Defendant Terrence Carroll received a 30-month sentence after pleading guilty to one count of wire fraud in violation of 18 U.S.C. § 1343. The charge arose from a telemarketing “sweepstakes” scheme that Carroll devised and used to defraud the elderly. The sentence was based on a criminal history ranking that took into account two prior sentences Carroll had received for fraud and theft convictions in 1985. On appeal, Carroll raises-two objections to the consideration of these prior sentences in calculating his sentence for wire fraud. Finding no merit in either objection, we affirm the sentence.
I. Background
Beginning in September of 1995, Carroll implemented a fraudulent “sweepstakes” scheme that operated as follows. Using an assumed name, defendant contacted his victims by phone, identified himself as a representative of “Office Depot,” and informed them that they had won a cash prize (usually totalling between $100,000 and $250,000), in a sweepstakes conducted by his company. In order to receive the prize money, defendant continued, the “winners” would have to set up an account with the Internal Revenue Service (“IRS”) into which they were required to deposit an IRS fee totaling one *459percent of the prize. The defendant then offered to collect that money and set up the IRS account for the victim. Defendant instructed the victims to send the money in the form of a cashier’s check via United States Express Mail to “Tom Harvey, Marketing Office Depot, 2151 East 1st Street, Santa Ana, California.” The name was defendant’s alias, and the address was that of a hotel which defendant temporarily called home. Defendant made many interstate phone calls to his victims in furtherance of this scheme.
Before Carroll could dupe his 18th victim, his scheme was foiled by Kenneth Caves, a savvy would-be victim who, upon hearing Carroll’s spiel, contacted the IRS. Investigator John Glugos responded by setting up a sting operation which ultimately led to Carroll’s arrest on January 8,1996. On January 26, 1996, a grand jury returned a two count indictment charging Carroll with wire fraud in violation of 18 U.S.C. § 1341 and mail fraud in violation of 18 U.S.C. § 1343. Pursuant to a written plea agreement entered March 13, 1996, Carroll pled guilty to the wire fraud count in exchange for the government’s agreement to drop the mail fraud charge.1
Defendant was sentenced to 30 months imprisonment, based on a total offense level of 13 and a criminal history category of IV, and was ordered to make restitution for his fraud in the amount of $12,650. In arriving at this sentence, the court adopted the Pre-sentence Investigation Report (“PSR”) recommendation that defendant receive a total of eight criminal history points, four of which were attributed to sentences Carroll received for two prior convictions.
The two prior sentences that the district court counted in calculating Carroll's criminal history were the result of two offenses he committed in 1985. In the first offense, Carroll stole four checks from his supervisor at Nelson Aved Technologies (“NAT”) where he was employed. On April 22, 1985, Carroll forged these checks and deposited them into his bank account. For this, Carroll was convicted of forgery in Walnut Creek Municipal Court, Walnut Creek, California (“the forgery offense”), and was sentenced to 60 days imprisonment. In the second offense, Carroll (now an “ex-employee” of NAT for obvious reasons) visited a client-lessee of NAT on June 18, 1985, falsely represented himself as a NAT employee, and picked up laser disc equipment belonging to NAT. Carroll did not return the equipment to NAT. For this, Carroll was convicted of grand theft property in Merced Superior Court in Los Banos, California (“the theft offense”), and received a sentence of nine months imprisonment and three years probation.
On appeal, defendant makes two challenges to the use of these prior sentences in calculating his criminal history. First, Carroll contends that the 1985 sentences should be counted together in computing criminal history because the underlying forgery and theft offenses were “related” as defined in Sentencing Guideline § 4A1.2(a)(2). Second, he argues that the method used in § 4A1.2(e) to calculate whether the prior sentences are more than ten years old (and thus countable toward criminal history) is unconstitutional. We assess both challenges pursuant to our jurisdictional power under 28 U.S.C. § 1291.
II. “Relatedness” of Prior Convictions
The length of a defendant’s sentencing range under the Guidelines is determined by the combined consideration of two factors: the offense level of the offense of conviction, and defendant’s criminal history category. The latter factor serves to increase a defendant’s sentence to reflect his propensity for recidivism. In calculating a defendant’s criminal history, the Guidelines at §§ 4Al.l(a)-(f) instruct a sentencing judge to add points for each of defendant’s qualifying prior sentences. However, under § 4A1.2(a)(2), if a defendant has served prior sentences for cases that were “related,” those sentences are treated as only one prior sentence for purposes of calculating criminal history.
Our first question today is whether Carroll’s two sentences for forgery and theft were “related” for § 4A1.2(a)(2) purposes. *460The district court found that they were not, and accordingly counted both prior sentences separately in calculating Carroll’s criminal history. Carroll of course argues that they are related, and if he is correct, stands to benefit by receiving a reduced sentence.
A Sentencing Guideline Application Note that interprets or explains the application of a Guideline is binding on our court unless it violates the Constitution or a federal statute, or unless it conflicts with, or constitutes a plainly erroneous reading of, that Guideline. Stinson v. United States, 508 U.S. 36, 113 S.Ct. 1913, 123 L.Ed.2d 598 (1993). Application Note 3 to § 4A1.2 is such an interpretive Note, and guides our inquiry today. Application Note 3 provides that prior sentences are considered related “if they resulted from offenses that (1) occurred on the same occasion, (2) were part of a single common scheme or plan, or (3) were consolidated for trial or sentencing,” unless the sentences were for offenses that were separated by an intervening arrest. U.S.S.G. § 4A1.2, comment (n.3). Of these three bases for “relatedness,” Carroll relies on the second, arguing that the forgery and theft offenses in 1985 were part of a “common scheme or plan” to defraud his employer.2
Under the law of our circuit, offenses are considered part of a “common scheme or plan” only if the defendant can prove that he either “jointly planned” the crimes or planned one crime that “would entail the commission of the other as well.” United States v. Ali, 951 F.2d 827, 828 (7th Cir.1992). Carroll offers no evidence that commission of the forgery entailed commission of the theft, attempting proof only that the forgery and theft crimes were jointly planned. He supports this claim by first pointing to several common characteristics of the two crimes: the nature of the crimes was similar, the crimes occurred in close temporal proximity, and the crimes had the same “victim” — Carroll’s employer. Neither close temporal proximity nor similar nature of the crimes commands a finding that the defendant jointly planned the crimes. See, e.g., U.S. v. Woods, 976 F.2d 1096, 1099 (7th Cir.1992) (holding three robberies committed eight days apart not “related” — mere repetition of similar criminal conduct insufficient). And although we would agree that whether the same victim is targeted can be persuasive circumstantial evidence of joint planning, see United States v. Manarite, 44 F.3d 1407 (9th Cir.1995) (“same victim” part of multi-factor test for common scheme or plan); United States v. Williams, 10 F.3d 910 (1st Cir.1993) (finding relatedness in part on evidence of same victim), it is not persuasive here because Carroll gives no evidence that his victimization of NAT was a planned attack rather than the result of mere opportunity. At most, Carroll has shown that he took advantage of two distinct opportunities to add to his personal coffers, opportunities which arose through his employment relationship with NAT.
Defendant also argues that the crimes were jointly planned because in each case he intended to steal from his employer. This argument also misses the mark. Crimes are not related merely because each was committed with the same purpose. See United States v. Brown, 962 F.2d 560, 564 (7th Cir.1992). Forming the same intent at two distinct times to commit the same crime (such as steal from an employer) does not evidence joint planning. Rather, defendant must show that he either intended from the outset to commit both crimes or that he intended to commit one crime which, by necessity, involved the commission of a second crime. See Woods, 976 F.2d at 1099, citing Ali, 951 F.2d at 828.
We agree with the district court that Carroll has shown neither intention. Although the proper standard of review in “related offense” cases is cloudy, see Woods, 976 F.2d at 1099, we are clear that the district court’s decision survives even de novo review. While Carroll insists that his goal was to *461defraud his employer, his testimony to this effect was unpersuasive to the trier of fact, see Brown, 962 F.2d at 565, and the circumstantial evidence of joint planning he provides supports only the “similarity, not singularity,” of the offenses. United States v. Yahne, 64 F.Sd 1091, 1096 (7th Cir.1995).
III. Equal Protection
Defendant then claims that neither of his 1985 sentences should be counted toward his criminal history, because Guideline § 4A1.2(e), the section that measures the recency of prior sentences for inclusion as relevant conduct, is unconstitutional. The constitutionality of Sentencing Guideline provisions is a question of law which the appellate courts review de novo. United States v. Turner, 93 F.3d 276, 286 (7th Cir.1996); United States v. Robinson, 958 F.2d 268, 270 (9th Cir.1992); United States v. Guajardo, 950 F.2d 203, 206 (5th Cir.1991).
Not all prior sentences are counted in calculating a defendant’s criminal history category. Under §§ 4A1.1 and 4A1.2(e), whether a prior offense is used to calculate criminal history (and thus increase the sentence) depends on how recent the defendant received punishment for the prior offense. Under the Guidelines, because Carroll’s forgery and theft sentences were each over 60 days but under 13 months, they are not counted if they were imposed more than ten years prior to Carroll’s commencement of the wire fraud offense. U.S.S.G. § 4A1.1, comment (n.2). In calculating this ten year period, the sentencing court measures the time between the date that each prior sentence was imposed and the date on which defendant commenced the instant offense. U.S.S.G. § 4A1.2(e)(2). Following this measuring scheme, the district court found that both of Carroll’s prior sentences fell within the 10-year period, and counted both toward his criminal history ranking.
Carroll complains that measuring the ten-year period by reference to the date on which the prior sentences were imposed, as opposed to the date on which the prior criminal conduct occurred, violates the Fourteenth Amendment of the United States Constitution. In his reply brief, Carroll presents the hypothetical of two defendants, each now being sentenced for a recent offense, and each sentencing court looking back to a crime that the defendants committed on the same day just over 10 years ago. The defendant who pleads and is promptly sentenced could fall outside the 10-year inclusion period, while the defendant who chooses to exercise his right to a jury trial, thus prolonging sentencing, might not.3 This disparate treatment of similarly situated defendants, argues Carroll, violates the Equal Protection Clause. To avoid this problem, Carroll argues that the court must use the date on which the prior offense was committed instead of the date sentence was imposed. Not surprisingly, use of this alternative look-back date would result in neither of Carroll’s prior sentences being counted toward his criminal history.
We reject Carroll’s constitutional challenge. The use of any specific measuring scheme in calculating criminal history will necessarily involve iine-drawing likely to be offensive to the defendant who just misses the cutoff, as defendant did here. And although the unequal treatment scenario Carroll hypothesizes is possible, he must do more than suggest a different line-drawing technique that favors his situation. Because the group “convicted criminals” is not a suspect class requiring stricter scrutiny, Carroll must show that the challenged Guideline provision is not rationally related to a legitimate government interest. See Chapman v. United States, 500 U.S. 453, 465, 111 S.Ct. 1919, *4621927-28, 114 L.Ed.2d 524 (1991).4 He has not.
We find that the measuring scheme in § 4A1.2(e)(2) is rationally related to the legitimate government interest of administrative efficiency in application of the Guidelines. See, e.g., Richards v. Lavelle, 620 F.2d 144 (7th Cir.1980) (administrative efficiency is a legitimate interest on rational review). Use of the date of sentencing of the prior crime provides a court with an easily determinable look-back date which aids in the efficient application of the Guidelines, and supports uniformity at sentencing. It is certainly rational for Congress to conclude that requiring a sentencing court to determine when a crime committed ten years earlier first commenced — Carroll’s suggested alternative to the current scheme — is too burdensome a requirement. The sentencing court has before it the full trial record only of the instant crime. That record enables it to make findings of fact with regard to that crime, including the date the criminal conduct first occurred. In contrast, it is much more difficult for a court to determine the date of conduct for a ten-year-old crime. Instead of having the full court record of the past crime before it, the sentencing court is often aided only by the record of judgment in the prior offense. Even if the full record of the prior criminal trial were brought before the sentencing court, the prior court may not have made relevant findings on the specific date of criminal conduct. Although additional evidence could be presented on this issue before the sentencing court, Congress rationally could conclude that judicial efficiency requires avoiding delving so deeply into collateral matters.
We need not comment on whether Carroll’s suggested alternative is also a rational measurement scheme, because the advancement of such an alternative, even if rational, does not disprove the constitutionality of the scheme now used in the Guidelines. The application of § 4A1.2(e) requires a sentencing court to look back precisely ten years to determine whether a prior sentence should be included in criminal history. Congress’ decision to set the look-back date as the date of prior sentencing, a date easily determinable and accessible to the sentencing court, is rationally related to their legitimate interest in the efficient application of the criminal history rules of the Guidelines. As such, the Guideline meets Constitutional requirements.
The sentencing determination of the district court is Affirmed.