MEMORANDUM OPINION

Introduction

In this appeal from administrative action, plaintiff contends he fell victim to an arbitrary and capricious Air Force promotion process. As a result, he claims he was unfairly and illegally denied promotion to the rank of Lieutenant Colonel. He seeks here the remedy twice denied him below by the Air Force Board for Correction of Military Records (AFBCMR): upgrading of certain performance reports or, alternatively, their removal from the record, and reconsideration by the Air Force of the question of whether he should be promoted to Lieutenant Colonel. Only a full recitation of the facts — which are essentially undisputed — will adequately illuminate the issues raised by plaintiff’s claim.

Facts

Plaintiff, Thomas J. Remy, is a retired Air Force officer. He was commissioned as a second lieutenant in the Air Force Reserve in December, 1967. In September, 1970, he was integrated into the regular Air Force. Remy served continuously on active duty from the date of his commissioning until his retirement. During this period, he was progressively promoted from second lieutenant to the permanent rank of major. Remy was first considered, but not selected, for advancement to the rank of Lieutenant Colonel by a promotion board which convened on July 26, 1982. Three subsequent promotion boards, from 1983 to 1985, also declined to select Remy for promotion. He retired with the permanent rank of major.

Defendant AFBCMR is an administrative board established within the Department of the Air Force pursuant to 10 U.S.C. § 1552(a). As Section 1552 provides, correction of military records is a function of the service Secretary who acts through a board of civilian employees of the executive part of the service department. Correction *1264of records is appropriate in those circumstances where it is necessary “to correct an error or remove an injustice.” 10 U.S.C. § 1552(a). Here, as the Air Force is the service involved, the appropriate board of civilians is AFBCMR. Initially, AFBCMR was the sole defendant. To eliminate any issue concerning whether plaintiff had named the proper defendant,1 the Court entered an agreed Order submitted by the parties adding the Secretary of the Air Force as a party defendant.

As noted, this appeal grows out of the Air Force’s decision not to promote Remy. Remy claims this occurred as a result of unfair and arbitrary and capricious ratings he received on certain Officer Effectiveness Reports (OERs). Hence, it is important to understand in some detail what OERs are and the regulations or policies that governed their completion at the times in issue. OERs are periodic evaluations of an officer’s performance and fitness for promotion. Until 1975, evaluations of an officer were completed by a “rater,” the subject officer’s immediate superior. These evaluations were in turn reviewed by an “endorser,” usually the rater’s immediate superior officer. Under this pre-1975 system, only the rater graded the subject officer; the endorser reviewed the rater’s work, but did not independently rate the officer under review. In the event an endorser disagreed with a rater, he or she merely noted this in the narrative comments.

This pre-1975 system used a two-digit rating system. The first digit represented a grade for overall performance and the second an assessment of promotion potential. The overall performance rating scale ranged from 0 to 9, the former signifying “unsatisfactory” performance and the latter “outstanding” performance. Appropriate qualitative descriptions existed for each of the numbers between 0 and 9. For example, a rating of 8 signified “exceptionally fine” performance. Promotion potential was measured on a smaller scale: 1 to 4, with a one awarded to those officers who “did not demonstrate any capability for promotion,” a two awarded to those recommended for promotion along with their contemporaries, a three for those demonstrating the capacity for promotion “ahead of contemporaries,” and a four for those meriting promotion “well ahead of contemporaries.” Thus, the highest rating one could earn, a “9/4,” signified outstanding performance and promotion potential well ahead of contemporaries. An “8/8,” the next step down on both scales, reflected exceptionally fine performance warranting promotion ahead of contemporaries. From 1969 until the system changed in 1975, Remy received the following ratings:2

January 1969 8/3

July 1969 9/4

October 1969 9/4

February 1970 9/4

July 1970 8/3

January 1972 9/4

July 1972 9/4

January 1973 9/4

July 1973 9/4

February 1974 9/4

May 1974 9/4

January 1975 8/3

In 1975, the Air Force changed its rating system.3 The new system called for three major changes. First, it used three levels of evaluation: a “rater,” a “second rater” and a “reviewer.” All three were required to evaluate the subject officer. The second change was a new, single overall rating scale ranging from 1 to 6, with one being *1265the highest rating available and six the lowest. Since all three — the rater, second rater and reviewer — were required to render an evaluation, each subject officer received a three-digit evaluation. Thus, a “1-1-1” OER means the officer received the highest rating from each member of the rating group. A “1-2-3” indicates the rater thought the subject officer worthy of the highest rating, while the second rater concluded he or she deserved the second highest rating and the reviewer awarded only the third highest rating.

The third major 1975 change is the genesis of this case. The Secretary, by regulation, imposed a quota on the top ratings. He did so, it appears, to control and remedy a serious ratings inflation problem that “had eroded the effectiveness of the OER as a management tool.” AF 36-10, Chapt. 6, § 6.1(a). The quotas imposed were straightforward. Of the officers being evaluated in a command, only 22% could receive a “1” and only 28% could receive a “2.” Thus a rating of “3” or below automatically consigned the subject officer to the bottom half of the evaluated group. This quota system was commonly referred to as the “controlled OER system.” The system imposed the quotas only on “reviewers” and “control points,”4 not on raters and second raters. Another significant, but unofficial, aspect of the controlled OER system was the use of seniority or time in grade as the principal factor in distributing the limited number of top ratings. The regulations neither permitted nor proscribed this ratings allocation practice; they were silent on this issue. Yet not surprisingly, perhaps, the practice was promptly engrafted onto the controlled OER system. Certainly it was used by the reviewer and second rater in this case. Apparently it was widely used. This is understandable. Those closest to promotion had the greatest need for the top ratings. If top ratings were to be limited, it must have seemed fairer that proximity to promotion be an important, if not decisive, allocation criterion.5

Significantly, the controlled OER system was short lived. It was discontinued on October 26, 1976, after only thirteen months of use. See Interim Message Change (IMC) 76-6. Evaluators were also instructed “not to use or consider” a number of factors including time in grade, date of rank and promotion eligibility status. Instead, they were instructed to consider job performance and, “peripherally,” such “whole person” considerations as education and career pattern. Id. Apart from the abandonment of the ratings quotas, however, the remaining essential elements of the 1975 changes continued in force.6 Under the post-1975 system, including the brief period of the “controlled OER system,” Remy received the following rat*1266ings:7

October 1975 2-2-3

September 1976 1-2-3

October 1977 2-2-2

April 1979 1-1-1

August 1979 1-1-1

February 1980 1-1-1

February 1981 1-1-1

February 1982 1-1-1

February 1983 1-1-1

January 1984 1-1-1

July 1984 1-1-1

July 1985 1-1-1

In 1975, Remy was a Captain assigned to Bitburg Air Force Base where he served as Base Airfield Manager. His immediate superior and rater was Major Melvin Hayes. Remy’s second rater and reviewer were, respectively, Colonel George Wehling and Colonel (later Brigadier General) Frederick Kyler. These individuals completed Remy’s October 1975 OER which is one of the two he complains of in this action. As noted, the numerical ratings were 2-2-3. A year later, Remy was still the Base Airfield Manager and he had the same second rater and reviewer, but a new rater, Major Roger Johnson. These individuals completed the second OER in issue, the September 1976 OER, which reflected a 1-2-3 rating.

Remy did not immediately complain about these OERs. Rather, it appears he resolved to persevere and do better. There is some evidence he succeeded. He went on to other duty stations. In August, 1979 he was promoted to the rank of major.8 Thereafter, however, in 1982 and 1983, Remy was considered but not selected for promotion to Lieutenant Colonel. The causes of this misfortune, he believed, were the October 1975 and September 1976 OERs.9 Accordingly, in May 1983, Remy filed a formal request to change or correct the October 1975 OER. In October 1983, he amended his request to include the September 1976 OER. In support of his request to correct these OERs, Remy submitted statements from Col. Wehling and Brig. Gen. Kyler, the second rater and reviewer, respectively, for both OERs.10 Both statements essentially allege that the ratings were determined, at least in large measure, by the quota system and by seniority. The following excerpts reflect the essence of each officer’s statement concerning the two OERs:11

October 1975 OER

Wehling: I concurred on the OER [dated October 31, 1975] and rated him a “2.” This was based on the quotas imposed by the OER system.... He had shown progressive improvement to the point that he was clearly one of the top captains in the wing and would have received the top rating if he were more senior at the time.

Kyler: My decision to downgrade Capt. Remy’s OER was solely influenced by the dilemma to distribute the available Is and 2s to officers who highly deserved promotion and were in close proximity to the primary cone. Capt Remy was one of the top captains in the wing, and should have received a 1, but due to his junior status, did not.

September 1976 OER

Wehling: I downgraded the succeeding report [dated October 31, 1976] from a one to a two for primarily the same reason — I gave a great deal of weight *1267to seniority in distribution of available ratings and Captain Remy was still junior. In retrospect I believe Capt. Remy should have received a “1” on that OER.

Kyler: I downgraded the succeeding report for primarily the same reason. I again gave a great deal of weight to the rating of more senior officers soonest eligible for promotion, to assure their promotion. Capt. Remy was downgraded because of his junior status. He should have received a “1” on that OER.

A staff reviewer for the AFBCMR recommended that additional information be obtained from Remy’s reviewer before disposing of the appeal. Specifically, three questions were posed. These questions and the reviewer’s responses are set forth here:

1. ...

(a) ...

Q. Were the controlled ratings awarded solely on date of rank? If not, cite all factors considered.

A. I considered factors that comprise the whole man concept such as leadership, scope of responsibility, depth of experience, job performance, performance under pressure, and future potential when awarding the controlled ratings. The [36th Tactical Fighter Wing] was manned with a highly select group of outstanding officers. Capt Remy was one of those officers. When the above factors were considered, those officers were equal in almost every respect and as a result I encountered a great deal of difficulty implementing the new controlled OER rating system. I discussed the system with my functional commanders and it became apparent that seniority of those officers soonest eligible for promotion became the deciding factor. Capt Remy was one of the top captains in the wing and should have received a 1, but due to his junior status did not.

(b) ...

Q: Were controls levied on the raters and additional raters? If so, what were the percentages?

A. My rating policy concerning OERs was the same as that established by the Air Force for wing commanders. In addition, I restricted each function commander, squadron commander, and division chief not to exceed the percentages as stipulated by AFR 36-10.

(c)...

Q. When was the practice of considering personal data in the evaluation process discontinued?

A. Personal data such as date of rank continued to play a very important part of my decision process until the controlled evaluation system was abolished.

2. Reference MPCAO letter, paragraph 3. The above factors were in effect during both 1975 and 1976 OER cycles.

Worth noting is the apparent inconsistency between paragraphs 1(c) and 2. The former suggests that the quota system was abandoned in timely fashion, while the latter indicates that it was used by the reviewer for the September 1976 OER even though IMC 76-6 was issued five days before the submission of that OER. The latter conclusion is supported by those portions of the reviewer’s statement set forth above.

Available to the AFBCMR for its consideration in this matter were some statistics concerning the number of “Is” Kyler, as a reviewer, awarded to captains during the 1975 and 1976 review periods. The statistics reflect that in 1975 Kyler awarded “Is” to three captains junior to Remy and in 1976 to fourteen such captains. Remy sharply criticized these figures as incomplete and as reflecting the end product of the decision-making, and not the process. Among the objections he raised was that the statistics compiled for 1975 inexplicably omitted captains in year group 1965, a group senior to him. This omission, however, does not affect the fact that a number of captains junior to Remy received the highest rating. Also available to the *1268AFBCMR were staff advisories12 commenting on the merits of Remy’s position and finding no regulatory violations and recommending denial of the application. One advisory pointed out that the 8/3 rating Remy received for the period immediately prior to the challenged October 1975 OER in fact placed him in the lower 25% of his contemporaries.13 Given this, the advisory opined that the “3” Remy received on the next OER, the October 1975 OER, was a “consistent evaluation trend.” Remy was afforded an opportunity to comment on all the staff advisories and he did so. Based on the entire record, including Remy’s comments,14 the AFBCMR denied the application, concluding that there was “insufficient relevant evidence” to “demonstrate the existence of probable error or injustice.” Remy’s claim, in the Board’s view, amounted chiefly to an impermissible challenge to, or attack upon, the regulations. (15 February 1984 Record of Proceedings [hereinafter “First Review”]).

Remy sought reconsideration. In support, he filed, by counsel, a brief arguing that the AFBCMR’s decision was based on inappropriate data. Specifically, Remy claimed that “recently discovered additional evidence ... seriously undermines” the statistics on which the AFBCMR relied. Remy’s new claim prompted yet another staff advisory. And again, the recommendation was “strongly” to deny Remy’s application. This conclusion was based on a more searching review of the ratings awarded by Remy’s raters and reviewer. It was again pointed out that in 1975 and 1976, captains junior to Remy received “Is,” indicating that seniority was not the determining factor, even in 1975 when it was a permissible factor. Moreover, the figures for both 1975 and 1976 showed that the reviewer awarded fewer “Is” and “2s” to the captain group than he had available under the quotas of the controlled OER system.15 According to the staff advisory, the figures also showed that, contrary to Remy’s claim, the raters apparently did not feel constrained by any quotas. In 1975, raters awarded “Is” to 28.3% of the group and “Is” and “2s” to 75.5%. In 1976, the comparable figures were 33.1% and 79.1%.16 There was, according to the staff advisory, no evidence of any regulatory violation. In sum, the staff advisory concluded that Remy’s claim was refuted by evidence (i) that captains junior to him received “Is” in both years, (ii) that his reviewer had “Is” and “2s” left over in 1975 and 1976, and (iii) that raters were apparently not constrained by any quotas. The AFBCMR accepted the staff advisory recommendation and again denied Remy relief. (10 February 1987 Record of Proceedings [hereinafter “Second Review”]). This appeal followed.

Analysis

Jurisdiction

Defendants initially claimed that this Court lacked subject matter jurisdiction. They argued that the relief requested by Remy — promotion to Lieutenant Colonel and award of back pay — violated the jurisdictional amount requirement of the “Little Tucker Act,” 28 U.S.C. § 1346(a)(2). This Act vests exclusive subject matter jurisdiction in the Claims Court for monetary claims against the government not sounding in tort where the amount in dispute exceeds $10,000. Jurisdiction is concurrent with federal district *1269courts only on claims of $10,000 or less.17 Significantly, however, where no monetary relief at all is sought, jurisdiction is exclusive in the district courts.18 Given this jurisdictional scheme, it is apparent that defendants’ jurisdictional attack would succeed if the relief requested had in fact included promotion and back pay. That, however, was not the relief actually sought. Instead, Remy relies on the Administrative Procedure Act, Section 702 (not the Tucker Act) and carefully limits the relief sought to removal or revision of the allegedly offending OERs and resubmission of his case to a Lieutenant Colonel promotion board.19 Should this relief be granted, in whole or in part, the matter would then be remanded to the Secretary for appropriate action, including possible submission of Remy’s case to a special selection board for promotion to Lieutenant Colonel. In any event, no monetary judgment would be required from this Court. Given this, it is readily apparent that jurisdiction is proper in this Court, but not the Claims Court.

Merits

Remy challenges an Air Force administrative decision. It is well settled that judicial review of such decisions is limited to determining whether the challenged decision “was arbitrary, capricious, contrary to law, or unsupported by substantial evidence.” Wronke v. Marsh, 787 F.2d 1569, 1576 (Fed.Cir.), cert. denied, 479 U.S. 853, 107 S.Ct. 188, 93 L.Ed.2d 121 (1986); Heisig v. United States, 719 F.2d 1153, 1156 (Fed.Cir. 1983); DeCicco v. United States, 230 Ct.Cl. 224, 677 F.2d 66, 70 (1982); Andrews v. Webb, 685 F.Supp. 579, 586 (E.D.Va. 1988). Remy’s challenge must be supported by “ ‘cogent and clearly convincing evidence.’ ” Wronke, 787 F.2d at 1576 [quoting Dorl v. United States, 200 Ct.Cl. 626, 633, cert. denied, 414 U.S. 1032, 94 S.Ct. 461, 38 L.Ed.2d 323 (1973)]. The analysis necessarily begins with scrutiny of Remy’s contention.

Remy’s contentions lack perfect clarity. Some arguably changed over time. Ultimately, the AFBCMR concluded, over Remy’s objection, that his contentions amounted, at least in part, to a challenge to the merits of the controlled OER system. As the AFBCMR viewed it, Remy was contending that a system of top ratings quotas allocated chiefly on the basis of seniority was inherently unfair and arbitrary and capricious. Such an attack must fail here.20 A rating and promotion scheme for Air Force officers, as well as a remedy for ratings inflation, is wholly within the power and discretion of the Secretary. See 10 U.S.C. § 8013 (Supp. 1987). This Court has neither the expertise nor the jurisdiction to substitute its judgment for the Secretary’s *1270with respect to an appropriate scheme for rating Air Force Officers. Id.; see also Boyd v. United States, 207 Ct.Cl. 1, cert. denied, 424 U.S. 911, 96 S.Ct. 1106, 47 L.Ed.2d 314 (1975). Accordingly, it is easy to see why Remy protests against construing his claims as a challenge to the merits of the OER regulations, for so construed his appeal would fail.

But Remy’s contentions, as the AFBCMR also plainly recognized, include more than a challenge to the regulations. They raise issues properly on appeal before this Court, concerning whether the challenged OERs were filled out in violation of any regulations and whether the AFBCMR’s decision was supported by substantial evidence or was arbitrary and capricious. Specifically, Remy’s contentions cognizable here on appeal raise the following questions:

(1) Whether the Wehling rating in the 1975 OER was awarded on the basis of a quota in violation of then existing regulations.

(2) Whether the ratings of Wehling and Kyler, as second rater and reviewer respectively, on the 1976 OER were awarded on the basis of quotas and time in grade or eligibility for promotion in violation of the existing regulations or instructions (IMC 76-6).

(3) Whether the 1975 and 1976 OERs are based on substantial record evidence.

The first and second questions have both legal and factual components. The first question requires this Court to determine whether, as a matter of law, the controlled OER system regulations prevented Kyler from imposing quotas on the raters and second raters below him in the rating chain. If so, the pertinent factual question becomes whether such quotas were imposed and followed in this case. Similarly, the second question requires that this Court determine whether, as a matter of law, the 1976 modification to the controlled OER system regulations prohibited the use of quotas and the consideration of seniority or eligibility for promotion in the assignment of ratings. If so, the Court must then determine whether either a quota, an improper seniority consideration or both were imposed and followed in this case. The third question inquires more generally whether the various conclusions reached by AFBCMR have adequate record support. With this discussion of the applicable review standard and the questions raised as background, the Court turns next to a separate consideration of each of the challenges Remy makes to the OERs.

A. 1975 OER

Remy claims that Kyler, as the reviewer, imposed ratings quotas on the rater and second rater in connection with the preparation of 1975 OER. Such quotas, if imposed, would violate the controlled OER system. The controlled OER System explicitly imposed quotas only on reviewers and control points.21 Moreover, the system forbade reviewers from forcing raters and second raters to meet a quota distribution. See AF 36-10, § 6-4 (“Reviewers may not return reports to additional raters and raters for the purpose of forcing them to meet a distribution.”). Undisputed in the record is that Kyler did impose quotas on raters and second raters. In response to a question propounded by AFBCMR about the imposition of quotas, Kyler answered “I restricted each function commander, squadron commander, and division chief not to exceed the percentages stipulated by AFR 36-10.”

Kyler’s imposition of quotas on raters and second raters, however, does not end the inquiry. It raises, instead, a second question. Were the quotas imposed by Kyler in 1975 actually followed by Remy’s rater, Hayes, and second rater, Wehling? Remy answers in the affirmative, presenting a statement from Wehling. In this statement, Wehling declares that Remy’s 1975 OER rating “was based on the quotas imposed by the OER system.” No contrary record evidence exists. Given this, Wehling’s 1975 rating must have been based, at least in significant part, on an *1271impermissible factor. There is, in sum, “clear and convincing evidence of factors adversely affecting the ratings which had no basis being in the rating process.” Guy v. United States, 221 Ct.Cl. 427, 433, 608 F.2d 867, 871 (1979). To ignore this uncontradicted evidence would be arbitrary and capricious. Under these circumstances, therefore, the 1975 rating prepared by Wehling must be deleted from Remy’s record.

Remy’s challenge to Hayes’ rating, however, stands on a different footing. Remy has not presented a statement from Hayes declaring that Hayes followed quotas imposed by Kyler. Moreover, defendants presented statistical evidence which demonstrates that raters did not, in fact, feel constrained by quotas. In 1975, for example, 28.3% of Remy’s peers received “Is.” 75.5% received either “Is” or “2s.” Remy’s challenge to the 1975 rating prepared by Hayes, therefore, must be rejected. Had similar evidence been presented concerning second raters and/or Wehling, Remy’s challenge to Wehling’s 1975 rating would also fail.

Remy seeks more than just deletion of the flawed 1975 Wehling OER rating. He asks the Court to replace the illegitimate rating with a “1,” the top rating. This request must be rejected. Wehling’s post hoc judgment, thirteen years after the fact, does not and can not support such an upgrade. Its post hoc nature deprives it of probative value. Wehling’s conclusions today concerning what rating Remy deserved thirteen years ago may be infected by irrelevant factors and events that have occurred since the date of the OER. Moreover, to credit Wehling’s post hoc judgment thirteen years later would open the flood gates and permit countless reviewers and raters to change their minds years later. Nothing could be more inimical to the fair rating system. It is vitally important that reviewers, raters and second raters perform their duties in the certain knowledge that the ratings they assign must reflect their honest and best judgment at the time and that they will not be permitted to issue revisions or express doubts years later. As the Court of Claims said in Tanaka v. United States, 210 Ct.Cl. 712, 713, 538 F.2d 348 (1976), cert. denied, 430 U.S. 955, 97 S.Ct. 1599, 51 L.Ed.2d 804 (1977), it is better “to attach more weight to an original OER than to a subsequent attempt by its writers to modify it as to matters of opinion only.” Accordingly, the Court rejects Remy’s request to upgrade the 1975 OER and also rejects Remy’s request to make Wehling’s and Kyler’s new remarks a part of the OERs. The Court, however, directs the Secretary to add a brief, nonprejudicial explanation to Remy’s record describing why the 1975 OER rating was removed. See Sanders v. United States, 219 Ct.Cl. 285, 314, 594 F.2d 804, 820 (1979) (ordering appropriate nonprejudicial explanation for gap in records due to removal of OERs).

B. 1976 OER

Remy challenges the 1976 ratings prepared by Kyler and Wehling on two grounds.22 First, Remy argues that both Kyler and Wehling impermissibly applied a quota in their 1976 ratings. Second, Remy argues that both Kyler and Wehling impermissibly considered eligibility for promotion and seniority concerns in their 1976 ratings. Remy’s challenge here is based on the 1976 revisions to the controlled OER system as announced in IMC 76-6. Defendants, however, argued in this Court that these revisions do not apply to Remy’s 1976 OER. This argument is without merit. IMC 76-6, dated October 26, 1976, ,is by its terms “effective immediately.” Kyler’s and Wehling’s 1976 ratings were completed and signed on October 31, 1976 — five days after the effective date of IMC 76-6. Thus, if Wehling and Kyler used quotas or considered seniority as a criterion in completing the 1976 OER, then that OER violates the regulations.

Statistical evidence persuasively demonstrates, however, that Kyler did not feel constrained by either quotas or proximity to promotion. Promotability clearly *1272was not the deciding factor in Kyler’s evaluations. In both 1975 and 1976, officers with less time in grade than Remy received “Is” from Kyler. In 1976, 14 of the 23 officers within Remy’s review group who received “Is” from Kyler were junior in rank to him. This is compelling evidence that Kyler did not use seniority as a factor in assigning ratings, notwithstanding his statement to the contrary. At the least, the AFBCRM was entitled to so conclude.

Similarly, the record contradicts Kyler’s assertion that quotas infected the 1976 OER. The record discloses that in 1976 Kyler awarded “Is” to only 14% of eligible officers. He could have awarded “Is” to 22% of these officers. Thus, there was nothing to prevent Kyler from awarding a qualified junior officer, like Remy, a “1” had Kyler concluded it was deserved. The AFBCMR’s conclusion that Kyler’s rating was not the result of the impermissible use of quotas is therefore not arbitrary or capricious. Indeed, Kyler actually downgraded the rating awarded by Wehling from a two to a three. Importantly, this downgrading was not required by the quota system. Kyler could award “Is” and “2s” to 50% of the officers he reviewed. In 1976, he awarded “Is” and “2s” to only 44.8% of these officers. The statistical evidence, therefore, proves that neither seniority nor quotas were factors in Kyler’s 1976 rating of Remy.

Statistical evidence, as it happens, does not exist for second raters, including Wehling. The AFBCMR claimed that it did not rely on statistical evidence in verifying the OERs. “The statistics, although of some value, were not the critical element. Of more importance were the OERs themselves and the statements provided by the individuals in the rating chain.” (Second Review at 3). There is no evidence, however, to contradict Wehling’s statement that he based his 1976 rating of Remy primarily on the quota system and seniority, impermissible considerations. “The process of evaluating officers by other officers is an inherently subjective process which neither the military boards nor this court will interfere with unless there is clear and convincing evidence of factors adversely affecting the ratings which had no business being in the rating process.” Guy v. United States, 221 Ct.Cl. at 433, 608 F.2d at 871; see also Skinner v. United States, 219 Ct.Cl. 322, 594 F.2d 824 (1979). There is clear and convincing evidence that Wehling considered such an impermissible factor in preparing Remy’s October 1976 OER. The 1976 OER prepared by Wehling therefore must be deleted and an appropriate, nonprejudicial explanation included in Remy’s record. For the reasons previously stated, however, the Court declines to upgrade the Wehling rating to a “1” or to make the qualitative statements of Kyler and Wehling a part of the 1976 OER.

C. Record Evidence

Finally, Remy argues that there is no substantial evidence in support of the challenged OERs. According to Remy, the AFBCMR conducted an imbalanced review of the evidence. A balanced review, according to Remy, demonstrates that the OERs are unsupported by evidence. Remy’s argument is meritless. There is substantial evidence in support of AFBCMR’s conclusion that the challenged OERs are valid, with the exception of Wehling’s 1975 and 1976 ratings. This evidence included Remy’s January 1969, July 1970, January 1975 and October 1977 OERs. These uncontested OERs demonstrate that the contested OERs were only a few of Remy’s OERs with less than superi- or ratings. More importantly, the AFBCMR studied the underlying descriptions within the challenged OERs and determined that these comments did not contradict the OER ratings:

[T]he comments are not particularly noteworthy, there is a definite lack of strong promotion recommendation, neither OER portrays an officer who would have been rated higher but for the constraints in existence at the time, and, most especially, both reports were downgraded by the reviewer.

(Second Review at 4).

The AFBCMR’s determination is based primarily on their interpretation of *1273the evidence contained in the challenged OERs. Wehling’s and Kyler’s recent statements about Remy’s abilities may not be considered as countervailing evidence. This is just the type of opinion evidence which was rejected by the Court of Claims in Tanaka v. United States, 210 Ct.Cl. 712, 713, 538 F.2d 348 (1976), cert. denied, 430 U.S. 955, 97 S.Ct. 1599, 51 L.Ed.2d 804 (1977). Tanaka also involved a challenge to OERs. There too the plaintiff submitted mitigating letters. The Tanaka court correctly excluded these post hoc changes.23 Wehling’s and Kyler’s new statements about Remy’s abilities are not competent to rebut their contemporaneous determinations as reflected in the challenged OERs. There is, in sum, a “strong, but rebuttable, presumption that administrators of the military, like other public officers, discharge their duties correctly, lawfully, and in good faith.” Sanders v. United States, 219 Ct.Cl. 285, 302, 594 F.2d 804, 813 (1979). Remy has not overcome this presumption in regards to the AFBCMR. The AFBCMR’s decision is plainly supported by substantial evidence.

Conclusion

Remy was apparently a competent officer. He was awarded the Distinguished Flying Cross and was nominated for Junior Officer of the Quarter.24 He received a Masters Degree in Executive Development in Public Service from Ball State University, and graduated, by correspondence, from Command and Staff College. In a perfect world, Remy may well have deserved promotion. On the other hand, Remy may have been merely competent in a group where exceptional or outstanding was the median. In any event, this is not a perfect world and it is not this Court’s role to substitute its judgment on promotion decisions for the Air Force’s. See Andrews v. Webb, 685 F.Supp. 579, 588 (E.D.Va. 1988). Rather, this Court’s role is limited; decisions of the Secretary and the AFBCMR are entitled to deference and must be upheld unless arbitrary and capricious, contrary to law or unsupported by substantial evidence. In the present case, with the exception of ratings by Wehling on the 1975 and 1976 OERs, there is no persuasive evidence of any procedural or regulatory violation to warrant reversal of the AFBCMR’s decision. Nor, with the exception of those specific ratings, is there any reason to conclude that defendants’ decisions were arbitrary or capricious or unsupported by substantial evidence. Thus, only Wehling’s ratings must be deleted from the challenged OERs. All other aspects of the OERs are correct and should be left intact and unrevised. It is unclear from the record whether the deletion of only one rating from Remy’s 1975 and 1976 OERs is sufficient to require the Air Force to resubmit Remy’s cause to a special promotion selection board. The parties did not address this precise issue. It was, perhaps understandably, not foreseen. The Secretary may have in place policies, procedures or regulations that govern or bear on whether deletion of a single OER rating warrants another bite at the promotion apple. Even absent any such guidance, the Secretary may prudently decide that resubmission is appropriate. In any event, this *1274issue is best resolved by the Secretary on remand. Should Remy disagree with the Secretary’s decision on resubmission, he may, of course, pursue whatever appeal remedies he may have.

For the reasons stated herein, therefore, the Court grants summary judgment to defendants on all claims except those pertaining to Wehling’s rating of Remy on the 1975 and 1976 OERs. With respect to those ratings, the Court grants summary judgment in favor of Remy and orders that those ratings be stricken from the 1975 and 1976 OERs. No other portion of those OERs should be stricken or revised. Nor should the statements of Kyler or Wehling be made a part of Remy’s promotion record.

Accordingly, this matter is remanded to the Secretary who is directed to delete Wehling’s rating from Remy’s 1975 and 1976 OERs, but to leave intact the remainder of both OERs, including all other ratings and all qualitative comments.

An appropriate order will issue.