This presentation combines Melody-to-Structure Licensing Constraints (MSLCs) with the hypothesis of asymmetric headedness. The goal is to improve the formalisation of phonotactics and especially those related to headedness. Szigetvári (2017) effectively proposes that the phonological grammar of English contains a restriction on the branching of the element |H|. This limits obstruent clusters by forcing them to contain at most a single fortis consonant. However, clusters such as /sf, ks, ps/ ([sfɪə] 'sphere', [fɒks] 'fox', [aeps] 'apse') and /ft/ (before stressed syllables) seem to disprove the claim: [fɪftʰíːn] 'fifteen'. Although the /t/ in these /ft/ clusters is phonetically a little reduced in its aspiration, it is by no means categorically/fully lenis: *[fɪftiːn] "fifdeen", and neither does the /f/ become lenis: *[fɪˑvtʰiːn] "fivteen". Continuing in the same spirit as Szigetvári (2017), but with a different hypothesis, we propose that aspirates and strident fricatives form a natural class defined by headed |H| (Kaye 2000, Backley 2011). We propose an MSLC that bans headed |H| from bipositional structures, effectively stopping headed |H| from branching. MSLC on |H| (English) |H| cannot be contained by a bipositional structure: Aspirates and strident fricatives (s, z, ʃ, ʒ) contain headed |H| Non-aspirates and non-strident fricatives (f, v, θ, ð) contain unheaded |H|. MSLC immediately excludes strident fricatives from preceding or following aspirate stops (*stʰ, *pʰs, *tʰs…). It also eliminates the very well-known restriction on s + aspirate: [pʰlɛis] 'place' vs. /mis-+ pʰlɛis/ → [misplɛis] ([*mispʰlɛis]) 'misplace'. And it excludes the far less well-known English ban on adjacent strident consonants: [stiːm] 'steam' vs. *[stʃiːm]. The exceptions to adjacent stridents are either the class of recent derived /t + u/ sequences: [stjʉː] or [stʃʉː] or [ʃtʃʉː] 'stew', or consciously known to be loanwords: [masʒid] 'mosque' (something which we suspect could be experimentally confirmed). Crucially the MSLC restriction in (1) does not exclude non-strident fricative + aspirate sequences: fi[ftʰ]een 'fifteen', [sfɪə] 'sphere', [fɒks] 'fox', [aeps] 'apse'. In order to work, our account relies on the phonological distinction between the natural class of strident vs. non-strident fricatives. The MSLC is a grammatical condition on representations. However, we notice that the typical Element Theory representation of headedness is highly arbitrary. In place of this, asymmetric headedness is proposed. A headed phonological expression also necessarily includes the unheaded version of headed element.