2014 set sleepers

However, it seems we're both happy with our methods. Someday, despite your reservations for the public leagues, it will be cool if we could match our methods and see how they come out...

I uh, wouldn't hold my breath waiting for that day to happen.

Also don't think it would prove a whole helluva a lot, cause were still talking about a game that involves roling dice, even if they're electronic and the bones roll funny for everybody at some point in time.

Also don't think it would prove a whole helluva a lot, cause were still talking about a game that involves roling dice, even if they're electronic and the bones roll funny for everybody at some point in time.

On that, we completely agree...

Coming in late, but anyway

((0.689* BB units )+(0.722* HBP units)+(0.892*1B units)+(1.283*2B units)+(1.635*3B units)+(2.135*HR units))/(AB+BB units+HBP units)

I've never seen Steve F formula, but I do know that any kind of formulas that predict run values can always be transformed to skip the 2b units/ 3b units without much errors. So you don't need to know the 2b/3b units. For example

(0.892*1B units)+(1.283*2B units)+(1.635*3B units)+(2.135*HR units) is roughly equal to

0.5*Hits +0.392*TB+0.067*HR

Only triples will be slightly off if you compare the two formulas, but they are so rare that you can skip off that difference---or if one wants to be picky, a slight adjustment could be done for the few players that have a lot.

Also, Steve said that superadvance takes off walks away from supercards such as Maddux, and that explained why walks are valued less. On this, two points

1-But in the good old days of Bernie H, he confirmed that this specific rule was OFF in online strat. So walks should be valued just as hbp.

2-But moreover, I don't think this is the reason why walks are valued less than than HBP. In most sabermetric formulas, which is based on real life, walks are valued less than HBP because walks tend to be more frequent in situations where sending one extra player on base cost less. Compared to HBP, Walks are much more prevalent with first base empty than with first base occupied---you know the typical semi-intentional walk.

If you look at any run expectancy matrix, consider for example this one based on 2014 values:

http://www.baseballprospectus.com/sorta ... id=1657937

you see that, with no one on base, no out, a walk has, on average, led to 0.36 runs. But with two outs, man on second, the walk has, on average, led to 0.11 runs, much less. In real life, whey you multiply these values with the frequencies with which walks happen, you get a value lower than 0.36 runs, in the range of 0.32-0.34 runs. But in Strat, walks happen independently of the situation (again, this is true because some of the superadvanced options available on the computer game are off in the online strat game---I'm thinking of the "pitching safely" mode or somethink akin to it). And it should be upgraded from standard sabermetric formulas based on real life---in steve's case, I would simply equal walks to hbp at 0.722.

Just to repeat: I'm not pushing for steve's formula, never seen it, so I can't judge whether the formula is right. But steve is a heck of a player, so I'm pretty sure he has some solid data to rely upon.

Thanks Marc. I can't take any credit for the formula, it's from SOMworld. I use it as is, but it's good to know that Strat online does not use some of the "max rules", they seem to make it less like the good ol Strat we grew up with. I'll have to consider giving the walk equal value based on your insight.

About the values on baseball think factory

There is a huge mistake in the cf/lf/rf sections: an incorrect huge jump from e0 to e1--which puts all outfielders way off of where it should be except for e0 outfielders.

Madsen's "when to play the slick player" has no such error, but it's based on OPS--onbase plus slugging, which we know is biased towards slugging, and I think it didn't take into account gbA.

The best published value for defense I know has been produced by our very own DeanC, which you can find here:

http://www.mfooz.com/bblog/wp-content/u ... efense.pdf (go see defensive charts on page 12-13)

Thanks for Dean's article once again. I believe I read it years ago when you posted it in a Strategy forum thread.

The data on the outliers -- either great fielders or awful ones -- is not surprising. What I did find intriguing, though, was that a hard-hitting infielder with a "3" range rating, for example, might be a perfectly fine choice over the "good field, no hit" guy often favored by Strat's salary calculators.

Great food for thought...

The Wiki shows "Correct Board Game Excesses" is turned on.

http://somonline.wikia.com/wiki/Maximum_Rules

This is what that does:

"This option improves overall statistical accuracy by adjusting for minor variances that cannot be accounted for in the board game.

Turning on this option improves the statistical accuracy of certain players who had extreme performances. For instance certain pitchers such as Greg Maddux allow very few walks. Using this option will enable Maddux to duplicate his real-life dominance in this area. A number of categories are affected by this option including home runs, walks and strikeouts. Also pitchers hitting will be affected by using this option, resulting in less walks and extra base hits by pitchers."

Okay, well I stand corrected. Thanks coyote for the catch. For some reasons, I thought they had removed this option with the Bonds controversy, but I guess they only removed the "improve statistical realism" option.

So I guess there is a case for giving less value to walks than to hbp. Although, frankly, this lesser value should be calculated only for extreme pitchers/hitters.

I also see that they have the "pitching around" option on, but I'm not sure how that translates in walks value. Actually, that rule probably lowers the probability of having a hit or a out, transforming some hits and outs into walks with first base open. But it probably doesn't affect the readings of walks and hbps. Just a guess though.

But then again, there's another thing I thought about after my first post. Walks contribute more to pitching fatigue; for a one-inning reliever, three walks and he's out. But hbp don't contribute to that rule. So walks (and hits) have an edge that hbps don't have.

So bottom line, I still believe that walks should equal just as much as hbp, after thoughtful considerations!!!

I remember those split decks - little orange cards - back from the 1970's. Man, I'm old.

So do I. I was in Coors a couple of years ago and lost a Homerun 3 times in a row. I posted what were the odds of that and some posted %5. In the orange cards I had at least 2 HR's

Marc,

At a basic level, the conversion you did to "Steve's" runs created formula is exactly how the strat data is converted into salaries. The actual values used have been posted before in the ATG threads and are based on Robert Johnson's (I hope I remembered the name right) linear runs created theories.