I wanted to share some thoughts about the sim engine and defensive statistics. I set a team's ratings to average values for fielding ability and arm accuracy to look at fielding percentages in the sim engine. I simmed ten seasons and recorded fielding stats.
While I am sure that many of you are aware that fielding percentages are too high, I wanted to point out the following:
I averaged the number of errors (in one season) for each position and compared to MLB statistics for 2015. Here is the breakdown:
POS......... 2015......... SIM
P............. 13.0.......... 7.5
C............. 9.4........... 5.9
1B........... 9.0........... 13.5
2B........... 13.0......... 16.8
3B........... 18.5.......... 9.3
SS........... 17.0......... 12.4
LF............ 5.2........... 4.6
CF............ 4.5........... 5.1
RF............ 5.0........... 9.3
Aside from pitchers, the one that stands out the most is 3B. In real life, there are nearly twice as many errors as what we see in this game. All in all, you can see the distribution is significantly off.
I bring this up in part because many have noticed that ERAs and run production are too low. The discrepancy in errors is large enough I think to consider it as a contributing factor.
You might think that more errors would lead to more runs but not higher ERAs, but consider that more errors means longer innings, more plate appearances, higher pitch counts and downward pressure on starter innings. I first noticed this when I was playing around with fielding ratings on a 1941 test roster. When I tried to get fielding percentages down a bit I noticed higher ERA's and a massive increase in run production - and I never got the fielding percentages down far enough to reach actual 1941 levels.
So I think there's something off here that may explain a number of deviations in the sim engine. My hope is that devs will look into this...maybe an easy fix, or it's equally possible that fixing this might cause a leak somewhere else in the boat.
On a related note, it's possible to see how errors generated are affected by manipulating the fielding and arm accuracy ratings. Setting one to 99 and the other to zero will give you an idea how fielding and throwing errors are generated. Doing this for catchers I found that setting arm accuracy to 0 and fielding to 99 generated on average about 21 errors. Setting the opposite yielded only 6 errors, however.
In real life, about 75% of all catcher errors are throwing errors, so the above figures would seem to indicate that this is backwards in the sim engine. If you have a catcher that is, say, 65 and 65 then he's generated way too many fielding errors (which are quite rare for catchers...just think about it) and way too few throwing errors.
I didn't do this for the other positions but perhaps that may explain the above chart.
Anyway, just some thoughts. I humbly request the devs to take a look if they haven't already.
While I am sure that many of you are aware that fielding percentages are too high, I wanted to point out the following:
I averaged the number of errors (in one season) for each position and compared to MLB statistics for 2015. Here is the breakdown:
POS......... 2015......... SIM
P............. 13.0.......... 7.5
C............. 9.4........... 5.9
1B........... 9.0........... 13.5
2B........... 13.0......... 16.8
3B........... 18.5.......... 9.3
SS........... 17.0......... 12.4
LF............ 5.2........... 4.6
CF............ 4.5........... 5.1
RF............ 5.0........... 9.3
Aside from pitchers, the one that stands out the most is 3B. In real life, there are nearly twice as many errors as what we see in this game. All in all, you can see the distribution is significantly off.
I bring this up in part because many have noticed that ERAs and run production are too low. The discrepancy in errors is large enough I think to consider it as a contributing factor.
You might think that more errors would lead to more runs but not higher ERAs, but consider that more errors means longer innings, more plate appearances, higher pitch counts and downward pressure on starter innings. I first noticed this when I was playing around with fielding ratings on a 1941 test roster. When I tried to get fielding percentages down a bit I noticed higher ERA's and a massive increase in run production - and I never got the fielding percentages down far enough to reach actual 1941 levels.
So I think there's something off here that may explain a number of deviations in the sim engine. My hope is that devs will look into this...maybe an easy fix, or it's equally possible that fixing this might cause a leak somewhere else in the boat.
On a related note, it's possible to see how errors generated are affected by manipulating the fielding and arm accuracy ratings. Setting one to 99 and the other to zero will give you an idea how fielding and throwing errors are generated. Doing this for catchers I found that setting arm accuracy to 0 and fielding to 99 generated on average about 21 errors. Setting the opposite yielded only 6 errors, however.
In real life, about 75% of all catcher errors are throwing errors, so the above figures would seem to indicate that this is backwards in the sim engine. If you have a catcher that is, say, 65 and 65 then he's generated way too many fielding errors (which are quite rare for catchers...just think about it) and way too few throwing errors.
I didn't do this for the other positions but perhaps that may explain the above chart.
Anyway, just some thoughts. I humbly request the devs to take a look if they haven't already.
Comment