Re: Baseball

On Apr 10, 8:18 pm, "C...@xxxxxxx" <C...@xxxxxxx> wrote:
On Apr 10, 1:24 pm, "mensana...@xxxxxxxxxxx" <mensana...@xxxxxxx>
wrote:





On Apr 10, 3:05 pm, beewo...@xxxxxxxxxxx wrote:

On Apr 10, 3:53 pm, "mensana...@xxxxxxxxxxx" <mensana...@xxxxxxx>
wrote:

On Apr 10, 2:45 pm, "john0...@xxxxxxx" <john0...@xxxxxxx> wrote:

To keep it simple, before 1960 the National and American Basball
Leagues used to have 8 teams each. Each team would play every other
team in its league 22 times for a total of 154 games. Usually the
first place team would lose about a third of its games and the last
place team would win about a third of its games.

Suppose you had eight people "play" an identical schedule of card
cutting or something where it is all luck and no skill. Aproxomately
how would you expect "teams" one thru eight to finish in wins and
losses, say averaged out over 100 "seasons?"

They would all be tied.

My guess is that the OP meant to ask in the 100 seasons, what was the
average number of games won by the best team each season. Likewise,
what was the average number of games won by the worst team each
season. Certainly this question is much more interesting than the one
you answered.

Then he should have asked it. Instead, he asked what I would "expect",
not what would have actually happened.

There is a difference between expectation for a specific team (any
team) and the winning team or losing team. For N teams with completely
random wins/losses, each team has a 1/N chance of being the winner or
the loser or any specific rank in between, and each team has the same
expected rank. Nevertheless, the stats for the winning team are
different.

For one season. Read the OP question again: "averaged over 100
seasons".

The result of that averaging is that everyone ends up tied even
though there may be a particular order in any given season.

.

Relevant Pages