RCAP and RSAA are pure RUN measurements.

So, in 2000, each run of RCAP and RSAA is worth less than in 1968.

It'd be relatively easy, wouldn't it,

yes, but i don't think it would help him figure out what kind of batting to assign to Stieb, if that was what he wanted to do.

"One thing I'm struggling with is how many negative BRAR to give Stieb for playing in the AL. What was the typical rate for BRAR for NL pitchers per inning during this time? I've guessed -75 BRAR for his 2900 innings but Stieb was a converted position player and probably a decent hitter for a pitcher."

Don't - use RCAP for pitchers. Works, MUCH better. Pitcher is the one position where replacement level for hitting is what the average pitcher does.

Pitcher is the one position where replacement level for hitting is what the average pitcher does.

And Dave Stieb was precisely an average pitcher. Zero RCAP (or maybe some trivial negative value). As I said above, it makes no sense to me to speculate on his untested offensive abilities - this was his production.

Measuring his pitching by RA+ or ERA+ or DERA already automatically corrects for the fact that he didn't get to face pitchers. My take on it is if that if you found a NL pitcher with the same IP, the same RA+ or DERA, and that NL pitcher had near-zero RCAP, then I'd declare them equal.

Measuring his pitching by RA+ or ERA+ or DERA already automatically corrects for the fact that he didn't get to face pitchers. My take on it is if that if you found a NL pitcher with the same IP, the same RA+ or DERA, and that NL pitcher had near-zero RCAP, then I'd declare them equal.

Except that the NL guy did get to face the opposing pitcher and an AL guy wouldn't. Advantage Stieb.

But RA+ or ERA+ is comparing him to other pitchers in the same circumstances. It washes out.

The gist of my question is what -BRAA would equal 0RCAP for Stieb? I don't have access to RCAP and I'm not overhauling the whole spreadsheet without it.

Anyone want to tell me why pitchers with the same PRAA are getting more support with fewer PRAR (type 3) than the one's I like (type 1)? If I distill the whole discussion on the Hardball Times article of this week it's RAA that really matter for pennants added. So take two pitchers with similar PRAA. Why wouldn't you prefer the player who had more time as an above replacement to average pitcher (Tiant) over the player with less bulk (Stieb)?

DL, could it be that more of the type 3s have other characteristics - possibly years with negative PRAA, which if zeroed out would greatly enhance their career PRAA Don't know, just thinking out loud. If I ever get my Sinins dsik back on my finally-fixed home computer, maybe I'll get to your very good question.

I think Tommy Bridges looks a little better than Dave Stieb, then you add the war credit and the gap grows.

Stieb's advantages over Bridges and virtually all other pitchers to whom he is being compared are two-fold:

1. Stieb played in a more difficult league.
2. Stieb compared more favorably to his contemporaries.

These are related concepts. Since the quality of the league was so high, players could not consistently match the level of statistical performance that had previously been deemed HOF-caliber. Thus, while Stieb's numbers may have only placed him as the fifth or tenth best pitcher in a previous decade, he was the best pitcher of the 1980s.

Stieb's career primarily lasted from 1980 to 1990, an 11-year span in which he threw over 2500 of his roughly 2900 innings. According to Dan R's data, the league adjustment for that period is 1.005 -- no single year before 1980 or after 1992 has a value that high.

Similarly, Bridges had the vast majority of his success over a 12-year span from 1932 to 1943 in which he tossed over 2500 of his 2800+ innings. The league adjustment for this period is just 0.837, or 83.3% of Stieb's era.

These differences in league quality are reflected in the statistical rankings. Bridges finished his career with a 126 ERA+ and Stieb with a 122, but while Bridges never led the league in that category, Stieb did so twice. Stieb and Bridges each had five Top 5 finishes in ERA+, though Stieb had significantly more competition for those spots in a 14-team, integrated league. In contrast, Bridges was in a league of just 8 teams during segregation.

Moreover, Stieb finished in the Top 3 in both ERA+ and Innings Pitched for 4 consecutive years from 1982 to 1985, leading the league in each statistic twice. While his WARP and Win Shares totals over that period may not appear historic by any means, this strikes me as an incredible accomplishment. Bridges never led his league in Innings and only finished in the Top 3 in 1934 and 1936. 1936 was the only year in which he was Top 3 (or even Top 9) in both categories.

I understand why Stieb didn't make the Hall of Fame: not a lot of Wins or Ks, never got a Cy Young, etc. However, I think it would be a mistake if he isn't voted into the Hall of Merit. If Stieb can't make it, does any starter who peaked in the 80s have a shot?

Why wouldn't you prefer the player who had more time as an above replacement to average pitcher (Tiant) over the player with less bulk (Stieb)?

Well, if it took a Type 1 pitcher more innings than a Type 3 pitcher to earn the same number of PRAA, then it may be the case that the Type 1 pitcher was never as effective as the Type 3 pitcher was. You just can't tell from the career stats whose peak was better, and, if the Type 3 pitchers had better peaks, then that would be a reason for some voters to prefer them to the Type 1's. Although the pitchers can be grouped as types, I don't think it can be said categorically that one type is better than another based purely on career stats.

Tom's speculation about negative PRAA seasons would be a contributing factor. People wll slice careers in different ways, and value various sets of seasons differently, also.

I'm with Dandy Little Glove Man on Stieb vs. Bridges.

If Stieb can't make it, does any starter who peaked in the 80s have a shot?

If you count Clemens as having a peak in the 80s, the answer to this question is obvious. Clemens aside, I think that the only other candidate with a 1980s peak is Brett Saberhagen. I'm not sanguine about him getting elected because his career was so fragmented by injuries: he never had two great seasons back to back. But when he was healthy, he was just an outstanding pitcher. His career DERA, 3.57, is closer to Koufax (3.51) than it is to Stieb (3.89). His will be an interesting case, but he's the only 1980s aside from Stieb, Clemens, and 1970s holdovers like Blyleven and Ryan who has a case, I think.

The type 3 guys may also have bigger years than the type 1 or 2 guys.

If you count Clemens as having a peak in the 80s,

Which peak of Clemens' are you thinking of??? ; )

If Stieb can't make it, does any starter who peaked in the 80s have a shot?

I haven't actually looked yet, but I would guess that Hershiser and Viola rank somewhere behind Saberhagen.

Our standards for how much peak is enough are such that we haven't elected Dean and we never seriously considered Chesbro or Joe Wood - but somewhere in here we've got to mention the owner of the single best pitching season of the decade: Dwight Gooden.

I think I'm going to have Stieb rather high. I always supported Billy Pierce, I think they are extremely close.

Stieb is definitely one of the most underrated players (by the general public) in the history of baseball.

I also like Bridges and vote for him regularly. I have him slighly ahead of Stieb. I have them as Shocker, Bridges, Pierce, Stieb, but all so close that I can't draw the line. Followed by Newcombe (with war credit), Walters, Grimes, Tiant and Trucks.

That's in an older version of the spreadsheet though, with the team stuff only updated through 1991. Once I throw in the 1992-97 standings, those guys are so close things could change a little.

Of course, Rich Reuschel is solidly ahead of that pack :-)

Joe,
What am I missing from my ballot-thread discussion comment below (repeated here, but worth it I think given how close these guys are to voting thumbs-up and downs):

ERA+s, must pitch 154/162 IP, and at least 100 ERA+ that year

DavStieb 171 45 43 38 35 30 24 17 13 11
TBridges 147 44 42 40 40 40 37 34 20 19 19 15 09

DavStieb top 10 in IP: 1 1 2 3 5
TBridges top 10 in IP: 2 2 5 8 10

DavStieb top 10 in adj ERA+: 1 1 2 3 5 6
TBridges top 10 in adj ERA+: 2 2 2 3 4 6 6 10 10 10

Bridges never threw 200 IP or finished in the top 10 in his league in IP past the age of 30, making some of his ERA+s misleading (technically I should even dump the 134 and 109 ERA+s from above for 148 and 151 IP seasons, respectively).

Bridges does have a five-year "star pitcher" window from 1933-37:
1933 140 ERA+ (2nd), 233 IP (10th)
1934 120 ERA+ (10th), 275 IP (2nd)
1935 119 ERA+ (10th), 274 IP (5th)
1936 137 ERA+ (3rd), 295 IP (2nd)
1937 115 ERA+ (10th), 245 IP (8th)
Wonderful 1936 season, probably 2nd-best in league to Grove. Harder was better in 1933, but Bridges is in the next mix with Grove and Marberry.

Stieb also has a five-year "star pitcher" window from 1981-85:
1981 124 ERA+ (12th), 184 IP (5th, strike year)
1982 138 ERA+ (2nd), 288 IP (1st)
1983 142 ERA+ (3rd), 278 IP (2nd)
1984 145 ERA+ (1st), 267 IP (1st)
1985 171 ERA+ (1st), 265 IP (3rd)
Stieb's latter four years seem to beat or at least match Bridges' best year, don't they?

From Bridges' peak years, we've elected Grove, Ruffing, Lyons, Ferrell, Hubbell, BFoster, Paige, and RBrown (plus hybrid Dihigo), with 3 years of Vance and one each of Faber and Rixey. Walters also is on deck to add 1935-37 to this list.
From Stieb's peak years, we've elected Seaver, Carlton, Niekro, and Sutton, with 3 years each of Palmer, Perry, and Jenkins. Plus we'll have Ryan and Eckersley and Blyleven, and don't forget Morris.
The Bridges cohorts are more in their prime than the Stieb cohorts, it seems.

Bridges is a bit underrated, but I think Stieb will make my ballot and Bridges will not. Interesting pair, though.

I don't know whether or not Stieb will be on my ballot, but I'm confident that his peak and prime were greater than Bridges.

Okay, going over their careers, I do like Bridges better than Stieb due to the former's consistency and longer career (I'm giving him WWII credit).

Neither one will be on my ballot, though Bridges has an excellent shot to make it in the near future.

Dandy Little Glove Man, as far as my league adjustment data is concerned, two quick points:

1. I think you have the old league adjustment numbers, which I derived just from NL data--they shouldn't be applied to AL players. Also, since I finished the AL I re-did the regression, and the numbers came out fairly differently. The new league-specific numbers are up at the Yahoo Hall of Merit group. The correct adjustment for the 1932-43 AL is .927, while the 1980-90 AL adjustment is .987. So it would be a 6% correction for standard deviation to compare Bridges and Stieb, except that

2. My league adjustment data were only derived for position players. I have no idea how the standard deviation for pitchers has evolved over time, or what league conditions/factors have statistically significant relationships to it. You'd imagine that certain aspects would be similar (expansion, population), but others might be different (runs per game). When I have pitcher WARP I'll do an analysis of them, of course, but for now I don't think the position player league adjustment numbers are necessarily applicable to pitchers.

Thanks for mentioning Reuschel. There's a direct comp with similar PRAA, 200 more PRAR, a good bat and a good glove. The only thing missing seems to be consecutiveness.

we never seriously considered Chesbro or Joe Wood - but somewhere in here we've got to mention the owner of the single best pitching season of the decade: Dwight Gooden.

1) Wood -- His 1911 is pretty good too
2) Gooden -- More career than Chesbro, less peak than Wood
3) Chesbro -- Sorry Jack, you'll have to wait a little longer

2) Gooden -- More career than Chesbro, less peak than Wood

In context, I would easily take Gooden's '85 over Wood's '12. But you might have a point about Wood's peak as a whole, Jim.

1) Wood -- His 1911 is pretty good too

And Gooden's 1984 was ... Serious question: what does Gooden 1984 look like in a DIPS-based system? He had 276 SO in 218 IP, and very few HR (just 7). The walks - 73 in the 218 innings - were slightly better than league average of 3.2 per inning. The hits were low, but was he unlucky that they weren't lower?

Minor side issue - Gooden wasn't that bad a hitter. In 1985, compare his .226/.265/.280 to Rafael Santana's .257/.295/.302 (with a dozen IBB). Not a lot of motivation there to walk Santana to get to Gooden, is there?

Looked at my notes, instead of relying on my (oft-faulty) memory.

Upon further review:

1) Gooden -- More career than Chesbro, more peak than Wood (just 1984 tho)
2) Wood -- His 1911 and 1915 are pretty good too, giving him the better "prime"
3) Chesbro -- Sorry Jack, you'll have to wait a little longer

If I went by WinShares only, Chesbro's the best of the 3. BUT, his teams played great defense behind him, so WARP is less impressed. Gooden 1985 and Wood 1912 are pretty similar in WARP1 stature, but Wood was only the third best pitcher that year (see Johnson and Walsh), while Gooden was the best player in baseball, so I retract Wood having a better peak. (I really was unaware of Gooden's 1984 as I didn't pay much attention to the NL at that time.)

1984 was a long version of Vida Blue 1970. I didn't pay any attention until Peter Gammons (or a caption editor?) started calling him "baseball's Mozart" in 1985.

Jack Chesbro was a pretty good pitcher. I like Tannehill, Leever, and Phillippe better, but Chesbro was not bad. He doesn't deserve the bad rap, there are worse pitchers in the Hall of Fame, although I'm not sure I can argue him out of the bottom five.

On the other hand among NL/AL contemporaries Frank noodles Hahn was outstanding; Ted parson Lewis, Guy doc White, and Ned Garvin were pretty good. Bill Bernhard was pretty good for a few years in Cleveland; there's a late bloomer (age 31).

Gooden's rookie year is one of the nastiest DIPS seasons evar, right up there with Clemens 1997 and the Unit in 1995/2001. Only Pedro 1999 (and maybe 2000, I'd have to check) is clearly superior.

Just a lurker, not a voter, but I do have a question:

It is now 1999. Stieb pitched in 1998. Does that (should that) make him ineligible for the 1999-2003 votes?

We've chosen to ignore token appearances like Stieb's 1999 for eligibility purposes, so as to better compare players to their peers.

We've chosen to ignore token appearances like Stieb's 1999 for eligibility purposes, so as to better compare players to their peers.

Luckily he didn't grow hair on his feet before 1999, I don't think we have rules for Tolkein appearances.

What did Stieb throw? His hit prevention on balls in play is off the charts--almost like a knuckleballer's.

My recollection on Stieb was a good fastball with movement but a wicked slider.

From The Neyer/James Guide to Pitchers:

Pitches: 1.Slider 2.Sinking Fastball 3.High Fastball 4.Curve 5.Change

Blue Jays manager Bobby Cox: "He isn't Nolan Ryan, who overpowers you. But he throws two outstanding fastballs, one that sinks and one he throws by batters up. He has that awesome slider, a curveball, and a change-up, throws them all for strikes and isn't afraid to throw any of them at any time."

His sinker was like Wang's. A lot of balls were hit into the ground, then the fielders would err and Stieb would glare.

the fielders would err and Stieb would glare.

He didn't allow many in the air
but check my ballot - Steib's there.

When did misspelling Stieb's last name become trendy? I before E, except after C, or when sounding like A as in neighbor and weigh!

Sorry, I'm a slave to the rhyme. Spelling is of no consequence.

I before E, except after C, or when sounding like A as in neighbor and weigh!

Is there a sceintific basis for such wierd spelling rules, or are they merely constructs of soceity?

Is there a sceintific basis for such wierd spelling rules, or are they merely constructs of soceity?

I'm not a linguist, but I took German. In German, where i and e are concerned, when one follows the other, the long sound of the second vowel is tyically pronounced, but the first vowel is not pronounced. For instance, Stieb is steeb, not sty-eeb or stibe or stib.

Some German examples:
Liebsche (lover) = lee-beh
bleiben (to stay, IIRC) = bly-ben

But the words in English where the combination is /cei/ usually are French words: receive, for instance. But not every French word does that: experience. Yet in these latter French words, both vowels in the dipthong are pronounced (ex-peer-ee-yence) as the tongue glides backward to accomodate them. So I wonder if you have a combination of German and French influences that described by a single, easy to remember rule that just happens to fit both.

With any luck a linguist or philologist among us can get into the nitty gritty of the Great Vowel Shift and other historical comings and goings and shed more light on the specifics of why exactly i virtually always preceeds e except in Frenchy words with a c in them.

Liebsche (lover) = lee-beh

whoops! meant to say leeb-sheh, i was going to use Liebe at first, but chose the more familar and forget to edit my comment.

Making rules based on a mishmash of source languages is asking for trouble. For that matter making rules for English is asking for trouble, period. English spelling and pronunciation is just a mess - but, by golly, it's our mess.

One source of confusion: all those names which end in -stein (including Stein itself as a name), occasionally Anglicized to -stine. Stein is the German word for "stone," and it's a very common element in German and Yiddish names. (Yiddish being a language built on a Germanic base.) And, of course, it's pronounced in German exactly as Eric stated in #141 - as "stine". (Well, no, it should really be "shtine.") Plenty of examples out there - for instance, Steinway the piano makers. What was that originally? Steinweg, maybe, in which case the current version is half-translated?

But just to mess us up, a significant number of Americans with names ending in -stein pronounce it as "-steen." A few even spell it that way. Is that more common with the Yiddish names than the German names?

For what it's worth, I do assume that the name Stieb is of German origin, surviving relatively unchanged. (As opposed to "Quisenberry" which is probably German and certainly didn't start out spelled that way.)

There are plenty of ordinary exceptions to "i before e" (weird, as Eric says) and it is a worthless guideline for proper nouns.
Leif Erikson
Madeleine

On the ballfield, take Teixeira for example or two.
Burgmeier
Eichhorn
Klein
Leibold
O'Neil
Reilly

Heigh-ho, heigh-ho, it's off to work we go!

Madeleine

And curiously, this one has two common pronunciations too, neither of which is ni the ee vein:

Ma-del-ine
Ma-del-in

The first with the long vowel i, the second with the schwa.

No wonder you guys can't get your ballots in on time. You're all ADD!

Check out the 1981 Blue Jay lineup: team OPS+ 74. http://www.baseball-reference.com/teams/TOR/1981.shtml

According to Dan R's warp calculations, the position players totalled a magnificent -5.7 warp 2.

No wonder Stieb was frustrated.

Yep, they were pretty stinky (although I seem to have them at -4.8, not counting guys with fewer than 50 PA--which file are you using for the 5.7?). That said, there have been worse. The 1920 Philadelphia Athletics combined the same 74 OPS+ with some of the worst team defense on record, so they take the cake at -9.7 position player WARP2 (tied with the 1899 Spiders). The rest of the bottom 10 are the '79 A's, the '63 Mets, the '09 Braves, the '19 A's, the '88 Braves, the '83 Mariners, the '04 Senators, and finally the '81 Jays. On the flip side, the teams with the best position players were the '01 Mariners, the '76 Reds (the '75 are #16), three of the 1890's Orioles teams ('97, '96, and '98, with '94 not far behind), the '27, '31, and '98 Yankees, the 1902 Pirates, and the '06 Indians (with monster seasons from Lajoie, Flick, and Terry Turner, they led the league in run scoring and were one of the best-fielding teams ever, but mediocre pitching held them to a third-place finish).

the '06 Indians (with monster seasons from Lajoie, Flick, and Terry Turner, they led the league in run scoring and were one of the best-fielding teams ever, but mediocre pitching held them to a third-place finish).

Really??
They finished at Pythagoras -9 and 5 games behind.
Pwins Wins
Cle 98 89
Chi 90 93
NY. 87 90

ERA+/OPS+ = 123/125 looks like a strong and reasonably balanced team. Three starters at 142-145-151 in 900+ innings and poor Bill Bernhard at 103 in 255 innings.

In their first good season, 1904, they also finished at Pythagoras -9, and 7.5 games behind.
Pwins Wins
Cle 95 86 (pyth 95-56)
Bos 95 95 (pyth 95-59)
Chi 92 89
NY. 84 92 (actual 1.5 behind)

And when they famously finished 1/2 game behind Detroit in 1908, that was Pythagoras -2.
Pwins Wins
Cle 92 90 (actual 0.5 behind)
Det 88 90
Chi 85 88

Cleveland AL, 1901-1910
OPS+ERA+: ERA+ for top four pitchers by starts
_86 _93 : _92 122 _94 _81
113 104 : 116 124 156 _87 (Joss rookie)
115 104 : 130 164 134 117
121 115 : 119 160 106 113 (Joss leads team in ERA+, first of six seasons)
108 _92 : 130 _99 _93 _82
123 125 : 142 145 151 103 (1906)
_99 111 : 137 122 110 108
101 118 : 205 135 109 112
_94 106 : 113 _93 149 106
_95 _90 : _88 102 100 _85 (Joss sixth, 107 inn @ 114)

This team generally enjoyed good pitching and in 1906, exceptionally, the top four pitchers worked most of the innings: Bill Bernhard #4, 255 innings at ERA+ 103.

Yes, Paul, but that's my point--that ERA+ is the product of the fielders, not the pitchers. The team had a monster 124 BP FRAA, led by what I imagine was the best single-season defensive keystone combination ever in Nap Lajoie and Terry Turner (although I'd love to be corrected if anyone can find a better one). The team's NRA was 3.94 and its DRA was 4.73, suggesting that while its run prevention was responsible for 10.1 wins above average, with average fielders its run prevention would have been 3.8 wins below average (and then just multiply the gap by 9 to get the same 124 FRAA). You can even see this from the raw defensive statistics. The team turned 72.7% of balls in play into outs, while the other 7 teams averaged 69.9%. And they turned 111 double plays thanks to Lajoie and Turner, while the other 7 teams were all between 69 and 86 DP's. It would be virtually impossible to have a below-average ERA on that team as long as you could throw strikes. You're right that they won the Pythagorean standings by 8 games and "should have" won the pennant anyways, but what I was getting at is that their collection of position players, by leading the league in runs scored and playing such exceptional defense, was among the 10 greatest groups of position players on the same team since 1893.