The reasoning in support of the panel’s decision, fully set forth in the panel opinion, needs no elaboration. See Brown v. City of Oneonta, 221 F.3d 329 (2d Cir.2000). Some of the judges dissenting from denial of rehearing in banc, however, have chosen this occasion to advance, for the first time, novel equal protection theories that, in my view, would severely impact police protection. While such new theories are common to the pages of an academic journal to which interested critics might reply in the fullness of time, their appearance in this venue requires a more immediate response.
The dissenters propose that when the police have been given a description of a criminal perpetrator by the victim that includes the perpetrator’s race, their subsequent investigation to find that perpetrator may constitute a suspect racial classification under the equal protection clause. Judge Straub’s view is that equal protection review is triggered whenever the police rely on a physical description provided by a victim or witness that includes race as the basis for conducting an investigation. Judge Calabresi believes that equal protection review arises in a slightly narrowed, yet related situation: when the police ignore the non-racial components of the provided description and question persons who, except for the racial descriptor, do not fit the description provided.
The fact that no legal opinion, concurrence, dissent (or other judicial pronouncement) has ever intimated, much less proposed, any such rules of equal protection confirms a strong intuition of their non-viability. But, for the benefit of anyone who in the future may be undeterred by the inability of these theories to attract judicial recognition, their practical difficulties and analytical defects should be recognized.
I. General Concerns
For better or worse, it is a fact of life in our diverse culture that race is used on a daily basis as a shorthand for physical appearance. This is as true in police work as anywhere else. The theories suggested by the dissenters would require a police officer, before acting on a physical description that contains a racial element, to balance myriad competing considerations, one of which would be the risk of being subject to strict scrutiny in an equal protection lawsuit. Moreover, the officer frequently would have to engage in such balancing while under the pressure of a time-sensitive pursuit of a potentially dangerous criminal. Police work, as we know it, would be impaired and the safety of all citizens compromised. The most vulnerable and isolated would be harmed the most and, if police effectiveness is hobbled by special racial rules, residents of inner cities would be harmed most of all.
There have been times and places in this country in which the police have tolerated crime in African-American communities. See, e.g., John Dollard, Caste and Class in a Southern Town (1937). They have done this on a' variety of assumptions, all of them degrading, and one of them was that the victims in these neighborhoods were somehow less important to the then dominant white community from which the police drew their support. Although still imperfect, the more racially diverse police of today generally strive to serve and protect those within African-American communities as they do those within every other community.
I have little doubt that the rules of constitutional law proposed by my colleagues would weaken police protection within all communities. Regrettably, the social costs of frustrating police investigations receive no mention in either dissent. In my view, it is a grave mistake to seize upon an idea that would alter police work and law enforcement procedures fundamentally without fully considering its effect on those most vulnerable to crime.
*772In addition to potentially chilling police protection, and tying up officers in added court proceedings, these new rules would be implicated in many ordinary police investigations. As a result, these rules would likely undermine the strict scrutiny standard itself, because apprehending dangerous criminals in almost all instances would constitute a compelling state interest. Frequent satisfaction of strict scrutiny as police go about their daily work of investigating crime would likely have spillover effects into other areas of equal protection law, diluting the standard’s efficacy where we would want it to retain its power.
II. Judge Straub’s Proposal
The panel determined in this case that the police interviews of African-Americans in Oneonta — conducted in the hope of finding a person fitting the victim’s description of the perpetrator, a young, African-American male with a cut on his hand — did not trigger equal protection scrutiny because the officers, by acting on the basis of a description provided by the victim, proceeded in a race neutral manner, and limited their search for a suspect to persons fitting the victim’s description.
The rule that Judge Straub proposes is far broader and more trouble-prone than any possible emanations from the fact-specific holding of the panel opinion. Judge Straub suggests that whenever the police use a racial descriptor, they are employing a suspect racial classification, and should therefore be subject to strict judicial scrutiny. Indeed, under his theory, the police would face litigation even where the racial descriptor is combined with other common descriptors such as age, gender, and a physical marking (as was the case here), even if the police adhere faithfully to that description, and “regardless of [its] source.” See post at 790 (Straub, J. dissenting) (emphasis in original).
This rule would impede the most routine police work, particularly criminal investigations. A description of a perpetrator received by the police will often include only general attributes such as the perpetrator’s race, height and sex. If strict scrutiny were triggered by the mere presence of a racial descriptor, an officer would be subject to a lawsuit simply for trying to supplement a sparse description — and thereby narrow the field of potential suspects — by interviewing other persons who fit the racial description but are not yet suspects. Indeed, the police would be employing a suspect racial classification (and thus would be required to show a compelling state interest) whenever they “use” a racial descriptor, whether it is in an internal report, an equal opportunity employment database, a criminal investigation, or in simply recording or relaying a witness’ description.
The right to equal protection is an individual one. See Village of Willowbrook v. Olech, 528 U.S. 562, 120 S.Ct. 1073, 1074-75, 145 L.Ed.2d 1060 (2000). Under Judge Straub’s theory, therefore, the right not to be questioned (absent a compelling state interest and by means narrowly tailored to the pursuit of such interest), would be offended whenever the police act upon a description that includes race, regardless of whether the person questioned was the two hundredth approached by the police or the only one. Nothing in Judge Straub’s opinion suggests that his rule would be limited to exclude a situation in which an officer observes a crime on the street, follows the perpetrator into an empty restaurant, and questions the only customer there who fits the race of the perpetrator he observed. In short, it is difficult to discern limits to the impediments Judge Straub’s rule would place on ordinary police work.
III. Judge Calabresi’s Proposal
Judge Calabresi, although discomforted by the panel’s decision that equal protection considerations are inapplicable when the police follow a victim’s description that includes race, acknowledges certain difficulties with a contrary position. He cor*773rectly recognizes that under present law “[i]f an action is deemed a racial classification, it is very difficult, under the Supreme Court precedent, ever to justify it” and making such justification easier “in cases of police following a victim’s description” would lead to an undesirable spillover in other racial classification contexts. See post at 786 (Calabresi, J. dissenting). “In other words, were the requirements of strict scrutiny to be relaxed in the police/victim’s description area, it would be hard indeed to keep them from also being weakened in other areas in which racial classifications ought virtually never to be countenanced.” Id. He concludes, therefore, that “courts should recognize severe limitations on them competence to deal with victim’s racial descriptions.” Id.
Unable to discern a plausible jurisprudential basis for an equal protection claim when the police follow a victim’s description, Judge Calabresi proposes a variant rule that he says this case raises: strict scrutiny is triggered if the police disregard all but the racial component of the victim’s description. More precisely, he states the rule as “the police created and acted upon a racial classification [that triggers strict scrutiny] by setting aside all but the racial elements in the victim’s description.” Id. at 781 (emphasis in original).
A. In Banc Considerations
As noted above, Judge Calabresi’s rule has neither been proposed, let alone adopted, by any court. So far as I can tell (or Judge Calabresi claims), it has never before been thought of. But more to the point in this litigation, it was at no time argued by the plaintiffs and in fact is not supported by the pleadings. As a matter of in banc policy, substantive law, and sound jurisprudence, this court has appropriately declined to reach out and embrace this new untested rule on facts that do not put it at issue.
It is well-settled in this court that “[a] conclusory allegation without evidentiary support or allegations of particularized incidents, does not state a valid claim.” Kern v. City of Rochester, 93 F.3d 38, 44 (2d Cir.1996) (quoting Butler v. Castro, 896 F.2d 698, 700 (2d Cir.1990)). Judge Cala-bresi relies chiefly on a general allegation in the complaint that the police “attempted to stop ... any and every non-white person in and around the City of Oneonta.” Second Amended Complaint at 4; see also id. at 30 (“the objective, as defendant Chandler [the State police investigator in charge of the investigation] told the newspapers Newsday and the Oneonta Daily Star, was to try ‘to examine the hands of all black people in the community.’ ”). Id. at 2. But these are just the sort of general, bald allegations prohibited by Kern and Butler. And the complaint has a strange way of alleging, as Judge Calabresi reads it, that the police stopped African-Americans without regard to their age and sex. Indeed, the allegation is hedged as an “attempt” and seemingly includes other “non-whites” (for example, Asians, Native Americans, Hispanics). Moreover, it is vague on precisely the subject as to which specificity from the plaintiffs would be expected: departure from the given description. In sum, no where does the complaint allege that the police actually did depart from the description given by the victim. That such a contention is absent is not surprising: it strikes me as nonsensical to believe that the police, who have been given a description of the attacker, would disregard the description and look for someone else.
Perhaps realizing the deficiency in the allegation he points to, Judge Calabresi scours the approximately one-hundred-page complaint and the record below to patch together scattered demographic references in an effort to tease out an inference that the police stopped numerous black women. His math is speculative and ends up presenting at most another insufficiently particularized allegation. See post at 782 (Calabresi, J. dissenting) Only one relevant allegation with the required specificity appears in the complaint: that the *774police stopped Sheryl Champen, a woman. To be sure, this one stop could have been in disregard of the victim’s description of the attacker as male. But it is far more likely that the police, who after stopping her did not ask to see her hands, were initially mistaken about her sex or, because she was boarding a bus, feared losing a person with relevant information. In any event, the solitary fact that Ms. Cham-pen was questioned cannot support the far broader claim that the police systematically strayed from the bounds of the description they were given and stopped “any and every” African-American in Oneonta.1
Judge Calabresi’s suggestion that we instruct the district court in this case to allow the plaintiffs to submit a fourth complaint which could then allege facts that could support his proposed rule introduces a jurisprudential danger. A footnote in the panel opinion permits repleading, see Brown v. City of Oneonta, 221 F.3d 329, 339 n. 9 (2d Cir.2000), but it does so only to the extent that the district court’s understanding of the law of this circuit is clarified by the panel opinion. Had this court actually adopted Judge Calabresi’s new rule and directed the district court to permit repleading, and had the plaintiffs still been unable to state a case, which in my view is likely, this court would have announced a novel rule of constitutional law on a fact scenario that Judge Calabresi has simply hypothesized from a creative (and strained) reading of the complaint. As a rule of constitutional law, it would be impervious to legislative change, drifting about, untethered by any factual anchor, turning up in odd contexts, uninvited and inapt. The case and controversy requirement of Article III, both as the supreme law of the land and as wise jurisprudence, requires more than a hypothetical scenario to introduce far reaching constitutional strictures into the law.
B. Criticisms
Putting aside the fact that Judge Cala-bresi’s creative proposed rule is not presented in this case, in my view it is flawed and unworkable.
Judge Calabresi’s opinion describes the proposed rule variously as: (i) “the police create[dj and act[ed] upon a racial classification [that triggers strict scrutiny] by setting aside all but the racial elements in the victim’s description” and (ii) “[the state is] creating an express classification that can only be approved if it survives strict scrutiny when state officers (like the police) ignore essentially everything but the racial part of a victim’s description, and, acting solely on that racial element, stop and question all members of that race they can get hold of, even those who grossly fail to fit the victim’s description^]” See post at 781 (Calabresi, J. dissenting) (emphasis in original).
If this proposed rule were adopted, it would mean that whenever the police are given a description that includes race, simply asking questions of a person of that same race who does not otherwise fit the description would violate that person’s constitutional right to equal protection unless the state could (1) show a compelling state interest and (2) that the questioning was essential to achieving that interest.
Innocent situations that could trigger liability under Judge Calabresi’s rule spring to mind. For instance, the proposed rule would apply to any situation in which the police were trying to find a fleeing suspect in a defined and limited area, such as a restaurant, a sidewalk, a parking lot, or a building, regardless of how many people occupied the area in question. In such a situation, officers often cannot know with complete certainty whether the person they question eventually might turn out to be a suspect, not *775least because they can never be sure of the accuracy of the victim’s description, or whether the person so described has somehow subsequently altered his or her appearance, perhaps by shedding tell-tale clothing.
Judge Calabresi’s rule also would apply to instances where a police officer forgets or confuses part of the description— whether the perpetrator was wearing a grey jacket.or a brown one, for example. In such instances, prudent officers would fear to question anyone at all lest they draw an equal protection lawsuit. Finally, and perhaps most troubling, Judge Cala-bresi’s proposal, were it adopted, might be used as a prophylactic device to invalidate the arrest of the actual perpetrator, if that person could successfully argue that when he was first stopped and questioned he imperfectly “fit” a victim description that included race.
C. Fourth Amendment Safeguards
The Fourth Amendment’s prohibition on unreasonable searches and seizures, carefully calibrated by the Supreme Court over two centuries, balances law enforcement needs against the rights of the citizen to be protected. See Graham v. Connor, 490 U.S. 386, 395, 109 S.Ct. 1865, 104 L.Ed.2d 443 (1989). Three levels of interaction between the police and private citizens have developed under Fourth Amendment jurisprudence: voluntary encounters that do not require justification, so long as the police do not intimate that their requests must be heeded, see United States v. Tehrani 49 F.3d 54, 58 (2d Cir.1995); investigative detentions, so called Terry stops, that do require a justification of reasonable suspicion of criminal activity, see Terry v. Ohio, 392 U.S. 1, 88 S.Ct. 1868, 20 L.Ed.2d 889 (1968); and finally, arrests that require a demonstration of probable cause, see Tehrani 49 F.3d at 58. This framework, arrived at over the years through case-by-case adjudication, in the context of concrete factual settings, strikes an appropriate balance between individual rights and the necessities of effective law enforcement. See United States v. Place, 462 U.S. 696, 703, 103 S.Ct. 2637, 77 L.Ed.2d 110 (1983).
Judge Calabresi’s proposal, by injecting equal protection analysis into police investigations that rely on racial descriptors, would upset this carefully crafted balance.2 *776Fearing personal liability through Section 1983, 42 U.S.C. § 1988, litigation from equal protection violations arising from their investigative activities, police officers would undoubtedly fail to act in situations where we would expect them to.
The indefensibility of accepting the social costs of chilled policing in this context becomes all the more apparent when one considers the present reach of the Fourth Amendment: the gap in constitutional protection that Judge Calabresi believes to be created by the panel opinion and that he intends his rule to remedy is a narrow one. Judge Calabresi’s proposed approach would impose equal protection analysis without regard to whether the person encountered by an officer was arrested, temporarily detained for questioning or simply asked questions while remaining free to walk away.
Fourth Amendment jurisprudence— though not a general bar to racial discrimination or classification — already prohibits arrests and Terry investigatory detentions in many situations with which Judge Cala-bresi is concerned, that is, where “state officers (like the police) ignore essentially everything but the racial part of a victim description, and acting solely on that racial element stop and question all members of that race they can get hold of even when those questioned grossly fail to fit the victim’s description.” See post at 564 (Cal-abresi, J. dissenting). Such stops based on racial considerations alone, absent compelling exigent circumstances, would almost never rest on the constitutionally required “reasonable articulable suspicion” of criminal activity needed to justify an investigatory detention, see, e.g., U.S. v. Brignoni-Ponce, 422 U.S. 873, 885-86, 95 S.Ct. 2574, 45 L.Ed.2d 607 (1975) (no reasonable suspicion for border agent to detain person based solely on apparent Mexican ancestry)3, and a fortiori would never rise to the level of probable cause for a warrantless arrest. Cf. Graham v. Connor, 490 U.S. 386, 395, 109 S.Ct. 1865, 104 L.Ed.2d 443 (1989) (applying Fourth Amendment analysis and not more generalized due process guarantee to arrests, investigatory detentions, and “other ‘seizure[s]’ of a free citizen ... [bjecause the Fourth Amendment provides an explicit textual source of constitutional protection” as opposed to the “more generalized notion of ‘substantive due process’ ”). Therefore, it is only for police encounters falling short of a restraint “by means of physical force or show of authority,” Terry, 392 U.S. at 19 n. 16, 88 S.Ct. 1868, that Judge Cala-bresi’s proposed introduction of equal protection analysis would supply constitutional protections presently unavailable under the Fourth Amendment.
I believe that any benefits from extending equal protection guarantees to such situations, where the citizen who is questioned is not deprived of his liberty even for a brief period of time and remains free at all times to walk away from the officer, are outweighed by the additional costs to *777effective law enforcement.4 Officers rely on their ability to act on non-articulable hunches, collected experience, intuition, and sense impressions — all of which are crucial in carrying out a criminal investigation. Officers would be forced to justify these intuitive considerations in order to meet an accusation that race was the sole factor motivating the encounter. The unworkability of such a regime is self-evident.
If some guidelines for officers conducting non-restraint encounters based on victim or witness descriptions, where race is a component of the description, are needed (about which I have no opinion), then they are more appropriately established by bodies with political accountability and with more experience and insight into the nuances of community policing than unelected life-tenured federal judges. State legislatures, municipal councils, and citizen oversight boards are far better suited than courts to balance the complicated considerations regarding the community’s policing needs and the sensibilities of minorities who may feel unfairly targeted.
Judge Calabresi is understandably troubled by the facts of this case, as no doubt were the plaintiffs to an even greater degree. But it seems to me that his proposed new rule has missed the real source of that discomfort. It is not that the police strayed from the description they were given, if stray they did. Rather, it is that the police conducted an investigation that went through the town seeking to stop every young African-American male fitting the description the victim provided. The proper judicial remedy for people who were stopped and questioned during that investigation, as I have noted, is under the articulated and available standards of the Fourth Amendment; and several of those stopped have pursued that remedy. The constitutional rule that Judge has fashioned, by contrast, would hamper police efforts in investigating countless individuals — investigations that bear little or no resemblance to the “sweep” that is alleged to have occurred in this case. The consequent crippling of law enforcement in the more individualized contexts of daily police work would exact severe societal costs, and nowhere more so than in minority neighborhoods. The obvious presence of such costs counsels strongly against deciding to go in bane in order to elaborate ex nihilo a broad new rule of equal protection law that would constrain the police in situations that do not implicate the Fourth Amendment.
For these reasons, I concur in the court’s decision to deny this petition for rehearing in banc.
. Judge Calabresi also thinks the police request for a list of black male students at SUCO creates an inference that they departed from the victim's description. But this fact does not support Judge Calabresi's claim that the police wanted to stop elderly matriculants as well as those supposed to be college-aged.
. Judge Calabresi also argues that our circuit's Fourth Amendment law concerning whether a stop has occurred needs to be clarified. Judge Calabresi maintains that it is unsettled whether the reasonableness determination-that is, deciding whether a reasonable person would have felt free to leave-is a question of fact or one of law. See post at 780-81 n. 1 (Calabresi, J. dissenting). I think the law is clear on this point. The question of whether a seizure occurred is a question of law, see Oneonta, 221 F.3d at 340; the circumstances underlying that determination are questions of fact for the jury. See id. Whether a seizure occurred is determined by asking whether a reasonable person would have felt free to leave. Because the seizure determination is a question of law, it follows a fortiori that ascertaining reasonableness is also a question of law. See United States v. Peterson, 100 F.3d 7, 11 (2d Cir.1996) (reviewing de novo the district court's determination, before trial, that a reasonable person would have felt free to leave); United States v. Montilla, 928 F.2d 583, 588 (2d Cir.1991). (“[W]e believe that ... whether those statements and acts resulted in a seizure is a question of law subject to de novo review.”); see also United States v. Espinosa-Guerra, 805 F.2d 1502, 1507 n. 18 (11th Cir.1986) ("[tjhe trial court's determination of whether a reasonable person .would have believed that he is not free to leave is a question of law” (internal quotation marks omitted)).
The single case from our circuit that Judge Calabresi says supports the view that a reasonableness determination under the Fourth Amendment is a question of fact is inapposite. See Posr v. Doherty, 944 F.2d 91 (2d Cir.1991). The question in Posr, in which I wrote the opinion, was not whether police action constituted a seizure, but whether an arrest had occurred. And, as Judge Calabresi notes, the question of when an investigative stop ripens into an arrest is for the finder of fact. See Tehrani, 49 F.3d at 58. Because in Posr the jury had already found that the police had used excessive force, we noted that as a practical matter the only determination left for the jury on remand was reasonableness. See Posr, 944 F.2d at 99-100. The fact that reasonableness was an issue for the jury in that *776ease followed from its peculiar posture and from the arrest issue. The law is otherwise clear that whether a seizure has occurred-that is, whether a reasonable person would have felt free to leave-is a question of law for the court.
. Lower courts are generally in accord that race alone will not support a Terry stop under the Fourth Amendment. See, e.g., U.S. v. Grant, 920 F.2d 376, 388 (6th Cir.1990) (no reasonable suspicion to detain person when agents' only basis for stop was that man of color wearing dreadlocks was illegal alien from Jamaica and traveled from drug-source city); Buffkins v. City of Omaha, 922 F.2d 465, 470 (8th Cir.1990) (no reasonable suspicion when informant’s tip merely described race of person and person carried toy animal that appeared to be resewn); Orhorhaghe v. INS, 38 F.3d 488, 497 (9th Cir.1994) (no reasonable suspicion for INS agent to seize person based solely on “Nigerian-sounding name”); U.S. v. Tapia, 912 F.2d 1367, 1371 (11th Cir.1990) ("no reasonable suspicion supported further detention of vehicle beyond citation for speeding when suspect Mexican had out-of-state license plates, appeared visibly nervous during confrontation with officers, and had few pieces of luggage”).
. I do not believe Judge Calabresi's proposed rule to be "superfluous,” see post at 787-88 n. 13 (Calabresi, J. dissenting), because of the existence of Fourth Amendment protections. To the contrary, my point is that the potential constitutional protections his unprecedented rule would afford to potential criminal suspects (who are in no way restrained but may be offended by the police encounters) would add to those that exist but would upset the Fourth Amendment’s careful balance of interests and entail unacceptable costs to society.