NL Rookie Perfectas

50 Replies

According to Wins Above Replacement (WAR, baseball-reference version), the top two rookies in the National League this past season were Jose Fernandez (6.3 pitching WAR) and Yasiel Puig (5.0 WAR). How often have two rookies in the NL both reached 5 or more WAR (baseball-reference version, pitching WAR for pitchers and overall WAR for everyday players) in the same season, prior to this year?

The last time that happened in the NL was all the way back in 1972. That season two pitchers, Jon Matlack (Rookie of the Year award winner) and Burt Hooton, both topped 5 pitching WAR, Matlack with 6.2 and Hooton with 5.1. Burt had a fine rookie season, including a no-hitter to start the season. That created quite a sensation, especially coming after two September cup-of-coffee starts the previous year that had produced a 15-strikeout three-hitter and a two-hit shutout to defeat Tom Seaver and the Mets. But Hooton finished 1972 with an unimpressive win-loss record of 11-14 that produced no Rookie of the Year votes. Dusty Baker also finished with over 5 WAR that season, and he would have been a solid Rookie of the Year candidate, having accumulated only 103 MLB at bats before 1972, but he had exceeded the maximum 25-man roster time before ’72, so was not considered a rookie for award purposes.

1968, 1967 and 1964 all produced two NL rookies with 5 or more WAR. In 1968, it was Johnny Bench (5 WAR) and Jerry Koosman (6.3 WAR), with Bench taking the Rookie of the Year award and Koosman coming in second in a very close vote.

In 1967, two pitchers, Gary Nolan and Tom Seaver, both reached 6 pitching WAR as rookies in 1967, Nolan at 6.3 WAR and Seaver at 6.0 WAR. Seaver won the Rookie of the Year award, with Nolan finishing third in the voting. Dick Hughes was second in the ROY voting — he had a lower WAR total but was a significant factor in the Cardinals’ capturing the pennant that season.

In 1964, Richie Allen (later known as Dick Allen) had one of the greatest WAR rookie seasons ever with an 8.8 WAR. And Jim Ray Hart of the Giants produced 5.3 WAR. Allen received 18 of the twenty votes for Rookie of the Year, with Hart and Rico Carty (4.8 WAR) splitting the other two.

Going back further requires a jump to the World War II era to find an NL season with two rookies (well, sort of) reaching 5 WAR. In 1943, Lou Klein (5.8 WAR) of the league-dominant Cardinals and Braves pitcher Nate Andrews (5.1 WAR and a fine 131 ERA+, but an unfortunate 14-20 W-L record) both topped the 5 WAR level. However, it looks like Andrews had more 25-man roster time before 1943 to have qualified as a “rookie” in 1943 under the modern definition. Neither Klein nor Andrews found much post-war success in the majors.

In 1941, Elmer Riddle had an excellent season (5.8 pitching WAR, best ERA in the league, 5th in the MVP voting) for the Reds, with only 33 IP under his belt in prior seasons. But he had been on the Reds’ major league roster the whole prior year, so that 1941 performance would not be considered a true rookie season under today’s rules. That same year, Ernie White had a pitching season for the Cardinals that was almost a clone of Riddle’s, just a slight step behind in WAR , ERA, wins and MVP votes (6th in the voting to Riddle’s 5th). In contrast to Riddle, White probably was a true rookie in 1941, having spent most of his 1940 MLB debut season in the minors.

Before that, you have to go all the way back to 1905 to find two NL rookies with 5 or more WAR. That season two NL pitchers made their MLB debuts in stunning fashion. Big Ed Reulbach and Irv Young (who was 27 years old when he turned pro) were two of the three best pitchers in the league (along with Christy Mathewson) in 1905, despite the fact that neither had pitched in the majors before. Young had 9.2 pitching WAR that season for a Boston Beaneaters team whose total team WAR was 10.3.

0 0 votes

Article Rating

50 Comments

Oldest

Newest Most Voted

Inline Feedbacks

View all comments

mosc

11 years ago

Cool piece birtelcom, very thought provoking.

I have to say though that I really have come to detest arbitrary WAR cutoffs as a point of interest. 4 WAR, 5 WAR, 6 WAR, they’re just this season’s latest brew on how to best measure total contributions. They’ve changed in the past and they’ll change again. Say what you want about older statistics but they don’t really change. I don’t have to look up how many home runs Hank Aaron hit but his career WAR? Who knows.

Author

birtelcom

11 years ago

Reply to mosc

Arbitrary cutoffs (be it WAR totals or home run totals) are always just for fun, to provoke some thinking (to use your phrase, mosc), never to prove anything. And WAR represents simply a good faith, and ongoing experimental, effort to summarize as many different aspects of baseball performance as possible in a common currency of win value. It gives us, among other things, a useful way to compare players across contexts using a common currency that is hard to do other ways. This sort of post would be much more difficult, and I think less useful, without something like WAR,… Read more »

mosc

11 years ago

Reply to birtelcom

WAR is the best thing we have, until something better is created. I agree. I have my own list of specific complaints with it don’t get me wrong but on the whole it is the best tool available.

I just think the worst way to utilize it is to assign importance or significants to WAR thresholds. That, and presenting absurd significant figures (I would express single season WAR with a single significant figure, career with 2 significant figures).

11 years ago

Reply to mosc

“I just think the worst way to utilize it is to assign importance or significants to WAR thresholds.”

Right, but are .300, 30/30, 20 wins, 100 RBI, 200 Ks also not arbitrary thresholds? It’s a shorthand, imperfect way of saying how someone performed that season. You have to create some floor for good performance, otherwise this article would just be “NL Rookies who were worth more than 0 WAR in a season”.

11 years ago

Reply to RJ

Or “NL Rookies”. 🙂

Editor

John Autin

11 years ago

Reply to RJ

… or “Not Adeiny Hechavarria.” 🙂
http://www.baseball-reference.com/players/h/hechaad01.shtml

mosc

11 years ago

Reply to RJ

But 6 months from now we’re not going to change the definition of a Strikeout now are we?

brp

11 years ago

Reply to RJ

@22, no mosc, we won’t, but a pitcher with a K/9 of 6.5 is a hell of a lot less impressive now than it was in 1970. And if we compile a list of guys with 9+ K/9, isn’t that also arbitrary? What about guys with 8.9 who don’t make it due to rounding? How is it fair to pitchers from the 1940s, when hitters had different approaches to ABs and relief pitcher usage was vastly different? Your complaints are valid but the point is you can make similar complaints about any list of any statistic or combination of statistics… Read more »

Voomo Zanzibar

11 years ago

Reply to RJ

How do we look at a league average for WAR/162 in a given year?

vincent

11 years ago

Reply to mosc

I really wasn’t much of a WAR fan in the beginning, but the more I hear people on this site talk about it, the more rational it becomes. The more I hear other, major media outlets pimp it (or go the other direction) the more I dislike said media outlets. It’s not proclaimed to be an infallable theory or magic bullet here, but rather a solid measuring stick.

bells

11 years ago

Reply to vincent

That’s how I’ve come to think of statistics in general (not just in baseball). I’ve sort of had an ingrained distrust of stats due to the perception that they can be easily manipulated. And they can, but now that I understand them better, I see them as a great way to summarize trends and start discussions around what those numbers mean. For something like baseball, stats are a really great conversation starter, or a way to frame a narrative around a player, era, comparison of players, whatever. Point taken by mosc that WAR is funny because the goalposts shift, and… Read more »

Paul E

11 years ago

Reply to mosc

Voomo Z:
Supposedly, a team of replacement guys plays at a .294 clip or roughly 48 wins. I have to believe, and I’m probably wrong, but the average team wins 81 games. So, apparently, or theoretically, there should be an average of 33 wins above replacement to be shared throughout the league – even adjusting for outliers since the sample size should eliminate them, no?

I’m probably dead wrong on all of this, but, hey, it almost makes sense to me this way 🙁

bells

11 years ago

Reply to Paul E

That makes sense to me, Paul. The average wins has to be 81, by definition.

The Arizona Diamondbacks were 81-81 this year; a quick sum of their WAR shows 30.2. Not a perfect correlation.

Atlanta had 96 wins, should have WAR of 48, has WAR of 45.5.

Houston had 51 wins, should have a WAR of 3, has a WAR of 8.3.

Miami had 62 wins, should have a WAR of 14, has a WAR of 18.

Hmm, not exactly perfect from an initial selective sampling; the lower guys are higher than expected, the higher guys lower.

bells

11 years ago

Reply to Paul E

addendum: quick look at wikipedia suggests that WAR team wins at a .320 clip. So that would be 52 wins, bringing Arizona and Atlanta pretty close, but inflating Houston and Miami’s expectations even more. Should add, though, that after looking a couple of places the correlation seems to be pretty strong historically in terms of WAR predicting wins. Where did you hear this .294 value? Has this changed recently, possibly causing yours or wiki’s data to be outdated? Hmm, I’m teaching a stats class right now and have statistical programs at my fingertips. Maybe I’ll start doing some comprehensive comparisons… Read more »

Paul E

11 years ago

Reply to Paul E

Bells:
Baseball Reference has a total of 1004.8 total WAR for pitchers and batters in 2013. Divided by the 30 teams equals 33.49 wins above replacement. Coincidentally, .294 times 162 games equals 47.63 replacement level wins. 33.49 WAR + 47.63 replacement level wins totals 81.12 wins.
If this is how it really works, I’ll be pleasantly surprised

Voomo Zanzibar

11 years ago

Reply to Paul E

@35, bells,

About halfway down the page here:

http://www.baseball-reference.com/about/war_explained.shtml

… you’ll see that they changed it to .294 in March of this year. Part of the meeting in the middle they did with Fangraphs.

11 years ago

Reply to Paul E

Bells – The change from .320 to .294 was part of the agreement between Baseball Reference and Fangraphs before the start of the 2013 season.

And yes, there were supposed to be 1,000 WAR in 2013. I assume the extra 4.8 comes from rounding at the individual player level.

More here (look under Version 2.2, March 2013)

http://www.baseball-reference.com/about/war_explained.shtml

Artie Z.

11 years ago

Reply to Paul E

bells @34 – I would guess that WAR would not perfectly correlate to wins because WAR is a runs based measure (Rbat, Rfield, etc.), and not a measure based on wins like Win Shares. And while runs scored and runs allowed by a team correlate fairly well to wins and losses, it’s not a perfect correlation. It may correlate better to Pythagorean expected W-L, although still there are issues because some times teams just win a lot of close games or have some games that really skew their overall amount of runs (either allowed or scored). Someone more knowledgeable can… Read more »

Author

birtelcom

11 years ago

Reply to Paul E

Artie Z @42, that is quite right, I believe. At the individual team level, B-Ref’s WAR, when you add up all the WAR generated for the team it will generally match quite closely the team’s Pythag Expectation wins rather than its actual wins. Whatever it is that has a team differ from its pythag expectation (luck? timing? Higgs particles?) can’t really be assigned with any reliability to individual players, so WAR leaves that difference out. (In his Win Shares system, Bill James dealt with this problem by taking the difference between a team’s pythag expectation wins and its actual wins… Read more »

bells

11 years ago

Reply to Paul E

yes, yes, I remembered that b-ref and fangraphs had consolidated their WAR numbers last year, after I had posted. Wiki is out of date, and the first article I googled on b-ref was from 2010 and said they used .320. Not obvious to the casual observer. Regardless, I did a basic analysis of the 2013 teams, and I got 1001.1 WAR total (why is it different from b-ref – rounding error? data entry error on my part? different accounting for the fact that Tampa and Texas played 163 games?) Expected WAR, I calculated at 1002.16, so that’s very close, and… Read more »

bells

11 years ago

Reply to Paul E

Also – does anyone know how to find batting + pitching team WAR summarized? I was going to each team’s individual page and adding the 2 numbers up manually, which of course is a lot more time consuming than copying it from a single table. One thing that might be interesting is to see what teams differed most in accumulated WAR and actual wins above expected replacement level. The teams that had the most WAR MORE than their (wins-expected replacement wins), according to my calculations, were: Detroit – 7.8 Chicago Cubs – 7.7 Colorado – 7.2 White Sox – 7.1… Read more »

Richard Chester

11 years ago

Reply to Paul E

@46 bells: There is an easier way.
Go to BR and click on:
Seasons
All ML
2013
Batting Player Value
Copy and paste the top chart into an Excel spreadsheet
Go back to BR and click on:
Pitching Player Values
Copy and paste the top chart onto the Excel spreadsheet alongside the Batting Player Value spreadsheet.
Create a column that adds the batting WAR and the pitching WAR columns. That is the total team WAR value. Create a column with the teams actual wins and with replacement team wins (.294 x 162). Then do your calcs for the differences.

bells

11 years ago

Reply to Paul E

Richard! Thanks for the help, belatedly. I guess I’ve never really tooled around b-ref other than general leaderboards and player info – I tend to ignore all the top tabs. I’ve been looking more thoroughly the past few days and it really is an incredible database. It’s also helped me understand how to shortcut ways of copying/pasting info.

Paul E

11 years ago

Regarding Mosc’s issues with WAR, I’ve attached the “non-subscribers version” of a PI search for NL OF’er WAR 2013. You know, a little upside-down: 9 Jayson Werth 4.8 8 Carlos Gonzalez 4.9 7 Marlon Byrd 5.0 6 Yasiel Puig 5.0 5 Jay Bruce 5.1 4 Starling Marte 5.4 3 Gerardo Parra 6.1 2 Andrew McCutchen 8.2 1 Carlos Gomez 8.4 One thing is apparent, WAR digs OF’ers (and infielders, for that matter) who can pick it. I realize everyone believes WAR to be the answer, however, come on – Marte and Parra are “nice” ballplayers, but really not the 3rd… Read more »

Joe

11 years ago

Reply to Paul E

“If you bat four times in a game and only three (catchable) balls are hit in your direction, how can the empty production of an 0 for 4 be outweighed by a nice catch (Gomez, Marte, Parra)?”

You don’t think saving an extra base hit in the field can be as valuable as getting an extra base hit at the plate?

11 years ago

Reply to Paul E

I think we’re talking about many different things now, but to address some individual issues: Parra and Gomez’s Rfield numbers were off the chart this year. It’s eminently possible their defense is being overvalued, but it’s important to note that the huge boost they got this season from their defense is not typical. Further, Marte and Gomez had fine years with the stick; talk of empty production is a mischaracterisation of their batting abilities. As for Sheffield vs Parra, I don’t think WAR disagrees with you. Sheffield put up 60+ WAR in a long career. Prior to this season, Parra… Read more »

11 years ago

Reply to Paul E

One other thing: bWAR has Puig’s baserunning at -2, so it definitely doesn’t think highly of him. The worst Rbaser season of all time is -8, done by guys whose CS totals were more than twice their SB figures.

Editor

John Autin

11 years ago

Reply to Paul E

I’ll just say that most critics of WAR hold it to a far higher standard than they use for more conventional stats. Doc Cramer had a .296 career batting average, with over 2,700 hits — both a little better than Gary Sheffield, or Tim Raines. Was he in their league as an offensive player? For sure, WAR should not be treated as gospel, any more than hits or BA. But knocking it for comparative results that you just “know” are wrong, doesn’t convince me. Parra’s range factor last year was about 1/4 of a play per game above average; that’s… Read more »

Luis Gomez

11 years ago

Reply to John Autin

John, sorry for being off-topic here, but the other day you mentioned something about the Mexican Winter League and I thought this might be an interesting read.

http://www.baseball-reference.com/bullpen/Mexican_Pacific_League

John Nacca

11 years ago

Reply to John Autin

Mark Belanger, who had a career OPS+ of 68, earned 41.1 WAR in his career (2016 games). Jimmy Rollins, who has currently a career OPS+ of 96, has earned 42.0 WAR in his career (1952 games). Which player has/had the better career?

Paul E

11 years ago

Reply to John Nacca

John:
Assuming that’s not a rhetorical question, I’ll take the decent SS who hit with some power, stole some bases, and fielded better than adequately over the guy with the great glove.
I just believe the credit guys are given for fielding well at the bottom of the defensive spectrum positions is overstated…as is the penalty for mediocrity at those positions. You know:

Wes Parker, Keith Hernandez or Dick Allen?
Hornsby or Mazeroski (a little extreme)?
Jeter or Andrelton Simmons or even Trammell?

Per WAR, Sheffield had only two seasons better than Parra’s 2013?

Editor

John Autin

11 years ago

Reply to Paul E

“…Keith Hernandez or Dick Allen?” — But what is the question, exactly? If the question is, can I believe that dWAR gives a reasonable estimate of their defensive contributions — Hernandez about 16 dWAR ahead of Allen, or about 1 WAR year — I say yes. On what basis should I say no? If the question is, can I believe that their career WAR rates are a reasonable estimate of their all-around contributions — per 650 PAs, Allen averaged 5.2 WAR, Hernandez 4.6 — I say yes again. I think Allen was a little better player, per year, but Keith… Read more »

Paul E

11 years ago

Reply to John Autin

John:
If a ballplayer hits a double, there is no doubt – it’s a double. If a RF’er runs into the corner to make a nice catch, it’s a “judgment call” by the “eye test” to say that catch saved two bases (or two runs) and the average guy wouldn’t have caught it. I just don’t believe we should put a great level of trust into that kind of judgment.

brp

11 years ago

Reply to Paul E

Except for doubles caused by fluky bounces. Or doubles that could be scored single plus error. Or base hits that could be called errors. Or doubles off the green monster that might be HR at other parks. Or the fielder slipped or fell down while trying to grab the ball. Or the batter beat the throw to second because it was hit to a noodle-arm like Johnny Damon.

But yeah, a double is a double…

no statistician but

11 years ago

Reply to brp

In the record book it’s a double, and I think that’s what’s at issue. What would or wouldn’t have happened re the hypothetical replacement player is based on assumption.

no statistician but

11 years ago

Reply to John Autin

JA: In contrast, the true believers in WAR often hold it to a far lower standard than they use for more conventional stats, or remain silent concerning its imperfections, or equivocate with the ever unanswerable argument that well, it’s self adjusting and getting more and more accurate, so what’s to complain about. I, like you, think WAR is a useful tool, but—and here’s the crux of the issue, as Paul E suggests at #17—the foundation of WAR and many of its most important structural components are assumptions. These assumptions most often have a mathematical element that gives them the legitimacy… Read more »

11 years ago

Reply to no statistician but

Whether you’re pro, anti or neutral WAR, Dave Cameron of Fangraphs and Sean Foreman of Baseball Reference wrote interesting articles earlier this year re: WAR.

Dave’s article is a response to a column by Jim Caple and focuses on the strengths and limitations of WAR and what WAR should be used for:

http://www.fangraphs.com/blogs/what-war-is-good-for/

Sean’s article focuses on the complaints that WAR is constantly changing, there’s more than one WAR, etc., making a comparison to GDP:

http://www.sports-reference.com/blog/2013/03/my-answer-to-i-dont-like-how-complicated-war-is-and-how-it-is-constantly-changing/

Both, in my opinion, are worth reading.

Richard Chester

11 years ago

Reply to Ed

Thank you Ed for those links, they are worth reading.

koma

11 years ago

He is no more a rookie, but amassed “some” WAR until 21 years of age:

http://bbref.com/pi/shareit/hf1k8

pretty nice group, huh?

koma

11 years ago

Ok, he is AL, but still pretty impressive.

mosc

11 years ago

I seem to have started an entirely non-rookie based discussion in this thread! Sorry birtelcom! Frankly though I find discussing WAR far more interesting than this year’s crop of rookies. 1) I agree wholeheartedly with JA. WAR is so much better than any other stat for talking about total value that it’s not even close. It compensates for ballparks, eras, workloads, injuries, so many things. But with that statement of assigning total value to a number comes a whole bunch of “what about this?” as well. Sorry WAR, you are too close to correct. You won’t get much press saying… Read more »

mosc

11 years ago

Reply to mosc

That, and post season stats matter. A lot. WAR thinks I’m crazy.

11 years ago

On an unrelated note…

I’m a bit behind of my television viewing so apologies to those who might have already seen this, but there was a great sight gag on a recent episode of ‘Parks & Recreation’: a fictional law firm by the name of ‘Babip, Pecota, Vorp and Eckstein’.

http://mlb.si.com/2013/10/25/david-eckstein-parks-and-recreation/

Paul E

11 years ago

JA @ #18: Allen averaged 130 runs created per 162 games or 27 more runs created than Hernandez in a “neutral” environment. I have serious doubt that a 1B of the most Golden Glove (Goldest Glove?)variety could make that up at such a bottom of the spectrum defensive position. Hernandez, apparently, didn’t – no matter how big a fan of Seinfeld we were…..and, in the last 45 years, by the eye test, Hernandez was the best fielder I saw at the position. Twenty seven runs is a tall order ( 1 per week). Even if you believe it’s possible, we… Read more »

Voomo Zanzibar

11 years ago

Continuing the WAR /average question down here…

My specific question is how to calculate the average ‘player’ WAR.
Will an average player give you 2.0 WAR/162? More? Less?

Is there a handy way to see that, yearly?

The purpose of the stat is to compare to the “replacement” player.
But does the quality of the “average” player change from year to year?

Is WAR telling us that Rico Petrocelli’s 10.0 in 1969 is exactly as valuable (compared to average) as Rogers Hornsby’s 10.0 in 1922 ?
Or is it merely expressing a contrast to replacement level?

Voomo Zanzibar

11 years ago

Reply to Voomo Zanzibar

Oh, wait a minute, I just figured it out.

WAA

Just calculate WAR – WAA = average player…

Voomo Zanzibar

11 years ago

Reply to Voomo Zanzibar

It looks like average WAR for a full-time player has historically ranged 2.0 – 2.5

Author

birtelcom

11 years ago

Reply to Voomo Zanzibar

And, indeed, when you look at the top of the WAR tables at b-ref it has a guide that suggests that 2+ is (generally) starter level, with 5+ being All-Star-type level and 8+ being MVP-type level.

Paul

10 years ago

As long as it contains defense, WAR is meaningless. Mark Grace (as indicated in the article here) has a -5.6 career dWar. He has two times as many years in the red as in the black, with none of the good years being very good. I watched Mark Grace play. Every game, every season. No one has played a better or more consistent first base than Mark Grace. Some have played as good a first base, but no one’s done better. He was an excellent defensive player – the one to whom all Cubs fans compare any other first baseman… Read more »

Lawrence Azrin

10 years ago

@49/Paul, Grace is actually rated by B-R as one of the BEST defensive 1Bmen ever; I think you have misinterpreated the components of WAR. Grace has a career Rfield of +76, the 6th best ever for 1Bmen. However, when you subtract his positional adjustment of -119, he does end with -(5.6)dWAR. That’s hardly a damning indictment of Grace’s defensive ability; Keith Hernandez, often considered the best defensively 1Bman ever (as he is by B-R), has a mere 0.6 dWAR, despite his +117Rfield. I wouldn’t say WAR is close to meaningless, just that it’s not going to evaluate every player totally… Read more »