Tuesday, July 17, 2012
The 1935 National League MVP
Who was the most valuable player in the National League in 1935?
If you are a student of baseball history, maybe you know the official answer to that question. If you're interested in sabermetrics, there's a good chance you have an answer off the top of your head. If both apply, it's unlikely that you'd name the same person to answer both questions.
But in any case, a number of players had great seasons in the National League that year. Just to set the context, teams averaged 4.71 runs per game, with a league-wide ERA of 4.02 (yes, I know it's a big difference). The eight teams in the league collectively batted .277/.331/.391, so while this wasn't the offensive bonanza many fans associate with the 1930s, it was hardly Deadball. Seventeen players hit .300, and four of them hit over .340. On the other hand, only two players hit more than 25 home runs, and no one hit as many as 35. Only five players had over 100 RBI, and only nine scored 100 runs. Perhaps most surprisingly, that year only one National Leaguer stole more than 20 bases.
What makes the 1935 National League MVP race so interesting? Well, you have Hall of Famers in or near their primes. You have very good non-HOF players having career years. You have great pitchers. And you have a curious MVP choice. Here are some of the players who didn't win the MVP Award.
Dolph Camilli in 1935 had probably the worst full season of his career. He posted a slash line of .261/.336/.440 (in the hitter's paradise of the Baker Bowl), was a subpar first baseman, and struck out more than anyone else in the majors (113), at that time the second-worst total in NL history. So why even mention him? Because Camilli got the same number of MVP votes as Paul Waner. Big Poison hit .321 and walked almost three times as often (61) as he struck out (22). He ranked among the top 10 in the National League in BB, OBP (.392), OPS (.869), triples (12), and runs scored (98). Waner received just one MVP vote, finishing in a tie for 24th with, among others, Camilli.
I don't think anyone would suggest Waner should have won the NL MVP Award in 1935, but this is a Hall of Fame outfielder having a good season, and all he gets is one 10th-place vote? The same as Dolph Camilli? In a just world, Waner probably should have finished in the top 10 or 15. (Camilli, of course, went on to become a very good player. He eventually won the 1941 NL MVP Award, and was an entirely reasonable selection. In '35, he was barely average.)
Also in 1935, Cardinals first baseman Ripper Collins batted .313, scoring 109 runs and driving in 122. He was one of only four players to both score and drive in over 100 runs, with the second-highest combined total (231) in the league. Collins was fifth in the NL in extra-base hits, and gained 306 total bases. He also excelled in less celebrated statistics that we recognize today as important. He walked 65 times, which might not sound like a lot, but in the NL in '35, that tied for sixth in the league. Collins grounded into only 5 double plays all season, in 578 at-bats — less than half as often as the NL average. Collins received 3 points in the NL MVP voting and tied for 21st.
Joining Collins on the short list of players who scored and drove in over 100 runs was New York left fielder Hank Leiber. He had 203 hits, led the Giants in doubles (37), ranked sixth in the league in HR (22), and posted an OPS of .901 (not that anyone was computing OPS then). Leiber gained over 300 total bases, and came in 11th in MVP voting.
Boston's Wally Berger led the NL in HR (34) and RBI (130). He had the second-most XBH (77) and third-most TB (323). Berger batted .295/.355/.548, and might have won the MVP Award after leading the league in two of the three Triple Crown stats, but the Braves lost 115 games that season. He could have driven in 230 runs and they wouldn't have given him the MVP. Though I suppose if he'd driven in 230 the Braves wouldn't have finished last. Anyhow, it says something about the respect for RBI that Berger did show well in the voting, tied for sixth.
The interesting question, I think, is how the Braves would have finished without Berger. Both Baseball Reference and FanGraphs rate Berger as about 5.5 Wins Above Replacement, so we can estimate that the team might have fallen to 33-120 (.216), seven games worse than the 1962 New York Mets (40-120, .250). From a certain point of view, Berger might have really been the most valuable player in the league, almost single-handedly keeping Boston from ranking as perhaps the worst team in history. The '35 Braves also had Babe Ruth in his final season. Ruth only played in 28 games, and he batted .181, but he still had power (.431 SLG, 250 ISO) and would still take a walk (.359 OBP).
I've mentioned players who had very good seasons, and several of them received surprisingly little support in the MVP voting, but Berger is the first one I've mentioned who, in many seasons, might have deserved to win the MVP based on his performance. The same applies to Chicago Cubs left fielder Augie Galan, who in '35 led the NL in stolen bases (22) and runs scored (133). He had 203 hits and walked 34 times more than he struck out. Galan had over 300 total bases and set a major league record which still stands, most at-bats (646) without grounding into a double play — 0 GIDP in 646 AB. As an outfielder, he threw out 12 runners, including 4 double plays, and his fielding percentage was above league average.
Unless you totally disrespect leadoff men, that's an MVP-quality season. Two hundred hits, 300 TB, led the league in two major offensive categories, fielded well, set a major league record, and played on a pennant-winning team. Dozens of guys have won the MVP with worse seasons than that. Galan was third on his own team in MVP voting, ninth overall.
One of the teammates he ranked behind was Billy Herman. In addition to superb defense at second base, Herman led the majors in hits (227), doubles (57), and sac hits (24). Talk about an old-school manager's dream: brilliant middle infield defense, great batting average (.341), lots of sacrifices, very low strikeout rate (3.9%). If neither Herman nor Galan was the most valuable player on the Cubs that season, it's no wonder they won 100 games. Herman was fourth in MVP voting.
What a season from Joe Medwick. He finished second in hits (224), runs (132), RBI (126), and slugging (.576), and no player bested him in more than one of those categories. Herman had three more hits, but Medwick had a higher batting average, and Medwick hit with power; he had more than three times as many HR (23) as Herman (7). Galan scored one more run than Medwick, but Galan was a leadoff man, and he only had 79 RBI, nearly 50 fewer than Medwick. Berger drove in four more runs, but Medwick hit 58 points higher, got 50 more hits, and out-slugged the NL home run leader by almost 30 points. Arky Vaughan slugged higher, but Medwick hit more singles, more doubles, more triples, and more home runs than Vaughan. He scored more runs and drove more in.
In addition to H, R, RBI, and SLG, Medwick also finished second in batting average (.353) and doubles (46). He ranked third in triples (13) and fourth in home runs (23). He led the league in XBH (82) and led by a ton in total bases (365), more than 10% ahead of second place (329). He wasn't an error-prone fielder, either. Modern defensive statistics show Medwick as one of the best outfielders in the league. He placed 5th in MVP voting.
Of course, we can't have a discussion about NL MVP snubs without mentioning Mel Ott, widely regarded as the greatest player never to win an MVP Award. In 1935, Ott scored 113 runs and drove in 114. He ranked 2nd in the NL in home runs (31), total bases (329), and on-base percentage (.407). He ranked third in bases on balls (82), extra-base hits (70), and slugging average (.555), and fourth in runs and RBI. Ott finished the season with just 4 GIDP in 593 at-bats — 0.67% of the time, and less than 1/3 of league average (2.09%). His low GIDP rate yielded 8-9 extra times on base and is roughly equivalent to 14 points of batting average, for someone who was already at .322. Ott was also one of the league's finest outfielders, with 17 assists, 7 double plays, and a .990 fielding percentage.
So where did the voters leave Ott in '35? Second? Third? Heaven forbid, eighth or ninth or something? Ott finished 20th in NL MVP voting. And please don't imagine that Ott and Medwick were punished for the poor performances of their teams. Ott's Giants won 91 games. The Cardinals won 96. Medwick didn't lead his own team in MVP voting, and Ott actually ranked 5th in the MVP voting among his own team.
One of the teammates Ott received fewer votes than was catcher Gus Mancuso. Ott's batting line was .322/.407 /.555. Mancuso's was .298/.342/.380. Ott finished the season with 329 total bases, Mancuso with 170. Ott scored and drove in a combined total of 227 runs; Mancuso scored and drove in a combined total of 89. Mancuso was regarded as superb at handling pitchers, but to be more valuable than Mel Ott, you'd have to turn everyone on the staff into Lefty Grove. It boggles the mind that a superstar like Ott bats .322, hits the second-most HR in the league, and gets a total of three points in the MVP voting.
That same season, Pittsburgh shortstop Arky Vaughan led the NL in batting average (.385), on-base percentage (.491), and slugging (.607). Vaughan in '35 and Chuck Klein in '33 were the only National Leaguers to win the "Slash Stat" Triple Crown between Sherry Magee (1910) and Stan Musial (1943). Not that OBP and slugging were recognized at the time, but you're obviously talking about exceptional performances. Vaughan's OPS of 1.098 led the league by 136 points, miles ahead of the second-place tie between Ott and Medwick (both .962).
Vaughan in '35 led the NL in walks, actually had more than five times as many walks (97) as strikeouts (18), and led the league in times on base (296). We all know that getting on base was undervalued in 1935 — hell, it's undervalued today — but .491 OBP? That's a Ted Williams season. Hey, maybe it's fitting that Vaughan didn't win the MVP.
But Vaughan, a shortstop, was much more valuable than Ted Williams in the field. He wasn't Ozzie Smith, but he was playing a demanding defensive position. If your shortstop is the best hitter in the league, you're going to win a lot of games. Unless you're the 2002 Texas Rangers or something. Anyway, Vaughan led the league in every major offensive efficiency stat and gained 303 total bases. He placed third in the 1935 NL MVP vote.
In addition to the tremendous seasons turned in by Berger and Galan, Herman and Medwick, Ott and Vaughan, several pitchers had pretty great seasons. Brooklyn's Van Mungo went 16-10 for a team that was 54-73 in its other games. He tied for the league lead in shutouts (4), and led the NL in K/9 by almost three-quarters of a strikeout (6.00). He didn't get a single vote in the MVP selection, despite that his name really was Van Mungo.
Reds pitcher Paul Derringer was 22-13 (.623) for a team that was 46-72 (.390) in its other games. He was third in the NL in innings pitched (276 2/3) and in wins. Derringer's 3.51 ERA ranked a respectable 15th, but his 3.37 FIP (Fielding-Independent Pitching) ranked an impressive 4th, and he finished among the league's top 10 in strikeouts. He tied Mancuso, the New York catcher, for 17th in MVP voting.
Phillies reliever Syl Johnson overcame the Baker Bowl, a tremendous hitter's park, to post the second-best K/BB ratio in the National League, 2.87-to-1. He's listed with 18 starts and 19 relief appearances (unthinkable today), and credited with a 10-8 record and 6 saves. He led the team in ERA (3.56) and easily in WHIP (1.22). Those numbers probably seem pedestrian, but the Baker Bowl in the '30s was comparable to Coors Field in the '90s. In away games, Johnson had a 3.21 ERA, 1.14 WHIP, and 3.89 K/BB ratio. That would have ranked him ninth in ERA, third in WHIP, and first in K/BB. On the road, Johnson was one of the best pitchers in the majors, but his home field was so brutal for pitchers that his overall stats are not impressive.
Johnson didn't get any MVP votes, which is not entirely surprising, but for reasons largely beyond my understanding, Johnson's teammate Curt Davis tied for 21st in the voting and shows in modern sabermetric stats as one of the best players in the NL, at any position. Davis was 16-14 with a 3.66 ERA and 74 strikeouts (2.88 K/9, 1.57 K/BB ratio). The Neutralized Pitching tool at Baseball Reference suggests that in an average NL park, Davis' ERA would have been 3.26. That seems totally plausible, and with his 231 IP, it certainly makes him a valuable pitcher; I just don't understand how it makes him as valuable as Dizzy Dean.
Pittsburgh's Cy Blanton had a tremendous 1935 season. He led the league in ERA (2.58) and WHIP (1.08), was third in strikeouts per inning, and was the best in the league at preventing home runs (0.11 per 9 innings). Blanton pitched 23 complete games (third-best) and hurled four shutouts (tied for the league lead). His 2.77 FIP led the NL by half a run, and sabermetric statistics suggest that he was the best pitcher in the league. Blanton went 18-13, though, and tied for 15th in the MVP voting.
Four Chicago Cubs pitchers won 15 games in 1935. Charlie Root was 15-8 with a 1.19 WHIP. Bill Lee went 20-6 with a 2.96 ERA. Lon Warneke went 20-13 with a 1.17 WHIP, and led the team in innings (261 2/3) and strikeouts (120). Southpaw Larry French was 17-10 with a 2.96 ERA. He pitched four shutouts, tied for the most in the league, and didn't throw a wild pitch all season.
Lee got one 10th-place vote, tying Waner, Camilli, and Giants third baseman Travis Jackson for 24th place. Warneke ranked 12th. Neither Root nor French got a single vote. For what it's worth, advanced statistics suggest that French was the most effective pitcher on the staff, with a team-leading 3.41 FIP and 133 ERA+ (the latter tied with Lee).
The Giants had a pair of great pitchers, as well. Hal Schumacher went 19-9, ranking 3rd in the NL in both ERA (2.89) and WHIP (1.17). Carl Hubbell ranked second in the NL in complete games (24), innings pitched (302 2/3), and strikeouts (150). He finished 23-12 and led the league in K/BB ratio (3.06). Hubbell tied for sixth in the MVP voting and Schumacher didn't get a vote.
The most famous tandem, of course, was the Dean brothers in St. Louis. Paul Dean went 19-12 and ranked among the league's top five in innings, strikeouts, and WHIP. Dizzy Dean went 28-12, and led the league in numerous celebrated statistics: wins (28), complete games (29), innings (325 1/3), and strikeouts (190). He even saved 5 games. Dizz ranked 6th in ERA and tied with Bill Swift of the Pirates for the league's second-best FIP (3.27). Dizzy placed second in MVP voting. Paul tied for 21st.
According to rWAR, Baseball-Reference.com's Wins Above Replacement, the 10 best players in the NL in 1935 were:
1. Arky Vaughan, PIT (.385 BA, 1.098 OPS)
2. Mel Ott, NYG (.407 OBP, .555 SLG)
t3. Curt Davis, PHI (231 IP, hitter's park)
t3. Dizzy Dean, STL (28-12, 190 K)
t3. Billy Herman, CHI (227 H, 57 2B, defense)
6. Cy Blanton, PIT (2.58 ERA, 1.08 WHIP)
7. Joe Medwick, STL (.353 BA, 132 R, 126 RBI)
8. Wally Berger, BOS (130 RBI, .548 SLG)
t9. Augie Galan, CHI (133 R, 0 GIDP)
t9. Syl Johnson, PHI (10-8, 3.56 ERA)
Modern statistical analysis shows Vaughan as easily the best player in the National League. Bill James has written extensively about Vaughan's season. Earlier, I wrote that Ducky Medwick "hit more singles, more doubles, more triples, and more home runs than Vaughan. He scored more runs and drove more in." That's all true. So how could Vaughan be more valuable?
The biggest differences are times on base and outs used. Vaughan walked 97 times, most in the NL, and got hit by 7 pitches. Medwick walked 30 times with 4 HBP. Vaughan grounded into 5 double plays, compared to 15 for Medwick. Just looking at BB, HBP, and GIDP, that's 80 extra times on base for Vaughan. Medwick had 224 hits in 634 at-bats, meaning he made 410 outs at the plate. Vaughan got 192 hits in 499 at-bats, giving him 307 outs. So Vaughan has 80 more times on base and 103 fewer outs. That's easily worth 62 total bases.
I never saw any of these guys play. I wasn't even alive in 1935. But it seems to me that at least half a dozen players had MVP-quality seasons, exceptional in different ways. Galan probably had one of the best seasons ever by a leadoff man. Berger was the league's best power hitter, the only good player on a truly terrible team. Medwick was the most prolific hitter, with a great average and good power. Blanton was the league's most efficient pitcher, Dizzy Dean an elite workhorse. Herman led the league in hits and was perhaps the most valuable fielder in baseball. Ott did everything well. And no batter was more efficient than Vaughan, whose season was so exceptional it can be regarded as historic.
The 1935 NL MVP Award was voted to Chicago catcher Gabby Hartnett. He was a good hitter (.344/.404/.545), a good fielder (MLB-leading 61% CS), and good handler of pitchers (Root, Lee, Warneke, and French all had good years). That's a very nice season, but Hartnett only played 116 games.
Hartnett would have ranked third in the league in BA, fourth in OBP, and fifth in SLG. That's terrific, especially for a good defensive catcher. But Hartnett didn't actually have enough plate appearances (461) to qualify for the leaderboards. He ranked ninth in the NL in RBI (91), didn't make the top 10 in any other statistic. Hartnett in '35 gained 225 TB, actually the second-highest total of his career. But that's outside the top 20 in the NL, almost 100 TB behind Galan, Vaughan, Herman, and Berger. It's more than 100 behind Ott, and almost 150 behind Medwick.
None of those guys were inefficient on offense; they all ranked in the top 10 in OPS and OPS+. And none of them were poor defensive players. Galan, Medwick, and Berger were good defensive outfielders. Ott was a terrific defensive outfielder. Vaughan was a decent shortstop. Herman was a sensational second baseman. Was Hartnett's 3/4 of a season so extraordinary that he should rank ahead of the NL's other standouts? The Cubs went 78-38 (.672) with Hartnett in the lineup, compared to 22-16 (.579) without him. Certainly that's a big difference.
My intention here is not to prove that Hartnett was a bad choice as MVP. It's just to highlight some great seasons, superb in different ways. I'm confident that Vaughan was the best player in the league, but beyond that, it's guessing. I mean, I'm three-quarters of a century removed from this season, and there's only so much you can get from analyzing statistics and reading about the players. But since we've gotten this far, here's my own MVP ballot, 77 years late:
1. Arky Vaughan, SS, PIT
2. Mel Ott, RF, NYG
3. Dizzy Dean, RHP, STL
4. Billy Herman, 2B, CHI
5. Joe Medwick, LF, STL
6. Cy Blanton, RHP, PIT
7. Augie Galan, LF, CHI
8. Wally Berger, CF, BOS
9. Gabby Hartnett, C, CHI
10. Carl Hubbell, LHP, NYG