re: The "X all time record over Y bodes well for X" argument is dumb.

It’s like trying to use historical weather data to predict what the weather tomorrow will be like. Even though your data is valid, there are simply too many factors at play to be able to speak confidently about the outcome of a single event.

Now, if you want to use historical data to predict future trends, that’s much more valid. We may not be able to say, “Team X beats Team Y about two thirds of the time, so there is a 66% chance that Team X wins this year’s game,” because that’s too specific. But if we predict that, say, out of the next twenty years, there will be a 66% chance that Team X beats Team Y in any given year, well, that’s much more likely to be close to the mark.