a condition imposed because of her 2007 convictions for stealing, forging, and cashing checks from elderly patients at assisted-living facilities where she had worked; a striking example is her having stolen checks and a credit card from a 96-year-old resident of a life-care center in Randolph, Massachusetts, taking her for thousands of dollars.

The duo could have received harsher sentences. Passport fraud in cases like theirs is punishable by up to 10 years in prison on each count, see id., and Vixamar’s advisory Guidelines sentencing range (“GSR”) was 12 to 18 months in prison, while Saintfleur’s was 18 to 24 months. But the district judge instead sentenced them both to 36 months’ probation, with 9 months of home confinement for Vixamar and 5 months for Saintfleur. The judge put a lot of thought into Vixamar’s sentence. Reviewing all the evidence from the sentencing hearing, the judge found that Vixamar had come up with the passport-fraud scheme — that is what Saintfleur had testified to — and had lied to investigators and to him under oath in a desperate bid to duck responsibility. But because (among other things) she was still nursing her infant daughter, the judge decided on probation, even though he was not convinced that she had grasped how serious her offense was. And he warned her and Saintfleur that if they did not live up to their probation obligations, they would probably go to prison.

Violations

Vixamar and Saintfleur did not stay out of trouble for long, returning to federal court in January 2011 on probation-viola tion charges. The gist of the charges was this: Randolph police had arrested Saintfleur for depositing into her account in September 2010 a $4,000 check stolen from Lois Gibbs, an elderly, mentally-incompetent Randolph resident receiving in-home hospice care. The check was dated August 29, 2010. Saintfleur had reported to probation that she had no checking account in her name and had received only $200 for that entire month. She also had said nothing to probation about the arrest. Saintfleur had no ties to Gibbs. But Vixamar did through her work as a certified nursing assistant (“CNA”) for Clinical One, a healthcare-staffing company that had placed her with Gibbs right around the time that the check went missing. The theory was that Vixamar had swiped the check and handed it off to Saintfleur. Compounding her problems, Vixamar had not filled in Clinical One on her criminal past, had falsified her job application— giving the company her husband’s last name (“Jacques”) rather than the one she normally went by (“Vixamar”), and jotting down a false social-security number and date of birth — and had created a false job reference too. She also had kept probation in the dark about her Clinical One job, and neither she nor Saintfleur had ever told probation about their interacting with each other. Making matters worse for Saintfleur, law enforcement had a video of her helping a “friend” deposit a $6,000 check stolen from an elderly resident of a Reading, Massachusetts, nursing home where she had worked as a CNA. Unsurprisingly, Saintfleur had never clued probation in to the fact that she had worked there for two months until she was fired in November 2010, and she had never notified probation that the Reading police had contacted her. All of this resulted in the pair’s getting hit with five probation-violation charges apiece, though some of Saintfleur’s charges had several subparts.

A magistrate judge held a preliminary revocation and detention hearing, see Fed. R.Crim.P. 32.1(b)(1), and, after listening to testimony from a probation officer, concluded that probable cause existed for two of the charges against Vixamar (arising from (a) her not telling probation about her Clinical One job and (b) her falsifying her Clinical One job application) and for four of the charges against Saintfleur (stemming from (a) her depositing the stolen Gibbs check and (b) her not telling probation how police had questioned and arrested her or how she had worked for and gotten fired by the nursing home). Pertinently for present purposes, the magistrate judge found that the government had offered no evidence to bolster the charges that centered on Vixamar’s involvement in the Gibbs check-cashing scheme or their associating with each other. And he ultimately released the two to home confinement with electronic monitoring pending the final revocation hearing.

Revocation and Resentencing

That hearing happened ten months later before the same district judge who had sentenced them originally. See Fed. R.Crim.P. 32.1(b)(2). Vixamar’s and Saintfleur’s counsel started off by saying that their clients would admit to all the charges that the magistrate judge had found probable cause for, but not to the others. And the prosecutor indicated that the government did not intend to proceed on these other charges anyway. Not so fast, said the judge. The magistrate judge had to decide only whether to detain or release the defendants, the judge added, but he (the district judge) had to decide whether to revoke their probation and re-sentence them. Given their different tasks, “whatever the magistrate did is not binding on me.” 2 “I regard this, across the board, as an extremely serious matter,” said the judge, “and I’m not going to let it go until I’ve reliably found the facts.”

Saintfleur was sworn in, and, responding to the judge’s questions, she admitted all the charges against her except the association one. She confessed to a lot of things, including that she had known all along that the Gibbs check was a stolen check and that she had later helped a friend deposit the stolen $6,000 check too. But, she insisted, neither she nor Vixamar had stolen the Gibbs check. A “friend of mine,” a woman named “Nerlande Sanon,” had, she said. 3

Hoping to get to the bottom of this, the judge set the matter down for a formal evidentiary hearing. Before the government presented its case, the judge questioned Vixamar under oath, and she conceded that she had falsified her Clinical One job application, had not come clean with Clinical One about her prior convictions, and had not told probation about her Clinical One job. But she denied the remaining charges. The judge then went over the ground rules for the hearing, stressing that he could revoke probation if the proof, “reasonably” viewed, “satisfied” him that a violation had occurred. See, e.g., United States v. Dilanni, 87 F.3d 15, 17 (1st Cir.1996). “But it may require a preponderance of the evidence to satisfy me,” he added. 4 The judge asked whether anyone disagreed with that. And no one did.

The judge then heard testimony from a number of government witnesses, including John Bringardner, a Randolph detective who spoke about the Gibbs-check theft; Melissa Knybel, a Clinical One director who talked about Vixamar’s placement with Gibbs; and Nerlande Sanon (one of two Nerlande Sanons living in the area, a government search revealed), who testified about knowing Saintfleur but who denied knowing Vixamar or Gibbs or having a hand in the Gibbs-check heist. Neither Vixamar nor Saintfleur took the stand again (so the government never cross-examined them at the evidentiary hearing), and neither called any witnesses either.

After hearing evidence and argument, the judge found that Vixamar and Saintfleur had proven themselves to be incorrigible “frauds and liars” with “a history of conspiring with each other to commit crimes by using false documents.” And he concluded that the government had proven “by far more than” a preponderance of the evidence that the two had committed the violations that they had admitted to, and also that Vixamar had stolen the Gibbs check and had worked with Saintfleur to have it forged and deposited. Vixamar’s fate would have been the same, the judge said, even without her prior convictions for using her health-aide positions to steal from the elderly — but they “reinforce[d]” his “conclusions.” The government, he added, only fell short of proving that Saintfleur had not notified probation about being questioned by Randolph police, because the record showed that Detective Bringardner had tried to interview her (they had played a lot of phone tag) but never did. As for Saintfleur’s story that a Nerlande Sanon had poached the Gibbs check, the judge did not buy it, even for a second, given that Saintfleur had zero credibility and that the Sanon who had testified here credibly denied taking the check. Also, the judge stressed, Saintfleur’s decision not to testify and undergo cross-examination (as she had every right to do) undercut her story even if it were credible, which, again, it was not.

Having revoked their probation, the judge turned his attention to resentencing. See 18 U.S.C. § 3564(e) (explaining how “[a] sentence of probation remains conditional and subject to revocation until its expiration or termination”); id. § 3565(a)(2) (discussing how a court can “revoke the sentence of probation and re-sentence the defendant”). The Sentencing Commission classifies three grades of probation violations — A, B, and C — with the grades determined by the “conduct constituting” any “federal, state, or local offense punishable by” various terms of imprisonment. U.S.S.G. § 7B1.4. It also offers a table listing recommended sentencing ranges for probation violations. U.S.S.G. § 7B1.4(a). Vixamar and Saintfleur had committed Grade B violations, the judge said, which, when combined with their criminal histories, put Vixamar’s revocation GSR at 6 to 12 months and Saintfleur’s at 8 to 14 months. But the judge added, and the parties agreed, that he could also factor into his resentencing decision what their GSRs had been for the passport-fraud crimes (12 to 18 months for Vixamar; 18 to 24 months for Saintfleur) and what the statutory sentencing range had been for those crimes too (up to 10 years on each passport-fraud count). And he noted that he could depart upward if “the original sentence was the result of a downward departure____” U.S.S.G. § 7B1.1 cmt. n. 4.

Pausing at this point, the judge asked defense counsel whether he had calculated the GSRs correctly, and they answered yes. The government then asked for a sentence at the top of each defendant’s GSR. But Saintfleur’s lawyer floated the idea that the government’s recommendation might breach some prosecutorial promise that had persuaded her to plead guilty to passport fraud in 2009. See Santobello v. New York, 404 U.S. 257, 262, 92 S.Ct. 495, 30 L.Ed.2d 427 (1971) (holding that prosecutors must honor their end of any plea agreement that they make). So, in response, the judge said that he would not rely on the government’s sentencing proposal. Vixamar’s lawyer asked the judge to sentence his client to a year and a day in jail and six months of home confinement. 5 And Saintfleur’s attorney requested that his client get probation again, with the condition that she serve a year and a day in a halfway house.

The judge resentenced them each to 36 months in prison and 36 months of supervised release, however. And he explained why, hitting the relevant sentencing factors in 18 U.S.C. § 3553(a). 6

Turning to Vixamar, the judge reminded her that he had given her a real sentencing break in 2009 — remember, her record had shown multiple convictions for stealing from the elderly, and she had been the driving force behind the passport-fraud stunt — something he had done even though he had had serious doubts about whether she had learned anything from her repeated brushes with the law. Vixamar then not only lied to probation (about her job situation, for example) and to Clinical One (by handing over a false social-security number and date of birth, among other things), but she also committed another especially “despicable” crime — stealing from a person in her care who could not protect herself. “[Yjou’re a menace, a tremendous danger to other people,” the judge said, “particularly to [the] very vulnerable,” and he thought that her sentence would deter her and others from crimes like this and protect the public too. And he noted that 18 months (the high end of the GSR) would have been perfectly reasonable for her passport-fraud crimes. But he concluded that 36 months was justified because she had violated the “trust” that he had “placed” in her.

Speaking to Saintfleur, the judge said that he “could have locked [her] up, very reasonably, for two years in November of 2009,” and if he had — i.e., if he had not given her a sentencing break too — the Gibbs-check theft would not have happened. Like Vixamar, she had lied to probation and had abused the trust that he had placed in her. Also like Vixamar, she had shown herself to be “a very dangerous person” — she had gotten worse over time, graduating from shoplifting to passport fraud to fleecing the elderly — and he concluded that her sentence was necessary for deterrence and public protection as well.

In his written statements of reasons in support of the sentences, see 18 U.S.C. § 3553(c)(2), the judge, among other things, again referenced the § 3553(a) factors and noted too how Vixamar and Saintfleur had gotten downward variances at their original sentencing — all of which justified the above-Guidelines sentences, he wrote.

Issues on Appeal

Arguing in unison, Vixamar and Saintfleur raise two issues on appeal:

For openers, they say that the government presented insufficient evidence to support two of the judge’s violation findings — namely, that Vixamar had run off with the Gibbs check, which Saintfleur then deposited, and that the two had interacted with each other. They call this an attack on the procedural reasonableness of their sentences and analyze the issues through that prism. But because, again, what they are really doing is questioning the sufficiency of the evidence underpinning the two contested violations, a different set of cases controls. See United States v. Gallo, 20 F.3d 7,13 (1st Cir.1994). Under this regime, we review for clear error the district judge’s factual findings regarding whether the probationer violated a probation condition, and we review for abuse of discretion the judge’s decision that that violation warrants revocation of probation. Id. For Vixamar and Saintfleur to succeed on their sufficiency claims, they must get us to firmly and definitely believe that the judge made a mistake — a tall task, given that (a) the evidence must be viewed in the light most agreeable to the government, (b) the judge’s choice among competing but plausible inferences from the evidence cannot as a matter of law be clearly erroneous, and (c) credibility calls were his to make. See id. (discussing points (a) and (c)); United States v. Pacheco, 489 F.3d 40, 45 (1st Cir.2007) (discussing point (b)).

Vixamar and Saintfleur also say that their 36-month prison terms are too harsh, amounting to substantive unreasonability — an issue we review for abuse of discretion too, see United States v. Gallardo-Ortiz, 666 F.3d 808, 811 (1st Cir.2012), knowing that the defendants have an uphill fight, given that there is no single “reasonable” sentence in any one case but rather a range of sensible outcomes, see United States v. Clogston, 662 F.3d 588, 592 (1st Cir.2011), and that a sentence will stand provided that there is “a plausible sentencing rationale and a defensible overall result,” see United States v. Bunchan, 626 F.3d 29, 35 (1st Cir.2010) (internal quotation marks omitted).

Analysis

The Sufficiency Challenge

Vixamar’s and Saintfleur’s sufficiency arguments are straightforward enough. But before we get to them we need to discuss an issue that arose rather late in the game concerning the burden of persuasion for probation revocation. By way of a post-argument letter, see Fed. R.App. P. 28(j), the government argues that the reasonably-satisfied standard laid down in our caselaw —see, e.g., DiIanni, 87 F.3d at 17; Gallo, 20 F.3d at 14; United States v. Czajak, 909 F.2d 20, 22 (1st Cir.1990) — is a lesser standard than the preponderanee-ofthe-evidence standard, which is a more-likely-than-not rule, see Earle v. Benoit, 850 F.2d 836, 841 (1st Cir.1988). Not so says Vixamar in response, pointing to cases outside this circuit and insisting that “reasonably satisfied” requires a preponderance of the evidence. We do not choose sides on this issue today — an issue that has not gotten full-briefing treatment — because Vixamar and Saintfleur lose either way.

Here is why:

(a) The judge had seen and heard Vixamar and Saintfleur during the 2009 sentencing hearing and the 2011 revocation hearing, and he found them to be liars: Vixamar had lied about her role in the passport scam, the judge noted, and later had lied to probation about her Clinical One job and to Clinical One about her true identity. Saintfleur had lied to probation about her checking-account situation, her employment status, and her brushes with the Reading police. And, ultimately, we see no reason to second-guess the judge’s credibility findings.

(b) Dated August 29, 2010, the stolen Gibbs check had been taken from the middle of an extra checkbook tucked inside a box in one of Gibbs’s dresser drawers, and one could plausibly infer from that that whoever snatched it had had access to Gibbs’s home for parts of August 2010. That ruled Saintfleur out. But not Vixamar, who as Gibbs’s CNA certainly had access, and who by the way only had access because she had violated other probation conditions by working for a home-healthcare provider, lying to get that job, and cutting probation out of the loop.

Saintfleur fires back that others had entrée to Gibbs’s residence too, and she decries the fact that there was no evidence concerning the CNAs on duty when Vixamar was not. What she effectively wants is for us to use something over and above even the beyond-a-reasonable-doubt test, suggesting that we must reverse because the government did not disprove every possibility that might let her off the hook. Her argument goes nowhere, however, given the series of cases that hold that “[t]he government need not prove a violation beyond a reasonable doubt, but must merely satisfy the court that a violation occurred,” DiIanni, 87 F.3d at 17; accord Gallo, 20 F.3d at 14—which makes the standard that she essentially pushes for out of the question.

(c) Vixamar and Saintfleur also knew each other, an inescapable conclusion given the passport scam that they had run together. Desperate to poke holes in the judge’s analysis, Vixamar accuses him of exaggerating the “history” between her and Saintfleur, taking particular offense at his saying that the two had a “history of conspiring with each other to commit crimes using false documents.” According to Vixamar, the record shows only that she had recruited Saintfleur into the passport-fraud scheme and had gone with her to the passport office to file the false paperwork, which, she intimates, is not much of a “history.”

That is too a myopic view of the evidence, since the record shows that Saintfleur gave Vixamar the receipt so that Vixamar could pick up the passport the next day, and that the two then headed back to Saintfleur’s house, at which point Vixamar gave Saintfleur some of the money that she had promised her for helping out. Just as importantly, Vixamar glides over the fact that she and Saintfleur had been “friends” for five or six months before she sprung the passport-fraud idea on Saintfleur. No matter. The judge’s description of the defendants’ conspiracy history jibed with the evidence and does not come close to clear error.

As a fallback, Vixamar protests that the judge’s “history” comment shows that he was swayed by their “propensity to commit crimes together,” which is a Fed.R.Evid. 404(b)-type argument. Rule 404(b) holds that evidence of past crimes is inadmissible to prove that a defendant probably did whatever she is currently accused of, simply because she has shown a propensity to break the law. See, e.g., United States v. Tse, 375 F.3d 148, 156 (1st Cir.2004). But even assuming for argument’s sake that Vixamar is right about her propensity point (something we do not decide), she gains nothing because the rules of evidence do not apply to probation-revocation proceedings. See, e.g., Fed.R.Evid. 1101(d).

(d) Saintfleur admitted to depositing the stolen Gibbs check into her own account. She also admitted to helping an unnamed “friend” deposit a $6,000 check lifted from a patient at a Reading nursing home where she (Saintfleur) had worked as a CNA (without telling probation). Interestingly, the record reveals that Saintfleur had gone from stealing from stores to stealing from the elderly after teaming up with Vixamar on the passport scam.

Saintfleur reminds us that she had told the judge under oath before the evidentiary hearing that she had gotten the check from a “friend” named Nerlande Sanon. Convinced that this is some sort of trump card, she and Vixamar contend that her statement proves that Vixamar was not the thief and that the two had not interacted with one another as charged. We disagree.

For one thing, Saintfleur played fast and loose with the truth, the judge found, and even though prevaricators like her may not prevaricate all the time, see United States v. Williams, 216 F.3d 611, 614 (7th Cir.2000), he also found that her Nerlande-Sanon story was not credible either — and again, that was for him, not us, to decide, see United States v. Oquendo-Rivera, 586 F.3d 63, 67 (1st Cir.2009). For another thing, the judge could and did credit Nerlande Sanon’s testimony that she knew Saintfleur but not Vixamar or Gibbs, and that she had nothing to do with the Gibbs-check theft. Ultimately, these not-clearly-erroneous findings pour cold water on Saintfleur’s theory.

Wait a minute, Vixamar says. A government agent had found two Nerlande Sanons after running different internet searches but only one testified at the evidentiary hearing. Maybe the non-testifying Nerlande Sanon was the thief, Vixamar speculates. After all, the agent testified that this Nerlande Sanon lived in Randolph (just like Gibbs) and had worked as a CNA in 1976, and the defense introduced documents showing that she also had a dismissed larceny charge on her record from around that time, had filed for bankruptcy in 1996 but now owned a Mercedes, and has a daughter (Daphnee Germaine) who is a nurse. So maybe she or her daughter, the theory goes, had slipped Saintfleur the Gibbs check on the sly. Or because Nerlande Sanon is a fairly common Haitian name, Vixamar continues, maybe some .other Nerlande Sanon was the culprit, and if the agent had only run more or better searches, he would have found her.

This argument does not persuade. For starters, and to repeat ourselves, the judge rejected Saintfleur’s Nerlande-Sanon story, as he had every right to do. Also, Vixamar again sounds like she wants us to reverse because the government did not eliminate all doubts about whether she had done what she is alleged to have done. But again, that is not the test. See, e.g., DiIanni, 87 F.3d at 17; Gallo, 20 F.3d at 14. Cinching matters, the agent also testified that the non-testifying Nerlande Sanon, who was an investigator with the state department of social services, had said that she did not know Saintfleur or Vixamar (she did not recognize them from their photos) and that she had not passed any checks on to either of them. She had been “very helpful” to his investigation, the agent stressed, and “didn’t appear to be holding anything back.” And viewing the evidence in the light most favorable to the government’s theory of the case, we conclude that the judge did not stumble in spurning the suggestion that a Nerlande Sanon (either one of the two a government agent had found or one the agent had missed) was the Gibbs-check thief.

Relatedly, Vixamar blasts the judge for intimating that Saintfleur had falsely fingered Nerlande Sanon as the wrongdoer to falsely clear Vixamar. But taking the evidence in the light most friendly to the government, making all reasonable inferences in its favor too, we see no reversible error here.

Our bottom-line conclusion: Assuming (without deciding) that the reasonably-satisfied standard for probation revocation entails a more-likely-than-not threshold, we hold that the district judge did not clearly err in finding by a preponderance of the evidence that Vixamar had stolen the Gibbs cheek and that she and Saintfleur had associated with each other. Given the soundness of those factual findings and the leeway the law gives district judges for judgment in this area, we cannot say that the judge here abused his discretion by revoking Vixamar’s and Saintfleur’s probation. And so we move on.

The Sentencing Challenge

As for Vixamar’s and Saintfleur’s claims that their 36-month prison sentences are too long, their odds of winning are not great. See, e.g., Clogston, 662 F.3d at 592. Applying abuse-of-discretion review, we cannot vacate for substantive un reasonableness unless they show that the judge’s decision “falls outside the ‘expansive boundaries’ ” of rationally-available sentencing choices. See United States v. Vargas-Dávila, 649 F.3d 129, 130 (1st Cir.2011) (quoting United States v. Martin, 520 F.3d 87, 92 (1st Cir.2008)). This they have not done. We explain, beginning with a few basics.

Congress crafted a specific statutory scheme to guide a sentencing judge’s discretion in probation-revocation cases. At the risk of oversimplification, here is how it works. After holding a hearing under Fed.R.Crim.P. 32.1 and considering the relevant § 3553(a) factors, a district judge may either “continue [the] probation, with or without extending the term or modifying or enlarging the conditions,” 18 U.S.C. § 3565(a)(1), or “revoke the sentence of probation and resentence the defendant,” id. § 3565(a)(2). Of course, following Congress’s direction, the judge should consult the Sentencing Commission’s non-binding policy statements, see 18 U.S.C. § 3553(a)(5) — statements that suggest, for example, that the district judge should revoke probation for a Grade A or B violation, see U.S.S.G. § 7B1.3(a)(l), and should focus on punishing the “breach of trust” that the violation represents, “while taking into account, to a limited degree, the seriousness of the underlying violation and the criminal history of the violator,” see id. Ch. 7 Pt. A(3)(b). 7 Also, and again, the § 3553(a) factors that guide sentencing include the characteristics of the defendant and the nature of the offense, plus the need for the sentence to reflect the seriousness of the crime, to provide sufficient deterrence, and to protect the public. See footnote 6, above. A couple more things to keep in mind: First, the probation violation puts in “play the penalty provisions of the original charge,” United States v. de Jesús, 277 F.3d 609, 611 (1st Cir.2002) (emphasis added), meaning the judge can “impose any sentence [he] might originally have imposed,” United States v. Bynoe, 562 F.2d 126, 129 (1st Cir.1977) — even one that “differ[s] from the Sentencing Commission’s recommendation,” provided he “stay[s] within the range set by the statute[] of conviction,” Rodriguez, 630 F.3d at 41. And second, a judge may depart upward from the probation-revocation GSR if the original sentence for the original crime resulted from a downward departure. See U.S.S.G. § 7B1.4 cmt. n. 4.

To state the obvious, Vixamar’s and Saintfleur’s 36-month sentences are well below the combined 20-year maximum that each could have gotten after pleading guilty to two passport-fraud infractions apiece, see 18 U.S.C. § 1542, though their sentences are above the 18-month (Vixamar) and 24-month (Saintfleur) máximums suggested for the passport-fraud crimes under the advisory-Guidelines regime. And they are also above the 12-month (Vixamar) and 14-month (Saintfleur) máximums recommended for their probation violations.

Because judges typically impose within-Guidelines sentences, see Rodriguez, 630 F.3d at 42, a judge who does not risks creating “unwarranted” sentencing disparities, in violation of 18 U.S.C. § 3553(a)(6), so careful thought is required, see Gall v. United States, 552 U.S. 38, 54, 128 S.Ct. 586, 169 L.Ed.2d 445 (2007). And, as Vixamar and Saintfleur stress, we must consider the degree of difference between the sentences and the recommended Guidelines ranges. But we must also “give due deference” to the judge’s conclusion “that the § 3553(a) factors, on a whole, justified]” the sentences. Gall, 552 U.S. at 51, 128 S.Ct. 586.

The record here reveals no real reason to second-guess the judge’s exercise of sentencing discretion and plenty of reasons why it deserves our respect. Vixamar’s perjuring herself during the passport-fraud proceedings was bad enough. But after getting a sentencing break, she then lied to Clinical One (using a false social-security number and date of birth, and making up a job reference), and she lied to probation too (keeping quiet about her Clinical One job), all so she could prey on elder-care patients — the very persons the judge had sought to protect through certain probation conditions. Saintfleur was hardly any better. Like Vixamar, she chose to parlay her sentencing break into an opportunity to steal from the defenseless. And she was not straight with probation either (regarding her checking account and job situation and her arrest and other contacts with the police). Obviously the probation sentences did not make Vixamar and Saintfleur mend their ways— they had actually gotten worse, the judge found. “[V]ery dangerous” were words he used. “[IJncorrigible” was another. The need to deter them from committing future crimes — and to deter other healthcare workers from copying them — justified above-Guidelines sentences selected to make the message sink in and hopefully stick. Cf. Vargas-Dávila, 649 F.3d at 131 (finding the defendant’s 24-month prison sentence substantively reasonable, even though his GSR was 5 to 11 months, noting, among other things, his “checkered record of non-compliance” with his supervised-release conditions).

The above-Guidelines sentences were also justified given what the Sentencing Commission says about how upward departures for probation violations may be warranted following downward departures at the original sentencing. See U.S.S.G. § 7B1.4 cmt. n. 4. The judge at various points in the record called each probation sentence the product of a downward “departure.” But he also called them the result of a downward “variance.” Now, the policy statement relevant here speaks only of “departure[s].” Id. Departures and variances do differ. See Irizarry v. United States, 553 U.S. 708, 714, 128 S.Ct. 2198, 171 L.Ed.2d 28 (2008) (explaining that a “ ‘[departure’ is a term of art under the Guidelines and refers only to non-Guidelines sentences imposed under the framework set out in the Guidelines”); United States v. Gibbons, 553 F.3d 40, 42 (1st Cir.2009) (clarifying that a variance results from the judge’s consideration of the § 3553(a) factors). But their differences do not matter for present purposes — the policy applies just as well to one as to the other.

Vixamar and Saintfleur make several counter-arguments, none of which they raised below, so we review only for plain error. See, e.g., United States v. Edelkind, 467 F.3d 791, 796 (1st Cir.2006). Plain error, of course, requires “(1) error, (2) plainness, (3) prejudice, and (4) an outcome that is a miscarriage of justice or akin to it,” id. at 797 — a hard-to-meet standard, to be sure, see, e.g., United States v. Tan, 674 F.3d 103, 111 (1st Cir.2012).

First up is Vixamar’s claim that the judge impermissibly “double counted” her breach of trust, an argument that goes something like this: Pointing to the Sentencing Commission’s policy statement that we quoted several paragraphs ago, she says that a judge in a revocation proceeding should impose a sanction for the “breach of trust,” leaving punishment for any separate criminal offense to follow a conviction in a separate criminal proceeding. Next she says that the judge reached the 36-month number by first giving her 18 months (the upper end of the GSR for the passport-fraud offenses) because she had breached the trust he had placed in her when he gave her probation and then by adding 18 more because, again, she had breached his trust when she violated the probation conditions. 8 And that, she insists, without citing a single case, amounts to forbidden double counting.

Vixamar gets no mileage out of this argument. Admittedly, her stealing the Gibbs check constituted both a probation violation and a separate criminal offense. And by reading certain selected statements of the district judge, one might be tempted to think that the judge did what the Sentencing Commission recommends against — sentencing Vixamar for the triggering crime. One would be wrong, however. The record, when fairly and comprehensively examined, makes clear that the entire 36-month sentence was sparked by her breach of trust — though the judge did note the seriousness of her new offense, which is a permissible § 3553(a) consideration — rather than by a desire to punish her for her latest crime. All of this tracks nicely with the Sentencing Commission’s advice. The judge had the right focus, in other words. And with his not having subverted the policy statement that she relies on, Vixamar is left with nothing (she points us to no caselaw, for example) supporting her position. Simply said, this is not the stuff of plain error. See United States v. Roy, 506 F.3d 28, 30-31 (1st Cir.2007) (explaining the plain-error standard with exquisite care).

Vixamar also accuses the judge of over-zealousness in pursuing public deterrence, claiming he had let his “personal distaste” for what she had done get the better of him. True, the judge called thieving from a woman in hospice care “despicable.” But we see nothing suggesting any impropriety in the way he weighed the § 3553(a) factors.

As for Saintfleur, she does argue that the judge obsessively focused on punishing her for depositing the Gibbs check — a serious crime, she apparently concedes. But what we have just said about the judge having trained his focus on the right considerations sinks this argument too. She also contends that her probation-revocation GSR (8 to 14 months) adequately covered the seriousness of the situation, and she faults the judge for not explaining why he thought that it did not. Again, from what we have said it is obvious that the judge did spell out why a within-Guidelines sentence would not work, so we need say no more about that. And her final claim— that the judge could adjust her sentence upward because of the original sentencing break, but only by one level — is a nonstarter, given that she cites no authority, nor can we think of any, suggesting that she is right.

Weighing the § 3553(a) factors is no easy task. It is far from a science. See Clogston, 662 F.3d at 593. And it involves many hard choices, with a lot hanging in the balance. But sentencing judges (unlike us) do this almost daily, year in and year out, seeing and hearing the defendants and other witnesses first hand. See id. Hence our deferential standard of review. See id. Taking advantage of his superior position, the judge here imposed defensible sentences, given his sensible view of the circumstances, and he offered plausible rationales for them too. Maybe a different judge would have made different choices. But because we cannot say that this judge’s choices fall outside the realm of “reasonable sentencing outcomes,” Vixamar’s and Saintfleur’s sentences stand. See id. at 592.

Conclusion

For the reasons given above we affirm. the judgments below in all respects.

1 . Saintfleur's name is spelled "Saint Fleur" in the record, but, for simplicity's sake, we use the spelling that her lawyer uses in her brief to us.

2 . Both Vixamar and Saintfleur play up the magistrate judge’s no-probable-cause findings on appeal. But neither argues that these findings bound the district judge. And rightly so. See generally 3 Charles A. Wright St Sarah N. Welling, Federal Practice and Procedure § 562, at 379 (4th ed. 2011) (noting that a "revocation hearing in a felony case must be held before a judge rather than a magistrate judge").

3 . The friend's name is spelled "Narlan Sanon’’ at one point in the transcript. We use what the parties tell us is the right spelling.

4 . The probation-revocation statute, unlike the revocation-of-supervised-release statute, says nothing about the level of evidence required. Compare 18 U.S.C. § 3565 (setting out no burden of persuasion for revocation of probation) with 18 U.S.C. § 3583(e) (declaring that preponderance of the evidence is the burden of persuasion for revocation of supervised release).

5 . Looking for leniency, Vixamar’s lawyer offered up a medical note indicating that his client was pregnant. "I’m a little concerned about admitting this,” the judge announced, because the passport-fraud and probation-revocation cases "have been about false documents.” So the judge called the doctor’s office from the bench on speaker phone and in open session. And the doctor confirmed that, yes, indeed, Vixamar was pregnant.

6 . These include a defendant’s background, the nature and circumstances of the offense, the seriousness of the offense, the need to deter criminal conduct, and the need to protecl the public. See, e.g., United States v. Rodriguez, 630 F.3d 39, 41 (1st Cir.2010) (paraphrasing § 3553(a)).

7 . We quote a little bit more from this statement:

While the nature of the conduct leading to the revocation [may] be considered in measuring the extent of the breach of trust, imposition of an appropriate punishment for any new criminal conduct [is] not ... the primary goal of a revocation sentence. Instead, the sentence imposed upon revocation [is] intended to sanction the violator for failing to abide by the conditions of the court-ordered supervision, leaving the punishment for any new criminal conduct to the court responsible for imposing the sentence for that offense.

Id.

8 . Here is what the judge said:

This [36-month] sentence is above the advisory guideline range, but any disparity between you and others who might be superficially similar is justified. I gave you a very substantial downward departure, a break, in November of 2009, from 12 to 18 months. I could have locked you up for 18 months in November of 2009 and then Mrs. Gibbs wouldn’t have had her money stolen, and you quickly violated that trust by committing serious crimes. The guidelines indicate that usually 18 months is reasonable the first time. Your new crime is worth at least another 18 months, because you also violated the trust — the more hope than trust, even — that I placed in you.