Mariners "elite" starting rotation xERA and SIERA

[Sibelius Hindemith]

Via fangraphs here is each starting pitcher's expected ERA and skill-interactive ERA so far this season...

Kirby 3.41 and 3.71
Gilbert 3.95 and 3.45
Woo 3.11 and 3.75
Castillo 5.20 and 3.96
Hancock 4.06 and 3.36
Miller 2.15 and 3.23 (only 16 IP)

MLB average currently is 4.13 for xERA and 3.90 for SIERA.

What this indicates to me is that the Mariners don't have any true aces or even solid #1 pitchers. It is basically 5 #2/3 pitchers and a #4/5 pitcher.

[Sibelius Hindemith]

Definitions from AI...

xERA (Expected Earned Run Average) is an advanced baseball statistic that measures a pitcher's performance based strictly on the quality of contact they allow, rather than the final results.It takes a pitcher's Expected Weighted On-Base Average (xwOBA) and translates it to the traditional ERA scale, allowing for a direct 1:1 comparison with their actual ERA.

How xERA Works

-Quality of Contact: Statcast metrics track the Exit Velocity and Launch Angle of every batted ball. xERA uses this data to assign a "hit probability".
-True Skill Evaluation: By focusing on what happens at the moment of contact, xERA strips away the variables of fielding, stadium dimensions, and weather.
-Total Picture: Like actual ERA, it accounts for strikeouts, walks, and hit-by-pitches.

SIERA stands for Skill-Interactive Earned Run Average. It is an advanced baseball statistic used to evaluate a pitcher’s performance by measuring only what they can control (strikeouts, walks, and batted balls), while removing external factors like defense, park dimensions, and luck.

How SIERA Works

Unlike older ERA estimators like FIP or xFIP—which primarily focus on strikeouts, walks, and home runs while largely ignoring other batted balls—SIERA incorporates the type of contact a pitcher allows. It accounts for ground balls, fly balls, pop-ups, and line drives, adjusting its evaluations accordingly.

-What it rewards: High strikeout rates, weak contact, and ground balls.
-What it penalizes: High walk rates and hard contact.

[bpj]

Have never seen anything productive come from "expected stats".

[bpj]

Is it predictive? Is that the claim?

Here are last years xERA. Did it accurately predict 2026 current outcomes so far better than actual ERA?

Would be curious to see whether future production tracks closer to ERA or xERA.

[Donn Beach]

It's not trying to be predictive of the future, it's being predictive of what likely should have happened with a batted ball rather than what actually happened. Pitcher A has a bunch of seeing eye dribblers, broken bar hits and pitcher B has a bunch of line drive outs. As xERA would be higher and Bs lower in terms of what would have been expected. It's taking defense out of the equation. How much of the difference between the ERA of A and B is B having better defense behind them

[Sibelius Hindemith]

xERA is sort of predictive, but only of the near future. You are correct though that it attempts to interpret how things would have played out if luck were eliminated.

[Donn Beach]

Yeah, it's oriented to small sample sizes really. The larger the sample size the more they should converge one would think. What's indicated there is Castillo could be running into some bad luck so far this season

[Donn Beach]

Via fangraphs here is each starting pitcher's expected ERA and skill-interactive ERA so far this season...

Kirby 3.41 and 3.71
Gilbert 3.95 and 3.45
Woo 3.11 and 3.75
Castillo 5.20 and 3.96
Hancock 4.06 and 3.36
Miller 2.15 and 3.23 (only 16 IP)

MLB average currently is 4.13 for xERA and 3.90 for SIERA.

What this indicates to me is that the Mariners don't have any true aces or even solid #1 pitchers. It is basically 5 #2/3 pitchers and a #4/5 pitcher.

Terms of them not having a true ace, yup. There's only a handful of true aces to begin with. But they do have talent through the rotation. If you're calling Castillo a fifth starter he's league average. I'd put him up against other teams fifth starters

[Seattle or Bust]

I don't love FIP and xERA as stats purely because limiting runs is in fact a skill.

You can pitch to contact and be effective in this league.

There is such a thing as pitching to a location to induce a type of out. Hancock has gotten pretty good at it.

"Expected ERA, or xERA, is a simple 1:1 translation of Expected Weighted On-Base Average (xwOBA), converted to the ERA scale. xwOBA takes into account the amount of contact (strikeouts, walks, hit by pitch) and the quality of that contact (exit velocity and launch angle), in an attempt to credit the pitcher or hitter for the moment of contact, not for what might happen to that contact thanks to other factors like ballpark, weather, or defense."

I don't know this but I imagine this stat is flawed here:

A right-handed pitcher misses his location by a foot to the outer half of the plate to a lefty hitter. The defense is shifted, anticipating a pull. The hitter hits a soft liner through the hole on the left side. Because of the velocity, it is ruled a "lucky" hit when in reality, the pitcher missed his spot and the hitter took advantage even with weak velo. It is seen as a positive performance by the pitcher even though he failed his assignment.

Conversely, a right-handed pitcher throws a perfectly located fastball on the inner half... anticipating the hitter to pull the ball into a shifted defense. The hitter hits 100 mph liner straight to the right fielder for an out. This is ruled a "lucky" out... even though the pitcher executed his assignment.

[Sibelius Hindemith]

I take it then you aren't a fan of xwOBA either (with regard to evaluating batters).

Which ERA estimator do you like the most?