Relationship between word frequency and word length

[Koen G]

The problem is that we don't know the length of the word. [Daiin] doesn't exist, it is just a way we (using the EVA system) talk about these glyphs. But arguments could be made to see [iin] or even [aiin] as a single glyph, which would make it a short word. 

Not that this would fix Voynichese word length statistics overall. But we can never make the mistake of equating the EVA system with "Voynichese", especially not when it comes to statistics like word length. EVA is fairly analytical stroke-wise.