I went back and read the delphi forum about fielding. SOM isn't a individual base situation game as much as a catch all the base situations game in one swoop. IOW, the x-chart doesn't adjust by bat hand or by base situation and how it effects each fielder in a given base situation or his committing an error by base situation. IOW, the base situation correlations are condensed into an equal overall correlation for all base situations or even strategies. For example, over 50% of pitcher errors are on pick offs or bunt errors. A range 1 pitcher with a e-rating of e-51 will have 88% of his x-chart rolls being errors with only 12% of the x-rolls on the actual x-chart for the out. Strat-o-matic and other baseball board games are in error by mixing fielding range and error ratings into a combined system. Basically one has very little correlation to the other. Another point would be first and second base have more errors with left handed batters and shortstop and thirdbase have more errors with right handed batters. The pitcher card x-chance rolls do not adjust up or down by the bat hand. Another example, a first base man will make fewer errors with runners on third versus a runner being on first and this also applies when considering the base situation sample sizes. A note about pitcher's in SOM will have more errors allowing the batter to reach first base versus all the pickoff errors that only allow a runner to advance a base or two. I always use SOM for these examples due to it is my game of choice and the game design known best for contributing real life data into their design. Best for comparing the reality of baseball. SOM is definately a BABIP game, since if you miss rolling a HR, K, BB, or HBP basically all that is left is the BABIP. However; my knowledge of the SOM game doesn't mean the other baseball board game sims are off the hook for not having some of the same misconceptions.
Thanks for your comments. You make some interesting points. The point about errors is interesting. I was having a discussion recently about the vintage SI Baseball game. (In that game, errors are completely generic and the defense aspect is a team concept that is loosely based on fielders' ranges.) Anyway, in SIBB, a little more than 1% of plate appearances will result in an error. Using 1971 NL data, that would work out to about 750 errors. In actuality, there were 1613 errors. According to Baseball Reference, in 1971, 930 NL batters "Reached on Error". That leaves 683 other kinds of errors. Now, SIBB is a simple game so there should be no explanation that SIBB would capture these other kinds of errors, or at least not many of them. However, other more sophisticated games probably attempt to capture some of these other errors. Examples would include throwing errors, errors on pickoffs (as you mentioned), etc. This is not something that I have studied extensively. My main reason for pointing out the Delphi discussion relates to how you integrate or separate fielding from pitching in a sim design.
@steve_etzel Strat-O-Matic "error count totals" will be correct via their x-chance system design. But in actuality, the correlation of when the errors occur by base situation, by bat hand, and the type error result (ie, pickoff errors, infielder throwing errors, and outfield throwing errors off hits) will be in error. The difference is playing the game as a kid versus playing the game as an adult with an increased knowledge of the game of baseball result occurrences. Strat-O-Matic could greatly enhance their game by implementing realistic error results just by expanding their x- charts where warranted by base situation, bat hand and etc. If only as programmed for the baseball pc game. An example of how off the error result occurrences are off. Runner on 1st base only and a pitcher x chance is rolled, half the pitcher pick off error e-ratings under runner 1st base only would be a e-51. This conclusion is arrived at by using real life data and pro rating the pitcher original e-rating to fit the base situation. IOW, if the pitcher x-chance is rolled on anything beyond a runner on 1st base only situation the pitcher e-rating would generally to greatly decrease from his original e-rating. It's like if the game was programmed to get more homers outside what the base situation sample sizes should permit. For example, the natural correlation for the amount of homers correlating with the base situation sample sizes. The same is true via player position, base situation, bat hand, type of error and etc. I noticed this youtube channel doesn't allow web links to be implemented. So, go to the delphi stratomatic forum and look under the category of board game rules and innovations. You should be able to find a link to my download page for applied design changes for SOM baseball bases on real life data. IIRC, my user name is Johnny Smoke.
Steve - it might be important to clarify...In Replay, the OF and IF defensive ratings are "inverted". So while its intuitive that Belanger and Brooks Robinson received the best ratings of "1", a rating of "5" for an OF is also gold glove caliber...so of these eight fielders, the only one that was rated average was Powell; all others were rated above average. Not sure what the catcher's ratings were but both Hendricks and Etchebarren were above league average in CS% - so even more "help"? I found the delta between McInally's FIP and ERA interesting...most years he was pretty close. Of course, we know now that Wins are a somewhat spurious statistic...but the average IP/start comment was also interesting. I wonder how that compares to the Billy Martin-era A's of the early 80s...???
Thanks so much for the clarification. Further evidence of how good the Orioles' defense must have been. So much time had passed since I started building the slide deck that I forgot how the Replay ratings worked.
Thank you Steve, always very interesting. The 1972 season ended up being a pitcher's year similar to 1968.
I went back and read the delphi forum about fielding. SOM isn't a individual base situation game as much as a catch all the base situations game in one swoop. IOW, the x-chart doesn't adjust by bat hand or by base situation and how it effects each fielder in a given base situation or his committing an error by base situation. IOW, the base situation correlations are condensed into an equal overall correlation for all base situations or even strategies. For example, over 50% of pitcher errors are on pick offs or bunt errors. A range 1 pitcher with a e-rating of e-51 will have 88% of his x-chart rolls being errors with only 12% of the x-rolls on the actual x-chart for the out.
Strat-o-matic and other baseball board games are in error by mixing fielding range and error ratings into a combined system. Basically one has very little correlation to the other.
Another point would be first and second base have more errors with left handed batters and shortstop and thirdbase have more errors with right handed batters. The pitcher card x-chance rolls do not adjust up or down by the bat hand. Another example, a first base man will make fewer errors with runners on third versus a runner being on first and this also applies when considering the base situation sample sizes.
A note about pitcher's in SOM will have more errors allowing the batter to reach first base versus all the pickoff errors that only allow a runner to advance a base or two.
I always use SOM for these examples due to it is my game of choice and the game design known best for contributing real life data into their design. Best for comparing the reality of baseball.
SOM is definately a BABIP game, since if you miss rolling a HR, K, BB, or HBP basically all that is left is the BABIP.
However; my knowledge of the SOM game doesn't mean the other baseball board game sims are off the hook for not having some of the same misconceptions.
Thanks for your comments. You make some interesting points. The point about errors is interesting. I was having a discussion recently about the vintage SI Baseball game. (In that game, errors are completely generic and the defense aspect is a team concept that is loosely based on fielders' ranges.)
Anyway, in SIBB, a little more than 1% of plate appearances will result in an error. Using 1971 NL data, that would work out to about 750 errors. In actuality, there were 1613 errors. According to Baseball Reference, in 1971, 930 NL batters "Reached on Error". That leaves 683 other kinds of errors.
Now, SIBB is a simple game so there should be no explanation that SIBB would capture these other kinds of errors, or at least not many of them. However, other more sophisticated games probably attempt to capture some of these other errors. Examples would include throwing errors, errors on pickoffs (as you mentioned), etc. This is not something that I have studied extensively.
My main reason for pointing out the Delphi discussion relates to how you integrate or separate fielding from pitching in a sim design.
@steve_etzel Strat-O-Matic "error count totals" will be correct via their x-chance system design. But in actuality, the correlation of when the errors occur by base situation, by bat hand, and the type error result (ie, pickoff errors, infielder throwing errors, and outfield throwing errors off hits) will be in error.
The difference is playing the game as a kid versus playing the game as an adult with an increased knowledge of the game of baseball result occurrences.
Strat-O-Matic could greatly enhance their game by implementing realistic error results just by expanding their x- charts where warranted by base situation, bat hand and etc. If only as programmed for the baseball pc game.
An example of how off the error result occurrences are off. Runner on 1st base only and a pitcher x chance is rolled, half the pitcher pick off error e-ratings under runner 1st base only would be a e-51. This conclusion is arrived at by using real life data and pro rating the pitcher original e-rating to fit the base situation. IOW, if the pitcher x-chance is rolled on anything beyond a runner on 1st base only situation the pitcher e-rating would generally to greatly decrease from his original e-rating.
It's like if the game was programmed to get more homers outside what the base situation sample sizes should permit. For example, the natural correlation for the amount of homers correlating with the base situation sample sizes. The same is true via player position, base situation, bat hand, type of error and etc.
I noticed this youtube channel doesn't allow web links to be implemented. So, go to the delphi stratomatic forum and look under the category of board game rules and innovations. You should be able to find a link to my download page for applied design changes for SOM baseball bases on real life data. IIRC, my user name is Johnny Smoke.
Steve - it might be important to clarify...In Replay, the OF and IF defensive ratings are "inverted". So while its intuitive that Belanger and Brooks Robinson received the best ratings of "1", a rating of "5" for an OF is also gold glove caliber...so of these eight fielders, the only one that was rated average was Powell; all others were rated above average. Not sure what the catcher's ratings were but both Hendricks and Etchebarren were above league average in CS% - so even more "help"?
I found the delta between McInally's FIP and ERA interesting...most years he was pretty close. Of course, we know now that Wins are a somewhat spurious statistic...but the average IP/start comment was also interesting. I wonder how that compares to the Billy Martin-era A's of the early 80s...???
Thanks so much for the clarification. Further evidence of how good the Orioles' defense must have been. So much time had passed since I started building the slide deck that I forgot how the Replay ratings worked.