Newb, but 30-years Scoresheet. I like their realism, which is evident here also. But here's an example I'd like better understanding as applied to game results and player stats across different eras and sometimes year-to-year :

PitcherA with league average 5.00 ERA in my rotation

PitcherB with league average 4.00 in my rotation.

Similar innings. Everything else considered equal, at the end of the year do they meet in the middle both with a 4.50 ? Or do they further split the difference of their raw on paper stats and end up closer to :

PitcherA 4.75
PitcherB 4.25

I have other examples, but any info/advice would square those as well in better understanding ranks within salary values, thanx -

You can look at normalized numbers in the draft center instead of raw numbers and get a sense of how the players will compare across eras. These are the numbers with the # at the end.

Note that ERA isn't a very useful stat to look at, it has a minimal impact on performance. If you just look at OAV, BB/9, and HR/9, that's almost all of what you need.

There is a stat in the draft center called ERC that combines all this info into one useful number. It's a Bill James stat. If you look at the normalized version of this, you will get a very good idea of how a pitcher will perform, and how pitchers from different eras compare.

I believe in the sim, ERA is used to help determine XBH - but yeah otherwise it’s not worth much salt.

ERA as well as Wins-losses means nothing here. They aren't factored in.

ERC# is your best example of what your pitcher is capable of.

Posted by DoctorKz on 5/20/2020 11:44:00 AM (view original):
ERA as well as Wins-losses means nothing here. They aren't factored in.

ERC# is your best example of what your pitcher is capable of.

While ERA isn’t a huge thing that I really pay attention to, I wouldn’t say they aren’t factored in, as it is used to determine XBH, at least so long as my memory serves.

I don't think it is part of the algorithm.

The decision tree relies more heavily on the batter's stats than those of the pitcher in determining what occurs in the plate appearance. Deadball pitchers have a built in advantage in limiting XBH.

Someone else can fill in the details.

I think jfranco77, italyprof and contrarian23 have answered this one before. The pinned threads probably have the answer. I can't look now for it.

Posted by DoctorKz on 5/20/2020 12:01:00 PM (view original):
I don't think it is part of the algorithm.

The decision tree relies more heavily on the batter's stats than those of the pitcher in determining what occurs in the plate appearance. Deadball pitchers have a built in advantage in limiting XBH.

Someone else can fill in the details.

I think jfranco77, italyprof and contrarian23 have answered this one before. The pinned threads probably have the answer. I can't look now for it.

ERA is listed here as an insignificant stat, but they do make a note that they use it to estimate 2B and 3B. A regression from ERA and starting with HR and H allowed rates is how they determine the 2B and 3B allowed rates for pitchers. With the rate on doubles allowed factoring in to the batter/pitcher equation at that point at around 15% and the rate on triples allowed factoring in on the batter/pitcher matchup factoring in around 11%. So, the ERA portion is very small and still mostly insignificant, but it is used to approximate the XBH allowed rates, which are then used in the Log5 portion of the tree at that step, though that step puts more weight on the batters rates, so in essence it's a % of a % of a %, so very small effect. But when rolling a 1,000-sided die, thousands of times, those small effects can be milked and built upon to deliver more consistent results.

Thanks, I stand corrected. I never pay attention to ERA but I guess I will.

Thanx for all the quick replies. I'll tweak accordingly. But, imo estimation, of the 1 example I listed of a handful of other topics I could've offered, to keep it thread relevant of the replies that came in, ERA is the most important of what *DID* happen (everything else considered equal, although I'm very happy there's league-to-league year-to-year era-to-era balance factoring), not overly reliant on saber what could/should/might have happened.

I'm as big a saber guy as the next, but with big picture sights as an aid of what will happen in the real MLB, and less of engine driven SIM could/should/supposed to happen. But I'm already hooked so I'll play along and learn more and game the stats, although it's less than 99 of a 100 on the realism scale I had hoped.

WIS doesn't use any saber stuff. They make stuff like ERC or Runs Created available as a general guide, but the sim just uses standard rate stats like OAV, HR/9, BB/9, K/9, etc. ERC is directly derived from those four rates. They're not doing defense-independent stuff and ignoring what happens on balls in play.

Those rate stats are IMO a better reflection of what really did than ERA. By itself, ERA has a lot of flaws. Runs that are entirely the pitcher's responsibility can be unearned. Example: a starter gets the first two outs, the third hitter reaches on an error, and then the next six batters hit home runs before the starter is pulled. All seven of these runs are unearned, but all six home runs were 100% his fault and he should be penalized for them.

When ERC and the underlying components of fip are offered, we are indeed speaking saber. Which I'm ok with. But not in a vacuum and certainly not comparable with reality. We can do this would've could've should've all day, but in the meantime, to my original query thanx again for a better understanding of how some of this works, reality or otherwise. I hope to see some of you soon; it can only make me better and I've backed off Scoresheet in large part out of boredom in rolling standard leagues.

What makes this worth playing are the many theme and progressive leagues, as well as the varying salary cap leagues. You get to use a wide variety of players from 130+ seasons. For many here the drafts and team building are more enjoyable, and count more than how it plays out.

Thanx DoctorKz. I'm already all in cuz of the endless possibilities. A bored with standard old-timer leagues core at Scoresheet has done recent outside the box adventures. For 'fun', a few of us returned to a kids table standard league (and an easy trophy), and got cracked on a few times. One of the literal quotes was who let the chess players in when we're ok with checkers ?! We'll, I already see this as an additional upgrade to 3D-chess. I've tried floating this to others, but most seem content playing in the sandbox. I already have themes in mind and would be happy to hear of others. I'm sure we'll talk soon.

First timer. From an earlier thread that started with comparing stats across eras and even year-to-year, it turned into advice that my team's arms are going to suck and why. Ok, still not a fan that there's little chance of success when I used ERA+ which does factor year-to-year league average, than the woulda coulda shoulda ERC# and K9# used here.


So, I ordered another team. Same bats, but much different arms doing it your way. There's no comparison which team did better irl even the zero factored W/L was better for my first team (TmA) 'wrong' selections, but look at the numbers and tell me how much better I should do with the new and improved TmB.

Disclaimer : both teams salary are on a shoestring budget, but I did increase from 13M to 21M and looked better at pitcher's D. Thanx in advance :

Tm#A = original arms
Tm#B = better (?) arms

ERA
A - 3.88
B - 4.04

ERC#
A - 4.63
B - 4.29

K9
A - 3.79
B - 5.56

K9#
A - 3.99
B - 5.17

ERA+
A - 109
B - 101

Defense
A: A=5, B=13, C=5, D=3
4 A+, 1 D-
B: A=10, B=8, C=6, D=2
9 A+, 0 D-

I like the increase in Ks, which I don't think is in ERC but I've heard helps (fewer balls in play; less homers ?), and Defense is much better, if that matters much as a pitcher. But both of these changes I did on my own. Other than that, did much else change for the better ? My first 3 games with Team#A, my cheap 'Aces' are 0-3. When Tm#B starts next week, they'll open with Gultz'77/Guidry'80, followed by a lot of nothing. Thanx for any additional info or advice.

I don’t like ERA+

Here’s an example where I think it’s flawed.

2015 teammates John Lackey and Michael Wacha

Lackey - 1.211 WHIP

Wacha - 1.213 WHIP

Lackey - 0.9 HR/9

Wacha - 0.9 HR/9

Lackey - 142 ERA+

Wacha - 116 ERA+

ERA is already a stat that relies on luck, and being taken out mid-inning could spell disaster to your ERA due to someone else’s bad pitching.

Now if we use Component ERA, which I think is a better gauge on one’s performance, we see that Lackey has a ERC+ of 114 and Wacha has an ERC+ of 116. Much more similar than what ERA+ gave us.


Also, an ERC# above 4 is nowhere near good enough if you want to compete.