Matt Cain had a very nice season this year...but his ERC# is lower than every single season of Roger Clemens' career.

As surprising as these results are, I believe them to be correct after checking and re-checking several times.

There are two ERC (component ERA) formulas I'm aware of.   This is the one we are using. It has two steps.

Step 1:
((((hits allowed+ walks allowed + hit by pitch allowed)*(((((hits allowed - HR allowed)*1.255) + 4 * home runs allowed)*.89)+((walks allowed - Intentional walks allowed)*.56)+(hit by pitch allowed *.56)))/cast(plate appearances as real))*9)/innings pitched

Step 2:
if result from step 1 >= 2.24 then Result - 0.56
else
Result * 0.75

The normalized value then takes into account the league average from the pitcher season, the all-time league average and the pitcher values in the standard log 5 equation.

I've run and re-run for players in 2011 and previous seasons. All values appear to jive.

2011 was a season where the elite pitchers dominated more so than in other previous seasons.

If anyone finds any flaws, I'm happy to look into more but I'm pretty confident in these numbers as being both correct and consistent with what we have put out for all previous seasons.

thanks

Tom, sorry to continue to beat on this, but I'm still wondering.  I've just plugged the raw numbers for Verlander, Kershaw, Weaver, and Hamels into Excel, and this is what I'm coming up with:

Verlander = 1.92
Kershaw = 1.997
Weaver = 2.27
Hamels = 2.23

These are very very close to the values reported on baseball-reference (all within 0.01).  But quite different from the results listed here.  Is it remotely possible that either (a) the formula has a parenthesis misplaced or something causing it to compute slightly differently or (b) perhaps one of the input values is off?

So tzent, by your math his 2.31 ERC normalized down to a 1.78 ERC#? In a pitchers' year? That's what seems suspect. It's a notorious pitchers' year, it seems highly unlikely for someone's ERC to normalize down that much in one.

edit: Oh, I just noticed that the WiS DB has Cain as a 1.75 raw ERC. Using the formula you just posted and simply replacing the words with Matt Cain's raw values from this year, I just got 2.31, which seems far more realistic.

OK, I am 99.999999% sure this is it.  You are applying the 0.75 reduction twice.

For all 4 pitchers, this is exactly the disparity between the numbers here and the numbers cited elsewhere.

Posted by contrarian23 on 10/4/2011 2:44:00 PM (view original):
OK, I am 99.999999% sure this is it.  You are applying the 0.75 reduction twice.

Great catch, contrarian.

Good job.  It was obvious there was something off; I'm just glad there was someone smart enough to say exactly what was causing the error.

I knew I called you the Great Contrarian for a reason.  :-)

Guys -- thank you for spotting this and being persistent.  I took for granted that a manual mistake hadn't been made running queries, but there apparently was one.

contrarian23 is correct -- the second step of the basic ERC formula was being applied twice.

We have updated the values and corresponding salaries this morning.  This resulted in all pitcher salaries going down. We have updated all teams in progress of being built and all teams that had already been built using the 2011 pitchers so that their salaries reflect the correct pitcher salaries.  New downloadable spreadsheet will be available shortly.

All data now JIBES.

Thank you

Good stuff, tzent.  Thanks for correcting it so quickly.

Thanks. The revised values make Verlander a pretty interesting play for modern leagues...

Great job guys, and thanks to TZ!