People follow a bell curve, so it's best if your sizing can reflect that. If you sell (for example) 3 XS, 10 S, 40 M, 35 L, it's going to be a lot harder to sell the extra XS (i.e. it's easier to sell 40 XS when you have other sizes in stock, than it is to sell 40 XS when XS is the only size you have).curiousgeorge said:
Depending on your market I would consider stocking XLs.
All of that said... if you're happy doing 40 per size for your first run, you'll get an extremely good idea of how fast each size sells and therefore how to fine tune it for the next run. It could cost you some money that way, but experience is definitely the best way to know what size breakdown is going to work for you and your customers. If you expect to reprint the same design (i.e. 160 is a first run, but there will be followup runs) then that's even more true - if a size doesn't sell very fast you can just keep what you already have on hand and replace the sizes that do.
I know "it depends" isn't a very helpful answer, but it is fairly accurate