Assessing poems

Orbis magazine invites readers' votes and brief comments. I never have voted, though I've been tempted to offer comments. I tend to assess in various contradictory ways. Over-simplifying, and depending on the situation, they include -

• Bottom-up - I give points for various features (use of sound, etc) or (as in diving) combine degree of difficulty with performance
• Top-down - I first decide whether I like the poem or not, then I list its obvious features showing how they support my opinion: e.g. if a poem has tight integration of form and content I can say that this reveals technical prowess (if I like the poem) or that the poem has stifling predictability (if I don't). A poem may be understated (if I like it), or lacking verve (if I don't).
• Emotion - a piece may move me though I know it's not a good poem - it may not even be a poem, or I know I'm moved only because it describes something I've experienced.
• Learning resource - a poem may open my eyes to new poetic possibilities, inspiring me to write. It may not be good.
• Best bits - it's tempting to judge a poem by its best (often last) lines. Sometimes ("Lying in a hammock at William Duffy's farm in Pine Island Minnesota" maybe?) the last line justifies the 'blandless' of the rest of the poem.
• Good of its type - however good some poems are, they're restricted by the type of poem they are.