One reason could be that there have been hardly any 17th century mis-spellings in even half-decent condition (ie as good as they come) on the market in the last 15 years. I can't speak so much for farthings, or transactions via less high-profile dealers but to the best of my knowledge here are the last appearances for the best halfpenny ones:
1699 TERTVS - 2006 (Gregory I) Peck Plate Coin. Best known
1696 TERTVS - 2012 (Lockdales, ex Nicholson) Peck Plate Coin. Best known
1699 GVILELMVS - 2004 (Nicholson). Best known
1700 GVLIEEMVS - ~2007/8 (Baldwins). VF -GVF, better condition than the other mispellings listed here.
1699 GVLIEMVS - 2004 (Nicholson). Best known
The 1700 GVLIELMS variety generally turns up in similar condition but only ~ Fine or less - I've got one about 15 years ago not too inferior to the Peck Plate Coin for about £70.
1672 CRAOLVS - 2007 (DNW, not admitted as, but is the Peck Plate coin). Best known
1673 CRAOLVS - 2004 (Nicholson - perhaps not quite as good detail as the SNC 1994 one, but best sold since).
Peck's 1699 GVLIEMVS Plate coin sat around for years in the Baldwins trays after it didn't sell at one of their Gregory auctions - it was very ropey though!
Though I know much less about farthings, the 1697 GVLIELMS in lustrous, virtually mint state (DNW, 2010) though not struck up at the bottom of the bust or Brit's head due to an uneven flan) must be the finest known of all the 17th century copper mis-spellings, by a mile.