Testing Diamond Mind's 1919 Season
A quick experiment
Testing Diamond Mind's 1919 Season
Diamond Mind Baseball came out with the 1919 season just the other day.
This is big news. It’s the first time Diamond Mind has created a deadball era season. And I wanted to see how well it worked.
I set up a single replay — one run through with the auto manager on, using real life transactions and lineups. I know that a single sample doesn’t necessarily tell us much — but I was still curious to see how well the game engine would do.
It was okay, I suppose.
Here’s what the real life 1919 American League batting stats looked like:
And here’s what the replay looks like:
The rate stats are dead on. However, runs scored aren’t. Diamond Mind runs hot by about 500 runs. I’m not entirely sure where the extra runs are coming from, since doubles, triples, home runs, and bases on balls are about the same. At bats are also quite close, which tells me that Diamond Mind is probably bunting often enough.
My guess is that runners are advancing too many bases.
Here’s the National League:
And here’s Diamond Mind:
It’s the same thing. All the rate stats and all the raw stats are very close, if not right on. And yet we’re running into a significant number of extra runs.
That has an impact on the pitching stats. Here’s the American League in real life:
And here’s the sim:
The ERA in Diamond Mind is too high - by a significant margin. Note as well that complete games are too low - about a hundred complete games are missing.
And here’s the real life National League:
And the sim:
We see the same thing here. ERA is about 20 points too high, and there are about 80 fewer complete games than there should be.
It’s not necessarily conclusive, but I think Diamond Mind likely has problems with a few things:
I doubt the computer manager in Diamond Mind Baseball will ever be able to manage the bullpen appropriately. You may want to control substitutions for both teams if you do a manual replay.
Diamond Mind Baseball seems to have an issue with runners either moving too quickly around the bases, or not being thrown out enough while on the bases. It’s hard to tell which it is, of course, since we don’t have complete real life records to compare with. My guess is that there are probably some outs on the basepaths that are missing.
I’d love to know what you think and what your experience has been.