2020 Forecaster - replacement levels

[kez]

The hitting replacement levels on p. 31 are identical to the numbers from the 2019 book. Is it accurate that those numbers were used to calculate RAR this year, or did that table miss an update in the editing process? Posting updated numbers would be appreciated.

As well, replacement levels used for calculating pitching RAR would be useful. It would be nice if those numbers were included in the book too, it's odd to show how you get to your H RAR numbers but not P RAR.

[RAY@HQ]

Oooh, nice catch. Yes, just an editing error. Here are the 2020 values:

AL PROJ
Ca 3.14
1B 4.22
2B 4.14
3B 4.41
Ss 4.33
Lf 4.44
Cf 3.92
Rf 4.56
Dh 4.68


NL
Ca 3.65
1B 4.88
2B 4.05
3B 4.73
Ss 3.93
Lf 4.58
Cf 3.89
Rf 4.49

As far as pitcher RAR, from the site glossary: (I suspect we use the actual Games projection now rather than the hack version in the formula below, but can't easily confirm that)

Start with the replacement level league ERA.

Subtract the pitcher's ERA. (To calculate projected RAR, use the pitcher's xERA.)

Multiply by number of games played, calculated as plate appearances (IP x 4.34) divided by 38.

Multiply the resulting RAR level by 1.08 to account for the variance between earned runs and total runs.

[kez]

Thanks for the H numbers. My question about Ps is not how to calculate RAR, but what replacement level feeds into that calculation. In a perfect world, the book would include something like:

AL Repl ERA
SP 5.00
RP 4.25

NL Repl ERA
SP 4.75
RP 4.10

[RAY@HQ]

League avg ERAs are in the Cheater's Bookmark in the back.

[JonE]

Are you saying, Ray, that replacement level is league average? I would think that player-pool average would be higher than league average, because players are at least modestly successful at not rostering the worst pitchers who bring down the league average. But I would not have guessed that leagues are only rostering pitchers at or above league average. I've never looked at the relationship between the two, and wonder whether you've done research.

[RAY@HQ]

Ah, good question. I failed to consider that. I'll have to dig in the Forecaster template and see if we have a different replacement ERA value configured in there.

But it's also a weird thing because there are so many relievers, etc., who may not actually be positive-value players but who have a good ERA. It's a bit of a rat's nest. Basically, I've not done much thinking about how we calc replacement levels for pitching.