Starters ERA vs Bullpen ERA

[jabail]

I've noticed the difference in starter's ERA and Bullpen ERA is much higher than in real life. In most sims, the difference is >.50 runs, which is a lot. For example, in my most recent sim, The starter's ERA was 4.23 and the bullpen ERA was 3.65, a difference of .58.

In 2022, MLB starter's ERA was 4.05, and bullpen ERA was 3.86, a difference of .19. In my test above, I was using 2022 league modifiers, so that's why I wanted to compare it to 2022 real-life stats. I thought comparing to 2022 was best since we don't know exactly how much more offense there will be this year due to the rule changes.

For reference, here are some more recent years:

2021
Starters: 4.34
Relievers: 4.17
Difference: .17

2020
Starters: 4.46
Relievers: 4.42
Difference: .04

2019
Starters: 4.54
Relievers: 4.46
Difference: .08

2018
Starters: 4.19
Relievers: 4.08
Difference: .11


So in the previous 5 seasons, the average difference was only .12, but in my example, it was almost 5 times that.

Anyone else noticing this? I feel this could use some tweaking

[jcard]

Originally Posted by jabail

I've noticed the difference in starter's ERA and Bullpen ERA is much higher than in real life. In most sims, the difference is >.50 runs, which is a lot. For example, in my most recent sim, The starter's ERA was 4.23 and the bullpen ERA was 3.65, a difference of .58.

In 2022, MLB starter's ERA was 4.05, and bullpen ERA was 3.86, a difference of .19. In my test above, I was using 2022 league modifiers, so that's why I wanted to compare it to 2022 real-life stats. I thought comparing to 2022 was best since we don't know exactly how much more offense there will be this year due to the rule changes.

For reference, here are some more recent years:

2021
Starters: 4.34
Relievers: 4.17
Difference: .17

2020
Starters: 4.46
Relievers: 4.42
Difference: .04

2019
Starters: 4.54
Relievers: 4.46
Difference: .08

2018
Starters: 4.19
Relievers: 4.08
Difference: .11


So in the previous 5 seasons, the average difference was only .12, but in my example, it was almost 5 times that.

Anyone else noticing this? I feel this could use some tweaking

For what it is worth, I noticed the same thing last year and believe at least included the observation in a post. I will see if I can find it.

[jcard]

Last half of second paragraph (hope you get a better response than did I):

https://forums.ootpdevelopments.com/...hlight=bullpen

[Lukas Berger]

I guess we might need to look at the third time through the order penalty here, and make sure it's properly calibrated.

Is this just in the roster set with the real 2023 MLB players, or the same in a fictional or even a historical league as well?

[Garlon]

The penalty for third time through the order is part of it, but I used to notice this as well and corrected it through adjusting the strategy settings.

The setting for SP Stamina is Very Low for modern seasons. Maybe try Low instead.

How many relievers are allowed on the active roster? Reducing the amount of relievers by 1 may also help to balance things.

[Rain King]

Originally Posted by Lukas Berger

I guess we might need to look at the third time through the order penalty here, and make sure it's properly calibrated.

Is this just in the roster set with the real 2023 MLB players, or the same in a fictional or even a historical league as well?

In the fictional league I am currently running the Starters have an E.R.A. of 4.41 and the Relievers have an E.R.A. of 4.31.

[Garlon]

Another thing to consider is the import settings. There are 30 teams and just to get to 150 SP who are not adjusted you have to look at the season and see that the pitcher with the 150th most innings had 91 IP that season. 91/4 = 23. You will need to lower the import settings significantly to get more SP with better ratings.

For historical s you really need to balance the import settings and strategy settings.

[Lukas Berger]

Originally Posted by Garlon

Another thing to consider is the import settings. There are 30 teams and just to get to 150 SP who are not adjusted you have to look at the season and see that the pitcher with the 150th most innings had 91 IP that season. 91/4 = 23. You will need to lower the import settings significantly to get more SP with better ratings.

For historical s you really need to balance the import settings and strategy settings.

Right, that all makes sense, but my main question is more which mode folks notice this the most in.

Historical does not have the third time through the order penalty (except in recent years), so I would guess this is less of an issue there than in other modes.

I also suspect it's possible this is the least close in a current 2023 game using the roster set, as you have the double challenge there of making sure to correctly balance the ZIPS ratings (and any edits) between SP's and RP's which would already have the third-time through the order penalty baked in (though we do adjust for that on the initial import), and then the actual third-time through the order penalty itself as applied in-game.

[md40022]

I’m not even at the all-star break yet in season 1 of my sim, so my sample size is small. But I’m definitely noticing a large ERA gap between starters and bullpen. I use modern day rosters. Haven’t really pulled any numbers yet since my sample is so small, but if others have noticed this as well and the numbers back it up I would definitely say it might need a little bit of tweaking

[jabail]

Originally Posted by Lukas Berger

Right, that all makes sense, but my main question is more which mode folks notice this the most in.

Historical does not have the third time through the order penalty (except in recent years), so I would guess this is less of an issue there than in other modes.

I also suspect it's possible this is the least close in a current 2023 game using the roster set, as you have the double challenge there of making sure to correctly balance the ZIPS ratings (and any edits) between SP's and RP's which would already have the third-time through the order penalty baked in (though we do adjust for that on the initial import), and then the actual third-time through the order penalty itself as applied in-game.

Not sure if it helps much, but I did sim a fictional season based on 2022 MLB and the starter ERA was 3.90 and the reliever ERA was 3.71, so much closer.