I was looking at some of the past single-season leaders in a long-running theme I'm in and noticed 1904 Jack Chesbro's coming up back-to-back in a lot of categories.  Check out these two seasons (from an 80M w/average AAA twist theme):

Past Season Totals - Regular Season
League	Owner	G	GS	W	L	SV	SVO	IP	BFP	NP	H	ER	HR	BB	SO	OAV	OBP	SLG	WHIP	ERA
90089	eastonest	63	63	31	21	0	0	498.7	2139	6985	493	177	6	144	216	.253	.306	.321	1.28	3.19
88648	eastonest	62	62	31	19	0	0	486.7	2103	6837	496	172	5	129	195	.258	.306	.322	1.28	3.18
Career	Totals	125	125	62	40	0	0	985.3	4242	13822	989	349	11	273	411	.255	.306	.321	1.28	3.19

Now that's astonishingly "unrandom"!!, looks like a pattern, and shouldn't even out like that in the short run.  An "Irrandomer" like me could have a field day with this info. Must be a catch (smile). Because I am respectful of the threads of others, I'll just leave it alone, keep this thread on topic, let others draw conclusions, and belatedly congratulate you on that "valid" OL worldseries win.

It gets less random as you approach 500 innings.

Now, some people I know dump "cold" pitchers after 3 or 4 innings.  But hey, that sort of logic doesn't mean they can't still make the playoffs (as long as they're in a division where the other three owners lose 100 games apiece!).

"...and shouldn't even out like that in the short run."

This is a statistically invalid statement. Inferential statistics say nothing about what should happen in the short run. In the short run, there is more variability, but things are still more likely to be close to average than far away from it. We shouldn't expect things to even out in the short run, but at the same time, we shouldn't be surprised if they do.

Now, for the long run. Let's say that in this particular league, Chesbro can be expected to post an OAV of .255. Then, 95% of the time, all things being equal (opposing talent, ballpark, etc), we can expect his OAV after 2100 batters faced (a round number close to the examples above) to be between .253 and .257 (2 standard deviation from the mean).

Across WIS, we seldom can cling to the "all things being equal" caveat. Talent varies widely from division to division, league to league, salary cap to salary cap.

I know this has been beaten to death in other threads/discussions, but thought I'd show the math above again. What strikes me (and I'm assuming what strikes crazy as well) is how similar the numbers are across the board.

Let eastonest beware--he's playing with fire. He's due for a bad version sooner or later!

Your point is well taken AKlopp, the statement IS invalid because there are no absolutes when dealing with random sequences. (except that they are absolutely random). "...My statement was refering to how strikingly consistant the numbers are across the board with all of the variables involved, not just the OAV.  Good information on Standard deviation, this topic needs more of that.  And as for Stengel's comment on numbers getting "less random"?  Now that may be as invalid as my statement.   Hmmmm. Sounds like something an "Irrandomer" (as you so 'eloquently' have called me) would say....... (fading back into silence)

How is saying that things get less random as samples get larger invalid?  That statement is the basis of all statistical studies everywhere.  It's the basis of the whole field of statistics.

crimsonblue, you can substitute the words "more meaningful" for "less random," if it helps you understand better. There's of course an element of randomness in every plate appearance; for a pitcher, results after 3 innings can look more random than they do after 500 innings. But then, some of us are smart managers who "know" 3 innings is enough to assess a pitcher's performance!

 Flip a coin 10 times and heads might be the result 20% or 80% of the time.  Flip a coin a trillion times and it's going to be heads 50% of the time.

Well dahsdebator, randomness, or the act of being random, is a finite term as it is best described as the opposite of deliberate by most dictionaries.  If a sample is not absolutely random then it is not random at all or it is non-random. Sort of like someone saying that someone is mostly or a little pregnant. Unless you believe that some events in the "Sim" are not random, then that's an area that I'm not going into again. "More meaningful" that's good Stengel, you're now approaching coherence. (humor),  As Aklopp accurately stated in his post "there is more variability".  Variability and randomness are not one in the same.............But because "it's all semantics"  I totally understood what Stengel was trying to convey........A little harmless cage rattling usually gets him going.  And no, 3 innings is not enough time to evaluate a pitcher, I usually wait until the 4th.(humor)..........(unless prompted again, fading farther into silence)

Did I say farther?  Next we'll be discussing the semantical usage of Further and Farther.  Whew. 

Farther is used to indicate geographical distance, while further is idealogical. I pressed my point further as we walked farther into the jungle. Now that was random.