[Richard Wallace]

Originally Posted by mocat1530

Wow! Richard, I think you have formally been out-statistic-ed!
I never though I'd see that day!

The last line of my starting post was in the form of a question, not a conclusion. The question contained a challenge, which Bongle nailed.

And my original post contained no mathematical reasoning; only some bare data, arranged as it might be seen in an advertisement.

For another example that illustrates how sound statistical reasoning trumps sloppy inference from bare data, see the birthday problem. Briefly, given a classroom with N students, what is the probability that two of them have the same birthday? Many a statistics professor has used this one on the first day of a new class. And many a student has been surprised by the result.

My inspiration for the challenge was a stochastic processes professor, long ago, who told our class that "gambling is a tax imposed on those with poor math skills." The highest praise we could get from that professor was "you will be tax-exempt."

Statistics won't predict the future but they will tell you how you should bet. Most of us have a hunch that the form of a team's number shouldn't be predict the team's performance. Bongle's post showed that this is more than a hunch -- it is mathematically defensible.

Playing hunches is reckless. Playing the odds makes you a gamer.