I’ve spoken at length about this topic with two audio pioneers who are some of the most knowledgable people on the planet when it comes to vinyl reproduction.
George Merrill (whom I assume needs no introduction here) strongly prefers 12-inch arms for the most obvious reason: geometry. If you focus on optimizing cartridge alignment, a longer arm is clearly superior when it comes to things like optimizing the placement of null points.
But JR Boisclair of WAM Engineering (that’s "Wallytools") disagrees that this benefit outweighs the real-world performance benefits of a shorter arm. For those not familiar with WAM, JR is known for his years of extensive testing and analysis of arm & cartridge characteristics. Sure, resonances & whatnot are factors to consider, but qualitative factors don’t automatically translate into quantitative differences in performance. If a cartridge is properly aligned -- something that JR insists must be done by means of microscopic analysis -- arm bearings become a crucial factor. JR has found that too-loose or too-tight bearings may be the most important cause when an otherwise good arm exhibits a lot of mistracking. (He even sells a WallyTool that checks bearing play.) And he’s found empirically that a 9-inch arm, all things equal, will almost always be more stable than an equivalent 12-inch model.
Me, I've always been in awe of George's expertise and still love the bespoke table he built for me many years ago. But I gotta say that JR is a brilliant engineer, is not given to hyperbole, and has a fanatical commitment to basing his conclusions on high-precision empirical measurements. So when I upgrade to a Hana Umami next month (after my unit has been analyzed by WAM, of course), I’ll probably be picking up one of George’s 9-inch arms to go with it.
