You're starting to sound a little defensive, and I'm probably the last guy to tell someone else they're doing anything wrong. Part of the reason WiS deliberately keeps their game mechanics opaque is to keep from having One True Way To Value Players.

I monitor stats all the time. I'm probably one of the few people that looks at minor league boxes every cycle as well. Mostly to ***** about underperformers, but that's part of the fun, too. My big crunches happen at the end of seasons, and that's where I look at running three-year averages to consider making changes to my ratings engine. I'm 100% sure I don't have it right, but I'm hoping that every iteration brings me closer to something I can use with confidence.

Not being defensive. Just failing to understand why you will insist on sticking with something that isn't working.

One piece of advice I give n00bs who are suffering losing streaks is "Do something". It isn't difficult to look at your team and say "We're 30th in batting average and 28 in Runs Scored. What can I do to make my team better?" Logic says you need to score more runs. Logic says you can't get hits. Logic says bench players who aren't hitting. At that point, you can stick with ratings that aren't working or bench players who aren't hitting. I bench players who aren't hitting.

I have a formula for hitting and pitching. The hitting formula works pretty well based on OPS. I can plug my 10 players with 100+ AB(my baseline) and there might be 1 or 2 that don't run in order. But I don't hesitate to deviate from the rating formula when the stats tell me I shouldn't be playing someone. I do fairly well at HBD.

By the way, three year averages are virtually worthless by nature. Owner changes, engine changes and random results can skew numbers in all sorts of directions. Look at a player's career numbers. That will tell you how he performs throughout all the changes.

To me, people who don't adjust their thinking when they're failing are the same people who insist on cramming the square peg in the round hole. It's not working. Put the hammer down and think for a moment.

Nonetheless, this is really getting away from the point of the thread.

If owners would like ST to have some meaning, even the tiniest, craziest bit of meaning according to saint and goose, my idea works. Rather than "try out" a player during the season, you can do it in ST in a controlled, by you, enviroment. To me, that would add some meaning to ST.

If it doesn't work for you, nothing hurt. You'll still get to play your games against yourself and you can use training camp pitchers and career minor leaguers to your heart's content.

Quote: Originally posted by MikeT23 on 5/21/2010By the way, three year averages are virtually worthless by nature.

So three year averages are worthless, but 18 games are meaningful?

All performance is relative to league norms. Each of the worlds I'm in performs very differently.

I guess I'm lucky enough to be in fairly stable leagues (well, at least the private ones I'm in), so it doesn't take long to recalibrate based on league averages, since they don't swing very far from year to year.

Controlled enviroment.

I've already said 50 AB is a small sample size. In fact, I've said an entire career is a small sample size. There are 180000+ AB in each world in each season. A player's career might equal 3.5% of that. Not exactly a huge sample size. But, in a controlled enviroment, I believe you can learn just as much in a small sample size as you can over a career in an uncontrolled enviroment.

I guess, as best I can tell, you have 3 complete seasons. I'm not sure what sort of baseline you can get on stable worlds since you're playing your 2nd season in exactly one of them. Unless, of course, you've done extensive research on the last three seasons in each of those worlds. But there's really no point in debating the quality of the worlds you're in vs. the quality of the worlds I'm in(you'll lose). This isn't a you vs. me situation. I think more info is good. You're opposed to it. I'm not sure why.

And, again, we're straying from the point of this thread.

If two players are a "coin flip decision" because their ratings are similar but different, what's the harm in giving them 50 AB each, in a controlled enviroment, and taking the better performer to the bigs? Seriously, why does the thought of this offend you so? Are you some numbers geek who just can't get past that 50 AB is a SSS(which I've said about 50 times in this thread)? Why is this such a big deal to you?

Quote: Originally posted by MikeT23 on 5/21/2010And, again, we're straying from the point of this thread.If two players are a "coin flip decision" because their ratings are similar but different, what's the harm in giving them 50 AB each, in a controlled enviroment, and taking the better performer to the bigs?   Seriously, why does the thought of this offend you so?  

It doesn't offend me. I've even said I don't it's a bad thing to work the way you do. Our only point of disagreement is that you think those 50 PAs are meaningful, and I'm saying you'll have just as much success differentiating them by literally flipping a coin as you will using those 18 games to decide a winner.

mike, one thing to note is that in considering sample size, you don't need the sample to be big relative to the overall population. a lot of political polls that you see, for example, only sample 1,000 or so people and can generalize successfully to the entire country. this isn't intuitive, but it's true from a statistical standpoint.

if a career is a small sample size, why have a hall of fame?

Saint, I'd prefer 50 AB in a controlled enviroment, you'd prefer to flip a coin. Fair enough. I think my way is better. There's a 50/50 chance we'd end up with the same conclusion.

Sched, I tend to agree that you don't need the sample size to be big in relation to the overall population. Which is why I'd be fine making a decision based on 50 AB. In the scheme of things, one player's career is a small sample size. He performed well, under the circumstances, and that is being recognized by the HOF.