This court adjudged defendant Johnny Newman guilty of two violations charged in an amended petition for revocation of supervised release: on April 27 and May 9, 2018, he tested positive for marijuana and admitted to using the drug. He now asks the court to sentence him as follows based on these two violations as well as a third positive marijuana test on June 14, 2018, reported by the U.S. Probation Office: imprisonment of six days, to be served in three two-day spans when he is not working. He also proposes that, for any future positive drug test, the court sentence him based on the "swift and certain" model of punishment pioneered by Hawaii's Opportunity Probation with Enforcement (HOPE) program. For the reasons that follow, the court will revoke Newman's supervised release and sentence him as he proposes for his three past violations; the court will also accept his proposed HOPE model for future positive marijuana tests, albeit as modified in this opinion.
I. BACKGROUND
Newman is 52 years old. In 2005, he pled guilty to being a felon in possession of a firearm, in violation of 18 U.S.C. § 922(g)(1). This court sentenced him to 151 months of imprisonment, to be followed by five years of supervised release.
Since beginning supervised release in July 2016, Newman has tested positive for marijuana use at least a dozen times. In December 2016 and March 2017, he admitted to using marijuana. He told the Probation Office that he uses the drug as a form of pain management for an employment-related knee injury. See Noncompliance Summary (doc. no. 59) at 2. Following a hearing, the court ordered continued supervised release and a full substance-abuse assessment. See April 2017 Order (doc. no. 65). He began receiving outpatient substance-abuse treatment at the Chemical Addictions Program (CAP) in Montgomery, Alabama.
Nevertheless, Newman has continued to use marijuana. In July 2017, after he submitted his sixth positive drug screen for marijuana, the Probation Office filed a petition for revocation of supervised release. See July 2017 Revocation Petition (doc. no. 68). Pending resolution of the revocation petition, the court modified Newman's conditions of supervised release. Among other modifications, the court required that he live in a residential re-entry center under contract with the Federal Bureau of Prisons (BOP) for a term of 180 days, and continue attending outpatient treatment at *1220CAP. See August 2017 Order (doc. no. 79) at 1-2.
On September 8, 2017, the residential re-entry center dismissed Newman for using synthetic marijuana. The Probation Office subsequently amended its revocation petition to include this incident. On October 13, 2017, after a hearing, and by agreement of all involved, the court dismissed the amended revocation petition, and modified the conditions of supervised release to require that Newman (1) continue attending outpatient treatment at CAP; (2) receive a medical assessment to diagnose his reported knee problem, and obtain treatment if necessary; (3) continue to maintain employment; (4) receive mental-health counseling at least twice a month; and (5) participate in an alcohol-monitoring program and abstain from using alcohol. See October 2017 Order (doc. no. 100) at 1-2. The court further ordered, based on the agreement of all involved, that Newman need not reside at the residential re-entry center.
Unfortunately, Newman's marijuana use persisted. In March 2018, he pled guilty to 11 violations, including testing positive for marijuana on six different occasions between September and December 2017. See Judgment (doc. no. 117). The court modified his conditions of supervised release by sentencing him to 15 days in jail, to be served on weekends.
This opinion arises out of another amended revocation petition filed on May 22, 2018, which alleged three violations. The first two are for using marijuana--based on his positive tests for the drug on April 27 and May 9, 2018--and the third is for failing to report for drug screening on May 17, 2018. This court adjudged Newman guilty of the first two violations and the government dismissed the third. See June Order (doc. no. 133). According to a status update provided by the Probation Office, he subsequently tested positive for marijuana on June 14, 2018, and admitted to using the drug. See June Status Report (doc. no. 136) at 1.
Newman reports that he uses marijuana to alleviate chronic knee pain, which is aggravated by his work at the Cash Saver grocery store.1 Although he has been diagnosed with "osteoarthritis of knee," Patient Care Summary (doc. no. 148-1) at 1, he has faced bureaucratic and financial obstacles to obtaining medical treatment, including orthopedic care. According to Newman, his limited hours at Cash Saver do not qualify him for employer-sponsored health insurance.
The most recent development in Newman's case is encouraging. On October 3, 2018, his probation officer reported that, since being placed on an alcohol-monitoring device in October 2017, Newman had undergone almost 1,400 alcohol tests without having any "issues with [him] consuming alcohol." October Status Report, (doc. no. 152) at 1. The probation officer recommended that the alcohol-monitoring device be removed, and the court had the device removed on October 26, 2018. See Order Suspending Special Condition (doc. no. 153) at 2.
II. LEGAL FRAMEWORK
As a general rule under 18 U.S.C. § 3583(g)(4), if a defendant "tests positive *1221for illegal controlled substances more than 3 times over the course of 1 year," the court must revoke his supervised release and sentence him to a term of imprisonment.2 But this rule has an exception. Specifically, when a defendant fails a drug test, the "court shall consider whether the availability of appropriate substance abuse treatment programs, or an individual's current or past participation in such programs, warrants an exception" to mandatory revocation and imprisonment under § 3583(g). 18 U.S.C. § 3583(d).3
Even if the exception to mandatory revocation is warranted, the court still has the authority to exercise its discretion to revoke supervised release and impose a term of imprisonment based on the defendant's violations. In fact, after considering several of the factors set forth in 18 U.S.C. § 3553(a), the court may terminate or extend supervised release, modify the conditions, or revoke the term and imprison the defendant. See 18 U.S.C. § 3583(e). The § 3553(a) factors the court must consider are (1) the nature and circumstances of the offense and the history and characteristics of the defendant; the need for the sentence imposed (2) to afford adequate deterrence to criminal conduct, (3) to protect the public from the defendant, and (4) to provide the defendant with needed educational or vocational training, medical care, or other correctional treatment; (5) the kinds of sentences and the sentencing range established by the Guidelines; (6) any pertinent policy statements issued by the Sentencing Commission; (7) the need to avoid unwarranted sentence disparities among defendants with similar records who have been found guilty of similar conduct; and (8) the need for restitution. See id.4
If, upon considering these factors, the court revokes supervised release and imposes a term of imprisonment, it may also require the imprisonment to be followed by a second term of supervised release. See 18 U.S.C. § 3583(h). There are no minimum lengths for any new terms of imprisonment and supervised release. See 18 U.S.C. § 3583(e)(3) and (h). For Newman, the combined length of any new terms of imprisonment and supervised release cannot exceed five years. See id.
III. ANALYSIS
Because Newman has tested positive for marijuana more than three times over the course of the past year, he is potentially subject to mandatory revocation and imprisonment under § 3583(g). Nevertheless, the court finds that both the "availability of appropriate substance abuse treatment programs"--namely CAP--and Newman's "current [and] past participation in such programs, warrant[ ] an exception" to the mandatory revocation of his supervised release and imprisonment under § 3583(g). 18 U.S.C. § 3583(d).
Just because Newman is not subject to mandatory revocation, however, does not mean that the court cannot exercise its discretion to revoke supervised release and impose a term of imprisonment based on his violations. See, e.g. , *1222United States v. Crace , 207 F.3d 833, 837 (6th Cir. 2000) (stating that the § 3583(d) exception to mandatory revocation "restore[s] discretion to the district judge" to decide whether to revoke supervised release). Instead, as discussed above, the court must consider several of the factors in § 3553(a) to determine whether his violations warrant modifying Newman's conditions of supervised release or revoking the term and sentencing him to imprisonment. See 18 U.S.C. § 3583(e).
In considering the § 3553(a) factors, the court will first analyze the applicable Guidelines range and policy statements. See 18 U.S.C. § 3553(a)(4)-(5). Since Newman's most serious violation is classified as Grade C,5 the Guidelines provide that the court may revoke, extend, and modify the terms of supervised release. See United States Sentencing Commission, Guidelines Manual § 7B1.3(a)(2) (Nov. 2018) (USSG). Nonetheless, Guidelines commentary generally recommends revocation for Grade C violations by defendants who, like Newman, had previously violated conditions and been kept on supervised release. See id. at § 7B1.3, comment. (n.1). In the case of revocation, Newman's Guidelines sentencing range is 8 to 14 months of imprisonment. See id. at § 7B1.4(a). This range is based on the combination of his Grade C violation and his criminal history category of VI, as determined when he was originally sentenced. See id.
Yet Newman's Guidelines sentencing range is advisory, not binding. See United States v. Hofierka , 83 F.3d 357, 361 (11th Cir. 1996). And importantly, as indicated above, the Guidelines range is but one factor among many enumerated in § 3553(a) that the court is required to consider before modifying or revoking supervised release. See 18 U.S.C. § 3583(e).
Based on the court's consideration of all the relevant factors enumerated in § 3553(a),6 the court will revoke Newman's supervised release, and sentence him to three two-night jail stays, to be followed by a new term of supervised release that will conclude in July 2021, when his original supervised release term was scheduled to end.7 As a new condition of supervised release, each time Newman fails a marijuana test, he will immediately be incarcerated for a brief period of time.8 The court considers the six nights in jail for past violations and new term of supervised release requiring short jail stays for future violations to be one integrated, package sentence.
The sentence adopts in large part Newman's proposal to impose swift, certain, *1223and short jail sanctions for his future violations. See Sentencing Memorandum (doc. no. 148) at 6-8. As discussed below, the proposal and sentence are based on HOPE, a probation program premised on the theory that certain and swift punishment for, relatively speaking, minor infractions may be more effective than severe punishment after an accumulation of such infractions.
A. The HOPE Program
A Hawaii state judge started the HOPE program in 2004. In contrast to typical probation systems where violations accumulate until a prison sentence is imposed, HOPE requires an immediate and proportionate sanction for each violation. See, e.g. , Eric Martin, A Hopeful Approach , Nat'l Inst. of Justice (Oct. 2017), https://nij.gov/topics/corrections/community/drug-offenders/Pages/hawaii-hope-demonstration-field-experiment.aspx (A Hopeful Approach , Nat'l Inst. of Justice).
In practice, HOPE begins with a judge delivering a formal warning to the defendant that any violation of probation will immediately result in a brief jail stay. See "Swift and Certain" Sanctions in Probation Are Highly Effective: Evaluation of the HOPE Program , Nat'l Inst. of Justice https://www.nij.gov/topics/corrections/community/drug-offenders/pages/hawaii-hope.aspx (last modified Feb. 3, 2012) ("Swift and Certain" Sanctions , Nat'l Inst. of Justice). After receiving the warning, the defendant is subject to periodic random drug tests, where he must take the test the same day he is notified he has been selected for screening. See id. If the defendant fails to appear for the drug test, a bench warrant is issued and served immediately. See id. If he fails the drug test, he is immediately arrested and brought before a judge within 72 hours. See id. When the defendant is found to have violated the probation terms, he is immediately sentenced to a short jail stay. See id. The typical sentence is for several days and is servable on the weekend if the defendant is employed; however, the length may increase after repeated violations. See id. A third or fourth missed or failed drug test may result in mandatory residential treatment. See id.
Since its founding in 2004, the HOPE model has spread to many jurisdictions across the country. As of 2016, "at least 160 HOPE-like replications" had reportedly taken place in the United States. Daniel S. Nagin, Project HOPE: Does it Work? , 15 Criminology & Public Policy 1001, 1005 (Nov. 2016).
While empirical evaluations of the HOPE model show mixed results, there has been reported success in some jurisdictions. The Department of Justice's research and evaluation agency, the National Institute of Justice (NIJ), has funded three studies of the HOPE program: the initial 2007 Hawaii HOPE experiments, a follow-up of the original Hawaii cohort, and a subsequent study of four jurisdictions in the mainland United States. See A Hopeful Approach , Nat'l Inst. of Justice. The first study produced very positive results. For example, compared to probationers in a control group, after one year HOPE participants were 55 % less likely to be arrested for a new crime, 72 % less likely to use drugs, and 53 % less likely to have their probation revoked. "Swift and Certain" Sanctions , Nat'l Inst. of Justice. The Hawaii follow-up study also had positive results, although to a lesser degree. Compared to conventional probationers, HOPE participants averaged slightly fewer new charges and probation violations, as well as fewer returns to prison. See A Hopeful Approach , Nat'l Inst. of Justice.
Nevertheless, NIJ reports that the latest study--"a strict replication of the HOPE program" in four jurisdictions in Oregon, Texas, Arkansas, and Massachusetts--"was *1224not able to replicate the earlier promising findings from evaluations of HOPE in Hawaii." Id. Even though the latest study "showed reductions in property and drug offenses, the [four jurisdictions] did not experience similar reductions in probation violations and revocations across the board as was seen in an initial evaluation in Hawaii." Id.
This most recent four-jurisdiction study was published in a special issue of Criminology & Public Policy , along with the results of two separate studies of HOPE-style programs in Washington and a small Delaware city. The Delaware city study did not find "evidence of HOPE's effectiveness compared to probation as usual," but the Washington evaluation did. Nagin, supra , at 1006-07.9 In addition to these studies, the Criminology & Public Policy special issue contains academic commentaries with diverse views of the HOPE model.10 The "uniting common theme" of these commentaries is that there is no "silver bullet" to "materially reduce recidivism"; rather, "[a]n amalgam of enforcement tactics and treatment alternatives is required that matches individual and community circumstances." Id. at 1006.
A variety of academic views on the HOPE model have also been published outside of the Criminology & Public Policy special issue. These range from praise for the model's effectiveness in addressing substance-abuse disorders, to criticism of its failure to take into account the individual circumstances of defendants.11
In sum, despite inconsistent results in different jurisdictions, there is evidence of HOPE's success in at least some locations. The HOPE model may have flaws and be far from perfect, but it has the potential to work effectively and fairly in certain individual and community circumstances.
B. Applying the HOPE Model to Newman12
Newman's circumstances make him a good candidate for the HOPE approach.
*1225Notwithstanding his repeated supervised-release violations, he is making progress re-entering society: he is employed by Cash Saver grocery store and has made headway obtaining medical treatment for his knee pain, which may be contributing to his marijuana use. Crucially, brief periods of incarceration--as opposed to a longer sentence--will allow him to remain employed. At this time, nothing seems to be gained from incarcerating Newman for longer, especially given that his marijuana use--and behavior in general--do not appear to pose a significant danger to him or others. Moreover, certain and immediate consequences for his marijuana use--combined with substance-abuse treatment, mental-health counseling, and medical care for his knees--have the potential to help him overcome his drug problem. These circumstances and reasons not only make the HOPE approach a promising match for Newman, but also demonstrate that a HOPE-like sentence here is appropriate under the factors set forth in § 3553(a). See 18 U.S.C. 3583(e).
The court is cognizant that, in general, the HOPE approach of setting automatic penalties for future violations may risk impeding individualized consideration of each defendant's circumstances. See Doherty, supra , at 334. Judge Posner raised a similar concern in a case outside of the HOPE context, where a judge followed through on his promise to imprison the defendant for two years if he violated any one of his 18 conditions of probation. See United States v. Tatum , 760 F.3d 696, 697 (7th Cir. 2014). Judge Posner wrote that a "judge can't be allowed, when imposing conditions of probation (or of supervised release) to commit himself to a specified penalty should there be a violation or violations." Id. In meting out punishments, judges must be able to consider "[a]ny significant changes in the defendant's situation, such as mental deterioration." Id.13
While the court is sensitive to these concerns, it finds that they are sufficiently mitigated by several features here, which are dispositive to this ruling. First, Newman is the one who is proposing the HOPE-like scheme of automatic jail time for future failed drug tests.14 It is hard to argue that the individual defendant's circumstances are not being considered when it is the individual defendant who is in large part crafting the sentence. Second, both the court and Newman retain the right to terminate his HOPE-based sentencing scheme at any time. Thus, the court retains the full discretion to consider "[a]ny significant changes in [his] situation," Tatum , 760 F.3d at 697, and to make sure that any sentence is always specifically tailored to meet his then-current circumstances. Third, if he tests positive for drugs three more times, the court will hold a hearing to re-evaluate the sentencing scheme. Thus, the sentencing scheme will be used only so long as it is proving to be effective. Finally, at any time, the government or Probation Office may petition to change or end the sentencing scheme.
Accordingly, the court will adopt a sentence that closely mirrors Newman's HOPE-based proposal, albeit with some modifications. For his past violations, the court will revoke his supervised release *1226and impose a sentence of imprisonment of six days, to be served in three two-day spans when Newman is not working. Those three jail stays correspond to the April 27 and May 9, 2018, positive drugs tests for which he has been adjudged guilty of marijuana use, as well as the June 14 positive test for which he says he wants to be sentenced at this time. See June Status Report (doc. no. 136) at 1. Furthermore, the court will impose a new term of supervised release that will extend to the same date when his original supervised release term would have ended. All previously imposed conditions shall remain in effect, including the regular, random drug testing, and required substance-abuse treatment and mental-health counseling. For each new instance in which he tests positive for marijuana15 --or fails to appear for a drug test without good cause--a warrant for his arrest will issue and he will turn himself in to spend two consecutive nights in jail, servable when he is not working.16 If Newman tests positive three more times, the court will hold a hearing to re-evaluate this sentence and treatment approach. In other words, as stated, the approach will be used only so long as it proves to be effective.
Newman filed a proposed written agreement that would give effect to his proposed sentence. See Conditional Waiver and Agreement (doc. no. 151). The court will adopt the agreement (assuming Newman agrees to the court's modifications) --and also have an on-the-record colloquy based on its text--albeit with minor modifications. In short, the agreement shall provide that Newman will voluntarily waive a slew of rights under Federal Rule of Criminal Procedure 32.1, with the following understanding: if he tests positive for the use of marijuana or fails to appear for a drug test without good cause, the court will issue a warrant for his arrest on the Monday following the positive test. Upon testing positive, he will be notified that he is to turn himself in before 5:00 p.m. on Monday. The court will hold a hearing the following Wednesday, when he will be released from custody. Prior to the court issuing a warrant, he will not have a right to a preliminary hearing, a detention hearing, or a revocation hearing. The days and times of the warrant and incarceration are subject to change if Newman's work hours change. Finally, both the court and Newman shall retain the right to withdraw from the agreement; however, if Newman withdraws following a positive test, a warrant shall still issue the following Monday.
* * *
When it comes to criminal sentencing and supervised release, there is no silver bullet for reducing recidivism; HOPE is not a one size fits all solution. Still, courts must experiment with new approaches. A HOPE-based sentence seems to fit Newman well and is worth trying here, especially since he has so strongly bought into the scheme himself.
Accordingly, it is ORDERED that sentencing shall resume on November 29, 2018, at 8:00 a.m., in Courtroom 2FMJ of the Frank M. Johnson, Jr., United States Courthouse Complex, One Church Street, Montgomery, Alabama. If this date and time interfere with defendant Johnny *1227Newman's employment, the court is willing to reschedule the date and time for the resumption of sentencing.
DONE, this the 26th day of November, 2018.