ORDER AND OPINION
HODGES, Judge.
Mr. Van Cleave was separated from the Navy in July 1997 with a ten percent disability rating for migraine headaches and $46,720.80 in severance pay. Plaintiff sought an adjustment from the Board for Correction of Naval Records in August 1999 to reflect a disability rating of at least thirty percent. The adjustment would have placed plaintiff on the permanent disability retirement list and entitled him to disability retirement pay.
The BCNR denied plaintiffs application for relief. The Board’s rationale for denial was arbitrary and capricious, and its ruling on the merits is not supported by the record. We grant plaintiff Van Cleave’s motion for summary judgment.
BACKGROUND
Mr. Van Cleave was plagued by severe headaches from July 1996 through February 1997. He experienced eight to ten such headaches per month and was prescribed medication for migraines. He had symptoms consistent with migraine headaches, such as photophobia, which required him to lie down in a dark room until the headache dissipated. A Medical Board recommended that Petty Officer Van Cleave be referred to a Physical Evaluation Board for medical discharge.
The Physical Evaluation Board ruled in April 1997 that Van Cleave was not fit for further military duty and awarded him a preliminary disability rating of ten percent for migraine headaches.1 The PEB noted that “the disability may be permanent.” The Navy notified plaintiff of the PEB’s action in May 1997, and a Navy Disability Counselor explained his alternatives. The counselor presented Mr. Van Cleave with a form containing three options: (1) accept the preliminary findings of the PEB and waive his right to a formal hearing; (2) accept the preliminary findings conditionally, subject to any reservations that he wished to state; or (3) contest the preliminary findings and demand a formal hearing. Van Cleave accepted the preliminary findings and waived his remain[676]*676ing rights. The Navy discharged plaintiff in July 1997.
Plaintiff discovered later that the Physical Evaluation Board had received an incorrect diagnosis from his Medical Board. Van Cleave’s doctors had been treating him for migraine headaches. He assumed that the Medical Board had made a clerical error in its use of the term “chronic headaches.”
Plaintiff appealed the PEB’s decision to the Board for Correction of Naval Records in August 1999. He argued that the Physical Evaluation Board did not understand that he had been diagnosed with migraine headaches rather than frequent or chronic headaches. Had the Board known that he had frequent migraines, plaintiff believed, his disability rating could have been thirty percent or more. The BCNR denied Van Cleave’s appeal by letter dated March 14, 2000:
As the rating you received for chronic headaches was based on the same rating criteria as those applicable to migraine headaches, the rating would not have been any higher had the headaches been formally classified as migraines. The Physical Evaluation Board based its ratings in large part on the contents of your medical board report, which indicates that ... your headaches ... ‘lasted many hours’ ... and occurred every three to four days____The Board was not persuaded that you[r] headaches were severe enough [to meet] the criteria for 30% rating____
The BCNR further concluded that the records before it did not establish that Mr. Van Cleave suffered from prostrating attacks averaging once per month in the months preceding his discharge. The BCNR said “prostrating attacks” meant those “of such severity as to be incapacitating, i.e., that you would have to stop what you were doing and seek medical attention.” The Board denied plaintiffs motion for reconsideration on July 26, 2000.
JUDICIAL PROCEEDINGS
Mr. Van Cleave appealed the Board’s decision to this court. He acknowledged that he signed the form outlining his options voluntarily, accepted the Physical Evaluation Board’s findings, and waived his right to a formal hearing. He stated that he would not have waived his rights had he known that he might have been entitled to a thirty percent disability rating. Van Cleave pointed out that doctors had treated him for migraines, and his Commanding Officer recommended that plaintiff be transferred “due to a pending disability discharge for severe disabling migraine headaches____”
Defendant took the position that plaintiff had waived all administrative and judicial review, but in any event he was not prejudiced because the Board based its disability rating on Veterans Administration Code 8100, which is the code for migraine headaches. See 38 C.F.R. § 4.124a. If the PEB received incorrect information, the PEB used the correct diagnosis in calculating Van Cleave’s disability rating. In other words, the Government’s position was then, and has been until recently, that any misrepresentation of plaintiffs condition was immaterial because he was rated under the correct diagnosis.
Van Cleave pointed out that the 8100 VA Code provides a range of ratings from zero to fifty percent depending upon the frequency of migraines. If the PEB had known that all of his headaches were migraines, it might have given him a higher rating.2
We dismissed plaintiffs appeal upon the Government’s motion because Mr. Van Cleave had waived further review voluntarily, in compliance with applicable regulations. See Van Cleave v. United States, 60 Fed.Cl. 291 (2004) (Van Cleave I). Plaintiff appealed. The Federal Circuit ruled that volun-tariness of the waiver was not the sole issue upon review. Van Cleave v. United States, 402 F.3d 1341 (Fed.Cir.2005) (Van Cleave II). The Circuit asked that we determine the scope of plaintiffs waiver in such circumstances:
Did it include a waiver of any further administrative review of the PEB’s now-[677]*677final decision? Apparently the Navy did not so read it, since it granted Van Cleave’s petition for a subsequent review of the ease by the BCNR. Did it include a binding waiver of any right to judicial review? That for us is the $64 question.3
Id. at 1344.
We found that plaintiff’s voluntary acceptance of the informal PEB decision waived review by a formal Physical Evaluation Board, but did not necessarily prevent his appeal to the Board for Correction of Naval Records, which that Board accepted and considered on the merits. See Van Cleave, 402 F.3d at 1344 (“Apparently the Navy did not so read [plaintiffs waiver as precluding further administrative review], since it granted Van Cleave’s petition for a subsequent review of the case by the BCNR.”).
REMAND TO THE CORRECTIONS BOARD
We have jurisdiction to review rulings of the BCNR, irrespective of how plaintiffs appeal came to that Board’s attention. See Van Cleave v. United States, 66 Fed.Cl. 133 (2005) (Van Cleave III). Moreover, the Federal Circuit’s suggestion that the Navy did not consider plaintiffs waiver to have encompassed the Corrections Board or the court was persuasive. See Van Cleave, 402 F.3d at 1344. This court’s review of a BCNR decision is limited, however. See Van Cleave III, 66 Fed.Cl. at 136 (“We do not review the underlying medical determination of the Physical [Evaluation] Board; review is limited to the Correction Board’s denial of Mr. Van Cleave’s application for review on the basis of the record before it.”) (citing Pope v. United States, 16 Cl.Ct. 637, 640 (1989)).
We expressed concern that Mr. Van Cleave might have been denied benefits improperly because the BCNR had rejected his appeal on the basis that his condition was insufficiently severe to justify a thirty percent disability rating. Van Cleave III, 66 Fed.Cl. at 137. The only factor that distinguishes a ten percent rating from a thirty percent rating is the frequency of the attacks, not their severity. If plaintiffs medical records demonstrate that he suffered from chronic headaches and the VA Schedule Code for migraines is the appropriately analogous condition, his disability rating should have been reassessed based on the frequency of his prostrating attacks. Instead, the BCNR rejected plaintiffs appeal because the Physical Examination Board found that Mr. Van Cleave’s condition was not sufficiently severe to justify a thirty percent disability rating.4 We asked the Board to clarify this apparent inconsistency.
The remand order pointed out that the record did not distinguish between attacks that were prostrating and those that were not, and did not address directly the frequency of prostrating attacks. If no reason appeared to doubt that all or most of Mr. Van Cleave’s attacks were prostrating, then his disability rating should have been reassessed based on frequency alone.5
THE BCNR’S RESPONSE
The Board concluded that “the evidence submitted was insufficient to establish the existence of probable material error or injus[678]*678tice [and that] [i]t was not persuaded that [Mr. Van Cleave] should have been assigned a disability rating of 30%____” The BCNR’s reason for this conclusion was that it did not believe Mr. Van Cleave described his migraine symptoms truthfully. For example, the Board stated that the words “migraine headaches” appearing in the medical records were derived from what Mr. Van Cleave told his physicians. Moreover, “[t]he fact that [Mr. Van Cleave was] given trials of numerous medications normally prescribed to treat migraine headaches d[id] not suggest a diagnosis of migraine headaches ____” Mr. Van Cleave “ha[d] not demonstrated that [he] suffered from prostrating headaches while in the Navy, with the possible exception of the headaches that occurred on 8 and 12 July 1996.” The Board wrote that Mr. Van Cleave’s “habit of lying down in a dark room when a headache occurred does not demonstrate that the headaches were prostrating.”
The BCNR claimed that Mr. Van Cleave’s “subjective reports concerning the alleged frequency and severity of [his] headaches cannot be a basis for relief because there [we]re significant questions about [his] credibility.” Mr. Van Cleave was “referred to medical and physical evaluation boards because [he] wanted to be relieved from sea duty and referred for disability evaluation,” according to the Board. Nothing in the record supports such a conclusion. The BCNR made other unsupported statements about Mr. Van Cleave:
The Board does not believe that your choice [to waive a formal PEB hearing] resulted from your ignorance of the findings of the medical board and PEB ... It is more plausible that your choice was motivated by your conclusion that it was in your best interest to accept the findings of the PEB. You had difficulty maintaining your weight and body fat within acceptable limits throughout most of your career, and your prospects for promotion to chief petty officer, and perhaps even remaining on active duty, would have been adversely affected had you not been able to lose weight.
DISCUSSION
Plaintiff asserts (1) that the BCNR often misconstrued facts, made conclusory and unsupportable findings, and used only self-serving facts in the record while neglecting to explain those supporting plaintiff; (2) that the weight of the evidence demonstrates that plaintiff suffered from migraines and that he suffered at least one prostrating attack per month; (3) that procedural mistakes by the agency rendered its decision arbitrary and capricious; and (4) that the Board’s ruling establishes that the Government acted in bad faith.
The Government maintains that we must uphold the BCNR’s decision. It argues that the Board considered all issues on remand as directed by this court, reviewed the supplemental materials provided to it, and correctly determined that no new evidence justified increasing Mr. Van Cleave’s disability. The Government contends that it did not commit prejudicial error or injustice, and that the Board based its conclusions upon substantial evidence in the record and applied the appropriate regulations. Defendant refutes allegations that the Government acted in bad faith.
Legal Standards
Rule 56.1(a) governs motions for judgment on an administrative record developed before an agency. See Banknote Corp. of Am,., Inc. v. United States, 365 F.3d 1345, 1352 (Fed. Cir.2004). Our review of the Government’s actions in military pay cases “is normally limited to the administrative record developed before the military board.” Bateson v. United States, 48 Fed.Cl. 162, 164 (2000). We do not review decisions involving correction to military records de novo. See Patterson v. United States 44 Fed.Cl. 468, 471 (1999). Generally, we afford the agency decision substantial deference. Pope, 16 Cl.Ct. at 641. “[R]eview of the administrative decision is limited to determining whether the ... action was arbitrary, capricious, or in bad faith, or unsupported by substantial evidence, or contrary to law ____” Heisig v. United States, 719 F.2d 1153, 1156 (Fed.Cir. 1983) (citations omitted).
We do not serve as a “super correction board.” Skinner v. United States, 219 Ct.Cl. 322, 327, 594 F.2d 824 (1979). ‘When sub[679]*679stantial evidence supports a board’s action, and when that action is reasonable in light of all the evidence presented, the court will not disturb the result.” Pope, 16 Cl.Ct. at 641 (citing Holman v. United States, 181 Ct.Cl. 1, 8, 383 F.2d 411 (1967)). However, correction boards must examine relevant data and articulate satisfactory explanations for their decisions. Yagjian v. Marsh, 571 F.Supp. 698, 701 (D.N.H.1983) (citing Motor Vehicle Mfrs. Ass’n of the U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983)). This includes making rational connections between the facts found and the choices made. Id. We may find a correction board’s decision arbitrary and capricious if the board fails to consider an important aspect of a problem, offers an explanation for its decision that runs counter to the evidence before the board, or “is so implausible that it could not be ascribed to a difference in view or the product of agency expertise.” Id. at 706 (quoting Motor Vehicle Mfrs. Ass’n, 463 U.S. at 43, 103 S.Ct. 2856).
The role of a correction board is to decide if an applicant “has demonstrated the existence of a material error or injustice that can be remedied effectively through correction of the applicant’s military record, and, if so, what corrections are needed to provide full and effective relief.” Roth v. United States, 378 F.3d 1371, 1381 (Fed.Cir.2004). ‘When a correction board fails to correct an injustice clearly presented in the record before it, it is acting in violation of its mandate.” Id. (citation omitted).
Review of the BCNR’s Decision
This court assumes the regularity of military records and the good faith of government officials. See, e.g., Doe v. United States, 132 F.3d 1430, 1434 (Fed.Cir.1997); Armstrong v. United States, 205 Ct.Cl. 754, 762 (1974) (stating presumption that “the BCMR and military personnel ... performed their functions properly”). However, the rulings of the Board for Correction of Naval Records in this ease are not acceptable. The Board did not resolve the inconsistent findings for which we sought clarification in the remand order. It attempted to address them not by reconsidering them on the basis of the record before it, but by launching personal attacks on plaintiffs character and credibility. Rather than showing how its findings could be read in concert with the facts of the case, it accused Mr. Van Cleave of malingering and lying about his symptoms. It also questioned the integrity of the military’s system for rating disabilities, a process of which the Board is a part. None of these accusations was substantiated by the record.
Regulations authorizing the BCNR to review cases such as plaintiff’s direct the BCNR to review the facts and the law to insure that a service member has not been treated unfairly. The Board’s “function is to consider applications properly before it for the purpose of determining the existence of error or injustice in the naval records of current and former members of the Navy ____”32 C.F.R. § 723.2(b). The Board must explain its reasons for denying an applicant’s claims “together with all the essential facts upon which the denial is based ____” Id. § 723.3(e)(4). The primary “essential fact” the Board used for denying plaintiffs application was its assertion that he could not be believed. The Board stated, “[yjour subjective reports concerning the alleged frequency and severity of your headaches cannot be a basis for relief because there are significant questions about your credibility.”6
The BCNR’s Conclusion that Plaintiff did not Suffer from Migraines is Error
The Board concluded that plaintiff did not have migraine headaches during his final year of active service in the Navy, but chronic, recurrent headaches. Navy physicians noted the possibility of migraines, however. A July 21, 1996 examination chart stated, “Recurrent [headache], consider migraine vs. cluster [headache].” A July 28, 1996 exami[680]*680nation chart is marked “migraine vs. cluster vs. tension [headache].” A March 4, 1997 clinical evaluation cites one of Mr. Van Cleave’s conditions to be “migraine [headache].” A March 19,1997 medical chart lists a “summary of effect and diagnosis” as “migraine [headache].”
The Board dismissed such evidence from plaintiffs medical records because the entries were derived from plaintiffs own descriptions of his symptoms. Physicians must rely on their patients’ reports of symptoms in making diagnoses of many ills. The record does not show that plaintiffs doctors had available to them other, more objective procedures or mechanisms for diagnosing migraine headaches.
Plaintiffs doctors treated him for migraine headaches. Depakote, Fiorinal, Midrin, Nor-tryptiline, Sumatriptan, Verapamil, and supplemental magnesium were among the drugs they prescribed for Mr. Van Cleave’s condition. He had a CT scan, an MRI, and a lumbar puncture to rule out other, more serious illnesses. The Board argues that because none of these drugs or treatments was effective in reducing the frequency of his attacks suggested that Mr. Van Cleave did not suffer from migraines. Despite having taken these drugs, he saw no reduction in the frequency or the severity of his attacks. The Board thought this was a “strong indication that [he] did not suffer from migraine headaches.” Such a finding based on this record is all but irrational. The Board could not reasonably have concluded that a service member did not suffer from migraines merely because his medication failed.
The Board emphasized that Mr. Van Cleave’s “trials of numerous medications normally prescribed to treat migraine headaches d[id] not suggest a diagnosis of migraine headaches ....” In other words, the fact that a series of doctors prescribed such medicines did not suggest that they believed Van Cleave to have been suffering from migraines; yet the failure of such medications to cure plaintiffs migraines proves that plaintiff did not have migraines. Where the record shows that plaintiffs physicians prescribed a course of treatment common for migraine headaches, it is reasonable to conclude that the doctors thought it likely that he suffered from migraines. We are not aware that physicians routinely prescribe medications for ailments from which they think their patients do not suffer, nor does the record show that Mr. Van Cleave’s doctors followed such a protocol here.
The BCNR claims that plaintiffs medical records do not evidence any of the symptoms that normally accompany migraine headaches, such as vomiting, nausea, vertigo, or photophobia. The record shows that in the emergency room following the onset of a severe headache Mr. Van Cleave reported having problems with his vision. His head pain was severe and unilateral, on his left side. When he experienced such attacks, he could obtain relief only by lying down in a dark room.7 Such behavior is consistent with photophobia—a classic migraine symptom. Migraine headaches typically occur over the patient’s left or right eye.
Plaintiffs private physician, a neurologist, stated that Mr. Van Cleave had migraine headaches, as did his VA doctors. Plaintiffs own description of his symptoms strongly suggest migraine headaches to anyone who has experienced them. The Government criticizes Van Cleave for obtaining diagnoses of migraines based solely on plaintiffs own descriptions to his physicians. As noted, however, the record does not show that doctors have objective tests for migraine headaches or other means of confirming this ailment. The Board does not say how doctors should have diagnosed plaintiffs condition without asking Mr. Van Cleave how the headaches affected him.
The Material Issue for the BCNR was Whether Plaintiff Suffered from Prostrating Attacks
A finding that plaintiff suffered from migraine headaches is not necessary to the [681]*681Correction Board’s finding that plaintiff was entitled to a higher disability rating. The Physical Evaluation Board assigned Mr. Van Cleave’s rating based on the migraine code in the VA Schedule for Rating Disabilities. Though the PEB considered plaintiff unfit for duty because he had chronic or recurrent headaches, the statutory scheme permits ratings by analogy. The analogous rating the PEB employed was that used for migraines. A service member must have experienced “characteristic prostrating attacks occurring on an average once a month over the last several months” to qualify for a thirty percent rating pursuant to the VASRD standards. A Defense Department Instruction defines a “prostrating” attack as one that causes the “[s]ervice member [to] stop what he or she is doing and seek medical attention.” See SECNAVTNST 1332.39.8
The BCNR said plaintiff did not experience prostrating attacks in November and December 1996, or in January and February 1997, the four months leading up to his Medical Evaluation Board. The BCNR found that plaintiffs visits with physicians were regularly scheduled appointments, and the records from those visits do not show that he was experiencing prostrating attacks at the time. Mr. Van Cleave’s appointments may not have been concurrent with his attacks, but this does not compel the conclusion that he had no prostrating attacks during that period. Migraines are debilitating; it is entirely understandable that he would be unable to see a physician while experiencing an attack.
Mr. Van Cleave’s physicians altered his course of treatment often because nothing seemed to alleviate his condition. Indeed, plaintiffs only effective treatment was Tylenol #3, a prescription pain medicine that gave him some relief. He contacted his physician several times during these months for refills. He sought medical treatment by taking his physician-prescribed Tylenol # 3 when he experienced attacks. The attacks were prostrating, but he did not, and perhaps could not, see a physician at the time. At the least, it would have been impractical for one experiencing a migraine to travel and dangerous to drive. The doctor had prescribed medication for him to take when such attacks occurred.
The BCNR stated that the frequency of plaintiffs attacks decreased during the Fall of 1996. The record shows that Mr. Van Cleave’s attacks decreased from occurring every three to four days during August and September to every four to five days in October. Such a range of attacks may mean that Mr. Van Cleave’s headaches did not decrease at all; he could have had four attacks per week all Fall. If they did decrease somewhat, one less headache per week is hardly the “significant reduction” BCNR describes. It remains a frequency that is sufficient to qualify him for a thirty percent rating under the VASRD. This standard requires that he endure “characteristic prostrating attacks occurring on an average of once a month over the last several months.” Record evidence shows that Mr. Van Cleave had to stop what he was doing and seek medical attention an [682]*682average of once a month over the last several months prior to his discharge.
Plaintiff Experienced Prostrating Attacks with Sufficient Frequency to Qualify for a Disability Rating Higher than Ten Percent
The BCNR’s first decision denied plaintiff relief based on the purported severity of his attacks. The Board wrote that it “was not persuaded that [his] headaches were severe enough [to meet] the criteria for the 30% rating ... as the available records d[id] not establish that [he] suffered from prostrating attacks averaging once per month____” We urged that the difference between a ten percent disability rating and a thirty percent rating rests on the frequency, not the severity, of his attacks. See Van Cleave III, 66 Fed.Cl. at 137.
Plaintiffs medical records showed that he suffered attacks more frequently than once a month. Id. His Medical Board concluded, “at the time of initial presentation, he had one [headache lasting many hours] every three to four days.” Id. Notes from an exam of Mr. Van Cleave on October 22, 1996 stated, “he ... had four or five [headaches] since his last visit on 1 Oct. 1996.” Id. Mr. Van Cleave had experienced four or five headaches in that three-week span. We found that he “qualified for a thirty percent disability rating based on headache frequency alone.” Id.
We made other findings in the remand order. For example, the Medical Board’s conclusion that “[t]he patient states clearly that he is unable to perform his duties deployed at sea because of [his headaches].” An October 1996 examination record reported that Mr. Van Cleave’s headaches “leave him unable to continue working effectively” and usually require that he “lie down in a dark room and allow the headache to take its course.” His medications are used for migraines rather than simple headaches. His “[m]edical notes contain references to migraines.” Id. at 138-39.
The Board’s response was to avoid such findings by making irrelevant attacks on Mr. Van Cleave’s character. When we expressed concern that the record did not distinguish between attacks that were prostrating and those that were not, the Board attempted to portray Mr. Van Cleave as a hypochondriac, none of whose attacks could have been prostrating because he was also a liar. The Board argued, “[t]he fact that a person retreats to a darkened room and lies down while suffering from a headache does not mean that the headache is prostrating.” It makes the gratuitous and irresponsible comment in a letter to Mr. Van Cleave that “a valetudinarian might be severely distressed by minor aches and pains and seek refuge in a darkened room, [but] a more hale individual might continue to work despite having multiple symptoms of a classic migraine headache____Your habit of lying down in a dark room when a headache occurred does not demonstrate that the headaches were prostrating.”
The Board does not believe that Mr. Van Cleave’s headaches were sufficiently “prostrating” because he did not seek medical attention each time, though it admits that Mr. Van Cleave’s first two attacks may have been prostrating. This means the Board concluded that Mr. Van Cleave developed migraine headaches suddenly, had only two of them, then just as suddenly outgrew the migraines but continued to have simple everyday headaches that any hale individual could tolerate and continue to work through. Plaintiff had no history of debilitating headaches before the two recognized by the Board. He immediately sought medical help when the first one occurred. Once he learned the gravity of his condition and obtained treatment from a neurologist, he had no need to see a physician each time an attack occurred. Mr. Van Cleave could only do as his physician instructed—take his medication and seek refuge in a dark place until the attack “ran its course.”9
Plaintiffs admission that he did not see physicians between January 15, 1997 and July 13, 1997 does not demonstrate that he did not experience prostrating attacks during this time. He had a prescription for Tylenol [683]*683#3 that he took upon onset of his pain. Between January 1997 and July 1997, his neurologist did not see plaintiff but already had directed him to take Tylenol # 3 when an attack occurred. He refilled Van Cleave’s prescription during this time, apparently recognizing that it was not necessary or productive for plaintiff to be examined in the doctor’s office for each headache. It is understandable that plaintiff would be unable to see a physician while experiencing a debilitating migraine headache.
The Board knew its argument that plaintiff did not have prostrating attacks after July 1996 could not be reconciled with the PEB’s award of a ten percent rating. The Board attempted to avoid this inconsistency by turning its attack on the PEB itself, claiming that the PEB awarded Mr. Van Cleave a higher rating than he deserved. The BCNR said “it is uncommon for a PEB to find a career service member unfit for duty because of a single condition rated at 0%, and as such a rating may not have withstood close scrutiny, [so] a rating of 10% was selected, even though it was not warranted.” (emphasis added). Such efforts to discredit the plaintiff, the PEB, and the entire ratings system to support its findings are transparent. They have served only to create internally inconsistent findings that are unsupported by the weight of evidence in the record, and to discredit the Board for Correction of Naval Records.
The Board now contends that plaintiff never was entitled to disability at any level, but the PEB awarded him the ten percent rating out of sympathy—or perhaps because a zero percent rating would not have “withstood close scrutiny.” Such explanations are not believable or plausible. We find no record support for BCNR’s conjecture that the PEB rated plaintiff out of sympathy or because the military does not usually find a service member unfit for duty then decline to award him disability.
BCNR Attacks on Government Officials and Procedures
The BCNR alleged that the PEB did not perform its duties properly. The Board guessed that the PEB assigned plaintiff the ten percent rating “out of sympathy” for Mr. Van Cleave, and not because he suffered from the requisite number of prostrating attacks. The Board attempted to preserve its findings by stating that none of plaintiff’s headaches was prostrating except possibly for two. As noted earlier, such a conclusion is not supported by the record. The BCNR claimed to believe the PEB issued a ten percent disability rating because a zero percent rating “may not have withstood close scrutiny” and that the PEB routinely assigns disability ratings that are “not warranted.”
The Board did not limit its criticism to the Agency it serves. It also attacked the Department of Veterans Affairs, effectively accusing the VA of improperly conducting its official duties: “VA rating officials routinely assign ratings based on the subjective reports of the person seeking a rating, even when such reports are clearly without merit. The Board has reviewed a number of cases where this occurred.” (emphasis added). If the Board has reviewed cases in which this occurred, that does not establish that it happened here. The Board gave no support for its serious accusations of impropriety by the Department of Veterans Affairs.10
The BCNR belittled the system that Congress ordered and the Secretary of the Navy created to provide counseling for military personnel who are making important career decisions. The Board decided that plaintiff could not be believed because “[he] relied on a summary of the [PEB’s] findings provided by [his] disability counselor, a mess management specialist chief petty officer, whose normal duties prior to becoming a counselor probably consisted of cooking, operating a dining facility, and/or managing military quarters.” Thus, one of the Board’s reasons for attacking Mr. Van Cleave’s credibility [684]*684was that he accepted the advice of a Navy counselor whose military background happened to be in dining hall management. This Disability Counselor nevertheless was an officer appointed by authority of the Navy Secretary to assist Mr. Van Cleave as required by law. See 10 U.S.C. § 1142(a) (“[T]he Secretary [of the Navy] ... shall provide for individual preseparation counseling of each member of the armed forces whose discharge or release from active duty is anticipated as of a specific date.”).
The Board’s efforts to impugn the integrity and ability of a chief petty officer “whose normal duties prior to becoming a counselor probably consisted of cooking____” is grossly inappropriate. The Board did not know this officer but decided that he could not be trusted because of its understanding of his previous military specialty. Stereotyping and prejudice cannot be proper bases on which to question the ability of a Naval officer or anyone else. The Board’s rejection of Mr. Van Cleave’s credibility because he relied on assistance from a benefits counselor assigned to him by the Navy is per se arbitrary and capricious.
The Board’s Personal Attacks Against Mr. Van Cleave
Perhaps recognizing that the record would not support its findings, the Board made extraordinary efforts to undermine Mr. Van Cleave’s credibility and his character. It attempted to show that the PEB was wrong in its characterization of the severity of plaintiff’s headaches and asserted that it awarded plaintiff a ten percent rating out of “sympathy.”
The Board stated that plaintiffs treatment with drugs often used for migraines did not reduce the frequency or the severity of his attacks, and this was a “strong indication that [he] did not suffer from migraine headaches.”11 Such a conclusion defies common sense. The record does not establish that any course of treatment is always efficacious in treating migraine headaches. Certainly, the fact that a course of such treatment did not “cure” plaintiffs migraines is not evidence that he did not suffer from migraines. If a physician prescribes a course of drug treatment common for migraine headaches, it is reasonable to conclude that the treating physician believed the patient suffered from migraines. The Board’s ruling suggests that doctors routinely prescribe strong medications for ailments from which they do not believe their patients genuinely suffer.
The author of the BCNR’s ruling trivialized plaintiffs migraine headaches, often using peculiar observations such as, “a valetudinarian might be severely distressed by minor aches and pains and seek refuge in a darkened room, [but] a more hale individual might continue to work .... ” Such a “more hale individual” would have to continue working through symptoms described by the Board itself as including “severe pain, nausea, vomiting, sensitivity to noise, photopho-bia and vertigo.” The Board does not say what sort of job a hale individual could continue performing while experiencing such symptoms.12
The Board referred to plaintiff as “a frequent consumer of medical services” and criticized him for seeking treatment of supposed “minor conditions” and reporting to emergency rooms “on multiple occasions for non-urgent care.”13 This portrayal of Mr. Van [685]*685Cleave was a further attempt to discredit plaintiff as well as the severity of his condition. The BCNR speculated that plaintiffs case was nothing but an attempt to be relieved from sea duty. This is an inappropriate, improper, and conclusory statement that has no support in the record. Moreover, it is inconsistent with the Navy’s evaluations of Mr. Van Cleave as quoted by the BCNR in its decision:
You reportedly exceeded all standards in the section of the form entitled “Performance Traits,” with the exception of military b e aring/character, which is attributable to your overweight condition. You were lauded for such things as providing a high-quality work product in a timely manner, quickly and effectively processing all assignments, and often working extra hours to ensure that “everyone’s job gets done.” You were recommended for assignment to independent duty, as well as early promotion and retention.
The BCNR charged plaintiff with malingering throughout its decision, yet when it suited the Board’s purpose, it commented on plaintiffs high standards of performance. Its purposes were to bring plaintiffs “overweight condition” to the court’s attention once again; and to suggest that his migraine headaches must not have been severe enough to interfere with his performance. The Board managed to use even Mr. Van Cleave’s positive performance evaluations against him.
Such skeptical evaluations of Mr. Van Cleave’s headaches are internally inconsistent and frivolous. The BCNR’s comment that “a more hale individual might continue to work despite having multiple symptoms of a classic migraine headache, such as severe pain, nausea, vomiting, sensitivity to noise, photophobia and vertigo” is so absurd as to suggest animus. The Board’s description of such heroic efforts to remain on the job by “hale” individuals is an effort to make an unflattering comparison with Mr. Van Cleave’s “habit” of lying down in a darkened room to wait out a migraine. The Board knew or should have known that the need to seek a darkened room is among the few options that one has when a migraine strikes. Referring to that need as a “habit” is a transparent effort to portray Mr. Van Cleave as a weakling who is unworthy of belief.
Another attempt by the Board to justify its notions of plaintiffs lack of credibility was its “[cjoncern about ... [his] letter of 6 December 2006” regarding a response that he submitted to an advisory opinion sent to him by the Board. The BCNR stated that plaintiff “falsely claimed that [he] had been required to submit [his] reply” prior to the thirty-day period to which plaintiff thought he was entitled. (emphasis added). Anyone with the Board’s experience in these matters would know that if the thirty-day period did not apply in Mr. Van Cleave’s case, his belief that it did was merely a misinterpretation of applicable regulations or BCNR directives— not an uncommon circumstance. Such a misunderstanding cannot properly be described as a “false claim.” The Board’s deliberate usage of this term to describe a minor and understandable error provides additional evidence of bias and animus against Mr. Van Cleave.
Plaintiffs Claims of Bad Faith by the BCNR
Plaintiffs who claim that the Government acted in bad faith must overcome the strong presumption that public officials act in good faith and are “conscientious[ ] in the discharge of their duties.” Librach v. United States, 147 Ct.Cl. 605, 612, 1959 WL 7633 (1959). Courts have refused to find bad faith unless the government officials involved were “actuated by animus toward the plaintiff.” Kalvar v. United States, 211 Ct.Cl. 192, 543 F.2d 1298, 1302 (1976) (citing Librach, 147 Ct.Cl. at 614).
The BCNR’s willingness to issue a ruling that is nothing more than a personal and vindictive attack on a petitioner’s character is [686]*686certainly evidence of bad faith. The Board made numerous conelusory statements that were not supported by the record. The excerpts from its ruling included in this Opinion and Order speak for themselves. The Board has not shown that Mr. Van Cleave lacked credibility, and it had no basis for making such a claim, though this was its stated reason for denying plaintiffs appeal. Other attacks on plaintiffs character are similarly gratuitous, irrelevant, or wrong.
A finding of bad faith is not necessary in this ease because the Board’s rulings also are arbitrary and capricious, and unsupported by substantial evidence. The decision of the Board for Correction of Naval Records is a malicious attack that reads as though it were the product of a personal vendetta by its author.
CONCLUSION
We asked the BCNR to address this court’s concerns that the Board’s earlier ruling was inconsistent and to correct plaintiffs military records if appropriate. Instead, it launched an attack on Mr. Van Cleave’s credibility and character, criticized the disability counselor appointed by authority of the Secretary of the Navy, and questioned the integrity of Naval medical personnel and the professionalism of the VA’s “general medical officers.” Corrections Boards generally have acted with professionalism and sensitivity. They have undertaken their duties in an honorable and conscientious manner. This Board presents a grievous and unprecedented exception to that tradition.
Mr. Van Cleave has argued his own case throughout a long, frustrating process. We noted in an earlier opinion that “[p]laintiff appeared pro se and argued his case as well as anyone could have.” Van Cleave I, 60 Fed.Cl. at 293. He has been well prepared and knowledgeable of the record. He presented his ease orally and on brief effectively. He took his turn during conferences and never interrupted government counsel to make a point, despite the irresponsible attacks made on his character by the Board. Never has he acted improperly in his dealings with this court. Descriptions in the record of plaintiffs migraine headaches and his efforts to treat them ring true, as the Board should have recognized.
The administrative record before us provided convincing evidence that Mr. Van Cleave met the requirements for a disability rating higher than ten percent. The BCNR’s refusal to alter the original rating of ten percent assigned to him by the PEB was arbitrary and capricious. The record does not include substantial evidence to support the Board’s findings; its denial of plaintiffs appeal was erroneous. The Board’s inappropriate use of conjecture and conclusions unsupported by the record are not entitled to credibility or respect.
The BCNR concluded its letter to Mr. Van Cleave by reminding him that “the burden is on the applicant to demonstrate the existence of probable material error or injustice.” Plaintiff needs little more than an objective review of the Board’s ruling to meet that burden.
Defendant’s motion for judgment is DENIED. Plaintiffs motion is GRANTED. Defendant’s motion for leave to file additional pages of the administrative record is GRANTED.14 The parties will consult and file a joint status report no later than May 26, 2006, proposing an appropriate procedure for implementing this Opinion.
IT IS SO ORDERED.