The same encyclopedia entry surveys serious objections to the design inference: the normalizability problem, observer selection, and the possibility that the constants aren’t independent. Isn’t citing it only for the phenomenon selective?

The entry is used for the narrow point that the constants sit in narrow ranges, which it confirms, and the objections are taken up in the exhibits that follow: the multiverse, the anthropic reply, and the firing-squad response. It is fair to say that the same source presents the objections, and you should read those sections too. The normalizability problem in particular is the strongest technical objection, and it is treated under the Science case’s fine-tuning claim.