I was intending to wait for the halfway point before delving into this, but I was asked questions about fielding and errors in the 1901 replay and at this point in the season I think there is more than enough data to pull some thoughts together on this topic.
First of all, in these old seasons there just are a lot of errors. A lot of errors. Player batting cards from this ERA come with two error numbers to help account for that, but is it enough? After reviews of my final error numbers from my 1930 and 1949 replays, I determined it wasn't going to be, but for a BBW replay I really can’t change the boards, so I was limited to changing batting cards. Players were by on-base percentage and the lowest 50% of plate appearances were assigned a third error number. I tried to parcel these out such that each additional individual error number was represented on a per-team basis. It didn’t always work out cleanly, but second base and shortstop were my focus here anyway.
First, a chart:
As you look down the actual numbers from 1901 to present you can see the tremendous drop in error results and percentages. Those numbers are the ones that I am shooting for as a re-player.
So how did I do? As I compiled these numbers it became quite apparent to me that any season with a normal distribution of error numbers and normal distribution of fielding ratings should have a very similar number of errors occur over the course of the replay. The fielding ratings for 1901 follow what I call a "normal APBA distribution" so following this logic if I got 1.5 errors per game with one error number per card then two error numbers per card should get me ~3.0 errors per game, and an extra error number should bump that up even more (a quarter-again as much?).
It's not quite working out that way. Based on my projections through Week 10 I am going to be off by almost 50% for the total number of errors. Conversely, though, my number of double plays per game is almost right where it should be.
The impact of this is felt throughout the replay. My ERA's are about a half-run higher than where they should be, but both leagues are well behind in their run totals. Having said all that, it certainly feels like I am getting plenty of errors - it is only afterward when I look at the totals that I see I am this far off.
When I look at the baseball guides from that season I find the following in the 1902 Reach Guide:
In the NL it specifies that there were a total of 14 games in which neither team made an error. It doesn’t make that distinction for the AL, but I can't imagine it would be that much different. I am not going to go back and do it now, but if I should roll an old season like this then this is something that might be worth tracking. My guess is that I get ~10 totally errorless games a week (~80 games a week).
If I was rolling dice I could easily have a separate fielding chart for error resultants and non-error number resultants. My non-error fielding resultants are OK, but I would want to change the fielding chart to ensure less "good" resultants would come up on error number checks. Even good fielders make a ton of errors in these old seasons. Obviously, I can’t alter this in the computer game, but as a point of reference BBW already makes some allowances for older seasons under its section on AIM rules, so it's not like there isn’t a placeholder for it:
Note: Thanks to Scott, an APBA buddy, for reminding me about this.
Regardless, I am happy where I am with 1901, but it does make me question what I might do for other replays. I am considering doing 1957 next - would I get enough errors with the defense as is, or am I likely to come up short again? I was thinking about possibly doing another "old" season after that. 1908? 1911? 1919? They would definitely need a bump in error occurrences. I don’t have to decide on that today though - more on that later.
Comments
Post a Comment