Posted by Trentonjoe on 3/6/2013 7:32:00 PM (view original):
I have rebuffing boogers advances for years. The guy just doesn't get it.

for the performance history, I don't really look. I imagine that the more seasons a guy has in history the more accurate it is.

I never play OL's and use performance history quite a bit.   I use it to judge  one players output relative to another, and get an idea of how many walks and XBH a position player actually gets, and how many walks and HR's a pitcher will give.  For fielders it gives me a good idea of how many +/- plays and fielding %.
It's useless for range.

Another thing you can infer from it is how often a player lasts a season.  If I see a HOFer or a guy with great raw numbers and a low amount of seasons used it may  be a clue that he gets waived a lot.

Posted by zubinsum on 3/6/2013 10:38:00 PM (view original):

Posted by frazzman80 on 3/6/2013 2:52:00 PM (view original):
I think it would have more impact on the fielding side than the range end with your C- vs B- example at the most important postions (SS/2B/CF) because the number of chances will be higher for these positions over the course of a season. So, a C- fielder (say .955 fielding percentage) is going to make many more errors than a B- fielder (.970 fielding percentage) due to the number of chances in the field. Your other positions would have fewer chances and thus fewer errors as whole.

While the C- fielder will make more "-" plays than the B- fielder at these positions, niether will have enough range to make any "+" plays and from my experience your are looking at a small difference in "-" plays between the 2 (say around 10-15 for the C- guy and 6-8 for the B- guy depending on the position). Thus, for range I wouldn't be concerned between the two as I would for errors.

This is all dependent upon position of course...as your 1B can get away with C- for either fielding or range and probably perform as well as the B- guy over the course of a season.

Here's my question: I often find that I rely on ERC+ and ERC# when drafting players and use this search stat on an equal level as OAV+/#, BB/9+/# and HR/9+/#, but I still have no idea if this stat is used in the engine's log5 formula or what would this number translate to (XBH allowed?). Is ERC+/# a worthwhile stat to use when selecting players and why or why not?

I see this question in two parts. First, I’ll answer the question: Is ERC used to determine the result of a PA?  The short answer is no, the primary stats used to determine the outcome of a PA are bb/pa, h/ab, hr/h.  In all cases both the raw stat and the pitcher's league average is used.  Earned Runs Created (ERC) is definitely not used at any significant level and is likely not used at all in any outcome calculations.  As someone else noted, other than HRs, pitchers have little influence on the hit type in the sim.

 

The second question is more direct: Is ERC+/# a worthwhile stat to use when selecting players and why or why not? ERC# is a good general estimator for a pitcher's ability to avoid (or allow) runs, but I would only use it as a rough tool to sort out potential pitchers. The reason is, that  ERC# is a composite stat and will be blind to your strategy, ball park or league. You are far better off focusing on the three stats: (BB/9#, oav# and HR/9#) that most directly correlate to what your pitcher can control. 

Thank you Zub...so falling in love with a high ERC+ should come second fiddle to the core components of (BB/9#, OAV# and HR/9#) all of which I also use. To me it's hard to turn down a high + number when drafting a player and I should focus on the # stats more.

The + number could be tough.  It is a great comparison if every player is from the same season.  But, it does not always mean much when comparing seasons from different era's

bottomlee, I think you are absolutely right about the + number but I think there might be one other case where it is useful, that is as a rough guide to dual seasons leagues, as often happens in progs or theme leagues. 

It is a kind of simple math way to estimate normalization - if you are in a league with two seasons, and need to choose between two players, since players will compete against both their own season's other players and those of the second season, it can give you some idea how the two players compare - one that is under the league average for his season is less desirable than one that was above that league average. 

As I say, it is a rough guide but I think the # normalization is somewhat less useful when dealing with only two seasons, though with 3 or more is the better guide I think. 

+ if the league player pool is limited to a set of years that are roughly equivalent

 # if the league player pool is the entire DB or a large fraction thereof

no easy answer if it's neither...

I agree with that Al.

On the subject of B/C vs, C/B, I don't think the letter grades alone give us an answer. The fielding grade doesn't give a true indication of the number of errors that might be predicted. And , the difference in the number of plus/minus plays from a B to a C OFer is so minor, it scarcely deserves mention.
   I always search for bargains in pitching. I often take the key stats of a pitcher I know to be successful, and search with them as guides, setting that pitchers $/IP as the max. I hope to find pitchers with similar key stats at a lower $/IP, and frequently do. I am of the opinion that dead ball era pitchers benefit more from ++ defense than do modern pitchers, but there are some modern pitchers who do have deadball era similarities. The trade off in ++ D has to do with value. We have to score runs to win games, and good offense and ++ D is very expensive. It would take a ton of plus plays to equal the positives that '81 Raines provides, for a very reasonable price. The variance in RRF #s also is key with D- fielders. Compare 2 cookie SSs, for example, Mark McLemore and HoJo. Huge difference there.