I.D can explain more of the observed facts evolution can: how evidence works.

[Donald Ian Rankin]

42. Against, of course, as is made plain in the OP - try actually reading it.

ID and evolution are both compatible with all the observable evidence.

ID would be compatible with any conceivable observable evidence, past, present or future. It therefore has no predictive power, and is useless.

Evolution would be incompatible with practically anything except what we actually do see. It can therefore be used to make predictions, and is useful.


The point I was trying - and clearly failing - to get across is that "evidence" is best understood in terms of Bayes' theorem:

(Estimate of probility of Hypothesis given Observation) = (Probability of Observation given Hypothesis) x (Initial estimate of probability of hypothesis)/(Initial estimate of probability of observation under average of hypotheses)

P(H|O) = P(O|H)P(H)/(P(O|H)P(H) + P(O|notH)P(notH))


In this case, the observation is "facts compatible with evolution". The probability of that given that evolution is a correct hypothesis is 1; the probability given that ID is a correct hypothesis is less than 1. So whatever your initial estimates of the probability of the two hypotheses (the P(H|O) term on the RHS), you should increase the likelihood you assign to evolution in light of it.

More generally "evidence for a hypothesis" is "any observation that is more likely if that hypothesis is true than if it isn't".