The base-rate predictor got 65% right by definition, the logistic managed to score 68% correct (bootstrap20 95% CI: 66-72%), and the random forest similarly got 68% (67-78%).
which doesn't really sound all that great.
But one useful way to look at it is to see this as indicating how good the base rate predictor is compared to random commenters or your gut feel - 65%? Did you know, in advance, before I spent like 3 months collecting the data, that the right rate was 65% rather than, say, 10% or 90% or 50%?
Am I missing something? What's so special about Translate?
Normal predictors for shutting down google services doesn't apply to it, because it is very different in its behavior than other google services. That doesn't make it a bad algorithm.
Is there any alternative out there? I would like to try it out.
Or maybe they're tragically out of touch.
(I'll go with the first one.)