dissenting:
Samuel Salomaa appeals the district court’s decision affirming the administrative denial of his ERISA plan disability benefits for Chronic Fatigue Syndrome. The majority holds that the plan abused its discretion; I disagree. Therefore, I respectfully dissent.
I.
I readily agree with the majority that the standard of review applied to the plan’s denial is abuse of discretion with some degree of skepticism. Opinion at 675-76. The abuse of discretion standard is typically a deferential one, see, e.g., Conkright v. Frommert,- U.S. -, 130 S.Ct. 1640, 1646, 176 L.Ed.2d 469 (2010), but this deference is often muted when an ERISA plan administrator both administers and funds its plan. See Saffon v. Wells Fargo & Co. Long Term Disability Plan, 522 F.3d 863, 868 (9th Cir.2008). Although this standard’s dualism between skepticism and deference may seem strange, it is the proper standard and must be applied carefully. Suffice to say, as the majority aptly notes, we evaluate the plan’s denial for reasonableness. Opinion at 675-76.
But it is in this evaluation where I depart from my colleagues. The majority highlights several points in support of its holding that the plan abused its discretion. I address each in turn.
II.
A.
The majority first draws attention to the fact that the doctors who personally examined Salomaa found that he was disabled. Opinion at 676-77. Those doctors who found otherwise, the majority explains, did not personally meet with Salomaa. Id. The majority then derives importance from this fact, suggesting (without outright stating) that doctors who personally examine claimants are somehow more reliable than doctors who do not personally examine claimants. See id. It then implies that the plan’s decision not to personally examine Salomaa evinces an abuse of discretion. See id. It is unclear from the opinion why the majority adopts these views, as it provides no clear reason why a doctor who personally examined Salomaa should be given more authority or attention than one who didn’t, and no clear reason why a lack of personal examination precipitates an abuse of discretion. See id. Perhaps no reasons were given because no reasons exist.
Consider Black & Decker Disability Plan v. Nord, 538 U.S. 822, 123 S.Ct. 1965, 155 L.Ed.2d 1034 (2003), an ERISA case in which the Supreme Court vacated a judgment of this court. In holding that the “treating physician rule” does not apply to ERISA plans, a unanimous Court stated, “Nothing in [ERISA] suggests that plan administrators must accord special deference to the opinions of treating physicians. Nor does [ERISA] impose a heightened burden of explanation on administrators when they reject a treating physician’s opinion.” Id. at 831, 123 S.Ct. 1965. The Nord opinion seems sufficient to dispel any notion of treating physician superiority, so it is puzzling why the ma*682jority nevertheless chastises the plan for not personally examining Salomaa.
In fact, I’ve found no published Ninth Circuit case stating that “personal examination” dictates whether an ERISA plan administrator abused its discretion. The Nord Court indicates why: “[T]he assumption that the opinions of a treating physician warrant greater credit than the opinions of plan consultants may make scant sense when, for example, the relationship between the claimant and the treating physician has been of short duration, or when a specialist engaged by the plan has expertise the treating physician lacks.” Id. at 832, 123 S.Ct. 1965.
Given Nord, the majority’s emphasis on personally examining physicians is misplaced. I do not think the plan’s lack of personally examining physicians indicates that it abused its discretion in denying Salomaa’s claim.
B.
Next, the majority finds an abuse of discretion in that “the plan administrator demanded objective tests to establish the existence of a condition for which there are no objective tests.” Opinion at 676. Though I agree with the majority that a plan administrator ought not be able to condition an award on the existence of evidence that is impossible to produce, I disagree with its conclusion that there was no objective evidence of Salomaa’s disability. See Opinion at 678.
Salomaa did, ultimately, provide the plan with objective evidence of his disability in the form of neuropsychological testing, but because this evidence was acquired after the elimination period, the plan administrator found it inadequate to prove Salomaa’s condition. This neuropsychological testing showed deficiencies in Salomaa’s mental activity — a known symptom of CFS. And even though this evidence was gathered at the wrong time, its very existence demonstrates that objective tests can be used to prove disability from CFS, and that Salomaa could have proven his disability with objective evidence.
But don’t take my word for it. Just ask Salomaa himself. In Salomaa’s opening brief, he admits that the belated neuropsychological testing “objectively proved the existence of’ his disability caused by CFS.1 Appellant’s Br. at 31. So it appears that the majority has rejected the factual understandings of both parties by deciding that no objective evidence of Salomaa’s condition could have been produced.
The majority gives only superficial attention to the neuropsychological testing that everyone but it views as objective evidence of Salomaa’s disability. It uses this testing to colorfully describe how debilitating Salomaa’s condition is, noting that the “battery of intelligence and other tests” revealed that Salomaa’s IQ was average, “which is fine for many people but shockingly low for a Harvard man with a career in computers.” Opinion at 672. Now I’m no “Harvard man,” and I can barely operate an iPhone let alone pursue a career in computers, but I don’t understand why the majority at once boasts how meaningful the neuropsychological testing is — having revealed in Salomaa “ ‘the hallmark cognitive symptoms’ ” of CFS — yet later dismisses the testing as merely “bad performance on an intelligence test” that could “be faked.” See Opinion at 671-72, 678.
*683I also find it peculiar that the majority speculates that Salomaa could be faking his neuropsychological testing results. This seems to be the only time that the majority acknowledges that Salomaa might be faking his disability. The majority wields no similar skepticism when considering the other aspects of Salomaa’s claim, such as the mountains of self-reported symptoms that prompted the plan’s desire for objective evidence in the first place. Furthermore, the assertion that Salomaa may have faked this testing seems to conflict with the majority’s own description of the testing, “which showed ‘a valid profile and that [Salomaa] was putting forth adequate effort.’ ” Opinion at 672.
Finally, I find the majority’s citation to the plan administrator’s parent company’s position paper to be unpersuasive. This paper states that “[t]here are no specific diagnostic studies ... or physical findings that are specific to the diagnosis of CFS.” ER 158-59. This is unhelpful, however, because this position paper says nothing about the existence — or lack thereof — of objective evidence of CFS-caused disability. See id. Indeed, the majority’s quoted material only discusses “specific” studies and findings rather than “objective” studies and findings, and it pertains to establishing a diagnosis rather than disability. The majority therefore conflates the notions of specific and objective (and diagnosis and disability) while ignoring the admissions in Salomaa’s own brief.
For these reasons, the plan’s demand for objective evidence of Salomaa’s disability does not indicate that the plan abused its discretion.
c.
The majority surmises that “the administrator failed to consider the Social Security disability award.” Opinion at 676. I see nothing in the record to support this finding.
The plan administrator’s failure to mention the Social Security award in its correspondence with Salomaa does not mean the plan administrator ignored the award altogether. In fact, there is no proof that the administrator failed to consider the award, just as there is no proof that it closely and deliberately studied the award. In short, the record is silent as to what the administrator did with the award. Instead of treating this fact as inconclusive, the majority assumes that it demonstrates an abuse of discretion.2
D.
The majority faults the plan administrator for failing to meaningfully communicate with Salomaa. Opinion at 676. While it is a close question, I agree with the majority that the plan administrator failed to engage Salomaa in a meaningful dialogue. However, this failure should lessen the discretion this panel accords to the plan and should not itself substantiate a conclusion that the plan abused its discretion.
The majority compares the administrator’s conduct to the violations addressed in Booton v. Lockheed Medical Benefit Plan, 110 F.3d 1461 (9th Cir.1997) and Sajfon v. Wells Fargo & Co. Long Term Disability Plan, 522 F.3d 863 (9th Cir.2008). However, the failures in communication in Booton *684and Saffon are far worse than those alleged in the present case. In Booton, a plan administrator denied a claim “without explanation” and without requesting “necessary — and easily obtainable — information.” See Booton, 110 F.3d at 1463-64. In Saffon, a plan administrator’s communications included late test requests, self-contradictory statements, failures to tell the insured about discussions with her doctor, and omissions of important language from doctors’ reports. See Saffon, 522 F.3d at 871, 873 & n. 5.
The plan administrator’s communication in Salomaa’s case, though faulty, was much less severe. For instance, it provided Salomaa with a series of denial letters that described the medical records at issue, it specifically addressed most of Salomaa’s evidence, it contrasted Salomaa’s reported activities with his alleged symptoms, and it outlined other information that Salomaa could provide. Because the administrator’s communicative failures are not nearly as egregious as the violations in Booton and Saffon, I would not analogize those cases to this case, and I do not view the administrator’s failure to meaningfully communicate as evidence of an abuse of discretion.
Any remaining problems with the plan administrator’s approach to Salomaa’s claim are minor and, in my view, they do not themselves demonstrate an abuse of discretion.
III.
The standard of review is important in this case, and it requires this panel to weigh carefully the plan administrator’s conflict of interest when reviewing the plan’s denial for abuse of discretion. In my view, the points highlighted by the majority do not demonstrate that the plan abused its discretion, even in light of the conflict of interest weighing against the plan.
The majority cleverly observes that “weighing” is only a metaphor, and that judicial “weighing” is much more difficult than the “weighing” one might do in a traditional darkroom with little brass weights and literal scales. Opinion at 674-75. I agree. Generally speaking, the “weighing” metaphor is misleading as it suggests that judicial balancing is an extremely precise — even mathematical — exercise. It is not. However, today’s decision demonstrates that the metaphor of darkroom weighing is not totally amiss. To arrive at today’s decision, the majority had to overlook binding precedent and turn a blind eye to inconvenient facts— almost as though it were looking at nothing at all, in a room of near total darkness.
I respectfully dissent.3
. In full, Salomaa's statement reads: “Why would CIGNA, if it was acting in good faith, ignore the testing which objectively proved the existence of symptoms that Mr. Salomaa had complained about for months?” Appellant’s Br. at 31. The answer, as noted, is because this testing was performed after the elimination period.
. This assumption is worsened by the facts that the plan administrator was not bound by the Social Security award whatsoever, see Montour v. Hartford Life & Accident Ins. Co., 588 F.3d 623, 635 (9th Cir.2009), and the standard used to determine a Social Security award differs considerably from an ERISA benefits standard. See, e.g., Nord, 538 U.S. at 832-33, 123 S.Ct. 1965 (noting differences between the two standards, including the fact that the former allows for more reliance on treating physicians’ opinions, and the latter is especially sensitive to the terms and design of each particular plan).
. Even if I agreed with the majority that the plan administrator abused its discretion, I would not reverse and remand with instructions to direct an award of benefits. To qualify for benefits, Salomaa has to prove (1) that he suffers from a sickness or injury, and (2) that his sickness or injury renders him "disabled.” See ER 125, 150. Both of these elements — diagnosis and disability — must be established. The district court affirmed the plan's denial of benefits based on Salomaa's failure to prove disability, and the court set aside the diagnosis issue, as it was "not well-suited for judicial determination.” Salomaa v. Honda Long Term Disability Plan, 542 F.Supp.2d 1068, 1076-77 (C.D.Cal.2008). Therefore, the diagnosis issue remains unresolved. Were I to find an abuse of discretion, I would remand with instructions to conduct further proceedings regarding the diagnosis issue.