The Hardball Times -- Brandon Isleib

Some things literally never change

Brandon Isleib — 2011-05-04T20:22:15+00:00

Click here to learn about THT's download subscriptions.

Weird history: 1987 San Jose Bees

Brandon Isleib — 2010-08-27T09:25:15+00:00

Click here to learn about THT's download subscriptions.

Kendrick v. Hawksworth and the Underwhelmonster

Brandon Isleib — 2010-07-20T16:20:15+00:00

Click here to learn about THT's download subscriptions.

PIT-thagoras

Brandon Isleib — 2010-05-08T15:04:15+00:00

Click here to learn about THT's download subscriptions.

Does this happen often?

Brandon Isleib — 2010-04-03T13:34:15+00:00

Click here to learn about THT's download subscriptions.

Report cards (Part 2)

Brandon Isleib — 2010-02-25T10:30:15+00:00

Last time, I put up the framework for several studies based on giving letter grades to team fielding, pitching, and batting components. Now it's time to take a broad look at one of those time-honored, rarely tested adages: Do pitching and defense win championships?

In distill of the night

There are a couple of questions wrapped up in the adage. It's relevant to ask whether they win ballgames in the first place, whether one can win without the other, and whether their winning ballgames is in different proportion to their winning championships.

There are 320 playoff teams and 99 world championships through 2005, where my edition of the ESPN Baseball Encyclopedia cuts off. The following chart shows the 320 teams sorted by frequency of grade per component (top to bottom is A to F):

So right there we can see a couple pertinent items. While a variety of skill sets can make the playoffs, if your team has a D or F it's going to be awfully hard to get there. If your available talent forces you to have a weakness, make it hitting, sure; but that's not the same as saying pitching and defense are the important things.

Now take a look at the frequency of world championship teams by grade:

Although the graph by itself says that fielding is the most indicative of championships (after all, 64 A fielders produced 31 winners), that's misleading due to the slight uptick in A fielders from the days before multiple-tiered playoff systems, due to a game that used to revolve more around its fielders.

The D/F playoff teams are remarkably absent from the world champions, save for offense. It's probably those weird teams that have propped up the adage far more than is deserved. Examples of champions with weak offenses include the 1906 White Sox (AAD), the 1969 Mets (BAF; the '73 Mets were also BAF, and they're the only two F-offense playoff teams), the 1995 Braves (BAD), and the 2005 White Sox (BAD, although their win is muted by the Angels and Astros also profiling as BAD).

What's probably more accurate to say, then, is that pitching and defense occasionally enable weak offenses to win championships. But even that doesn't paint a very realistic picture. Don't even think about a championship if you have D or F pitching. The lone exception is the 1913 Athletics, whose AFA profile is the only F pitching to make the playoffs. The D pitching came closest to winning a championship with the 1982 Brewers (CDA), but the rest of the teams mostly are bad modern entries, like the 1997 Mariners or the 2005 Padres. Realistically, you're going to need A or B pitching with at least C fielding.

But while D offenses sometimes win, the A offenses are most associated with winning, even more than A pitching. 54 A-hitting champions out of 149 teams is 36 percent odds, better than the 32 percent of A pitching (53 of 165). The difference is slight, but in testing our initial adage it's relevant; the mashers don't get into the playoffs as often as the aces, but they're at least equals once they're there. The 1932 Yankees (CCA), 1957 Braves (BCA), and 1976 Reds (ACA) are fine examples of teams winning with crushing offenses and little regard to who's on the mound.

Disjunctivitis

Of course, while we've answered some of the question, we haven't answered whether pitching and defense are linked, e.g. we haven't seen teams with only one or the other. We've seen that pitching and defense can compensate for bad offenses (1995 Braves and friends), but can pitching compensate for bad defense?

As it turns out, not really. Four teams have made the playoffs with F fielding and A pitching: the 1912 Giants (FAA), 1965 Twins (FAA), and 1983/85 Dodgers (FAC/FAB). The 1908 Tigers and 1971 Giants were both FCAs; neither fared well in October. Occasionally the reverse will happen and a team that's only good with the glove will win—the 1987 Twins (ACD) and the 2003 Marlins (ACB) lead the pack there—but that's not a model you want to emulate, especially the former.

It wasn't until 1921 that a team without A fielding won a World Series. Since then, fielding has decreased in importance, but that's not to say a sloppy team will
win; like pitching, fielding has a minimum level of competence for any championship team. That said, once the minimum's in place, it's anybody's game, and if you can meet those minima while having a team of mashers, you may be in the best configuration to win. If I had my druthers (and let's face it, who doesn't want their druthers?), I'd take my chances with a random Joe McCarthy or Sparky Anderson team over the '95 Braves or '05 White Sox, and I think you would too.

Conclusion

So while it's unproven that pitching and defense win championships—indeed, the best link is with offense—bad pitching and defense can sabotage championships in a way that bad hitting doesn't. But there's far more potential to rack up wins with an offense, and if your pitching and defense are serviceable, then the offense is the key to success. You can pull out miracle seasons with bad offense, but why count on that when you could just hit your way to glory?

Click here to learn about THT's download subscriptions.

The culture of debate

Brandon Isleib — 2010-02-21T16:06:15+00:00

Click here to learn about THT's download subscriptions.

Report cards (Part 1)

Brandon Isleib — 2010-02-11T10:40:15+00:00

It's always fascinated me how many ways there are to build a winning team. It's said often that pitching and defense win championships, but that idea is parroted when a team wins a low-scoring game instead of being analyzed from any angle.

Fortunately, the good folks who make the ESPN Baseball Encyclopedia have divided team performances into fielding, pitching, batting, and base-stealing win components. From Page 1383 of the 2006 edition, each of these is the "total number of wins the team achieved through its [fielding/pitching/etc.] compared to the average team in the context of the offensive level of the league and the team's home park(s)." In other words, however they figure it out, they've adjusted the component figures to neutral levels so we can compare teams' strengths and weaknesses across eras. Mind you, these statistics measure the component performances and not quite the actual wins achieved; Pythagoras isn't invited to the party.

What I chose to do with the stats is group the components into grades, as all of them are measured as wins above or below average. For this exercise, these grades accompany wins above or below average like so:

   Fielding/Running    Pitching/Batting
A  +1.5 and up         +6.0 and up       
B  +0.6 to +1.4        +2.1 to +5.9
C  -0.5 to +0.5        -2.0 to +2.0
D  -0.6 to -1.4        -2.1 to -5.9
F  -1.5 and down       -6.0 and down

This breakdown, which simplifies the statistic a little bit while de-emphasizing minute differences, gives a fairly even distribution of A/B/D/F grades with a higher frequency at C. It also gives us 125 fielding/pitching/batting grade profiles (baserunning differences are too small and too dissociated from winning teams to be included regularly) and 13 team GPAs, which allow for a host of interesting comparisons.

I know the above explanation was kinda murky, but it's an intuitive sort of breakdown that will make sense as I milk the data for every worthwhile configuration I can find. All profiles are listed in fielding-pitching-batting order, since that's how the encyclopedia listed them. The figures are from 1901 to 2005, because the latter figure is where my edition ends. That's still 2052 teams to find us some patterns. Let's blow this Popsicle stand and get to analyzing!

The valedictorians

The 4.0 profile is rare, happening 20 times but only three times since 1945. (As strikeouts have increased and playing conditions have standardized, fielding wins are a tinge harder to attain than in days gone by.) The surprising bit is that four of those 20 missed the playoffs. The complete list of AAAs for those so interested:

1902 Pirates
1904 Giants
1905 Giants
1906 Indians (3rd)
1909 Athletics (2nd)
1910 Athletics
1910 Cubs
1912 Red Sox
1915 Red Sox
1915 White Sox (3rd)
1919 Reds
1929 Athletics
1939 Yankees
1942 Yankees
1942 Dodgers (2nd)
1944 Cardinals
1945 Cubs
1969 Orioles
1976 Phillies
2001 Mariners

Although eight of the 16 AAA playoff entrants have won the World Series, it's only been one of the last six who have won. The '42 Yankees lost to the CAA Cardinals (since it's just the fielding that is different, that's roughly an even match); the '45 Cubs lost to the CAC Tigers; the '69 Orioles lost to the BAF Mets (!); the '76 Phillies lost to the ACA Reds (but the best hitting team without Babe Ruth of all time, by ESPN's rankings); and the '01 Mariners beat the CCA Indians but could not beat the CAC Yankees.

Ruthian/Fosterian offense aside, it seems that A pitching can hang well with A pitching regardless of hitting quality. But does that hold up? In later installments we'll uncover this and other oddities. But for right now . . .

The dropouts

36 teams are in the ignominious FFF profile. And unlike the now-rare AAA profile, the FFFs split half-and-half at World War II. The last half of the list has a bunch of teams you'd expect and a couple of surprises:

1945 Phillies
1949 Browns
1949 Cubs
1952 Pirates
1954 Pirates
1954 Athletics
1955 Orioles
1962 Mets
1963 Mets
1963 Senators
1970 White Sox
1972 Rangers
1979 Athletics
1982 Athletics
1993 Rockies
2002 Tigers
2003 Tigers
2004 Diamondbacks

The three big surprises to me are the White Sox, the '82 Athletics, and the Rockies. The A's were too busy stealing bases to get noticed for their futility (they got an A in baserunning), while the Rockies' newness and the Padres' stink covered up the mess, but the White Sox never seem to come up in those "worst of" lists. Their frontline talent was decent, at least for a 56-106 team; Ed Herrmann and Bill Melton were sluggers at catcher and third base, they had a solid front starter in Tommy John, and Wilbur Wood was a workhorse closer.

But their bench was atrocious; two of their hitters with over 100 plate appearances had an OPS+ under 30, a feat achieved by only 28 teams without Bill Bergen on them. It's not like the others were much better. The combined performance of the fifth through ninth bench players: 90 for 505 (.178) with a .255 SLG off 12 doubles and nine homers, walking 36 times and striking out 107 times. Those players combined for a season of stench, and with those hidden seasons on the bench, it's no wonder the team was failing.

Yes, hap is a word. I looked it up and everything.

So we've seen the happed and the hapless. Next time we'll look at some of the in-betweens and see if any of them shed light on the best way to construct a winning team.

Click here to learn about THT's download subscriptions.

Hughie Jennings and Hack Wilson

Brandon Isleib — 2010-01-28T10:40:15+00:00

               W-L   PCT  RS  RA
Start to 5/20 16-13 .552 148 119
5/21 to 7/6   32-12 .727 238 163
7/8 to finish 45-35 .563 471 359

The 1924 Giants won the pennant by a mere 1.5 games, and you can point to that midsection as the biggest reason why they won. But the odd thing is that the middle period is while John McGraw was away from the team. That 32-12 belongs to Hughie Jennings, and while it's a slightly lucky record, it still stands out.

So what was different in that stretch? The Giants started out 11-2 but tanked by the time Jennings took over. The only changes McGraw had made were to move Heinie Groh from second in the order to seventh, moving everyone else up, and to swap Billy Southworth to third and Ross Youngs to leadoff. Groh eventually worked out in the seven-hole, but the Youngs-Southworth swap didn't make much sense. Southworth was the centerpiece of a deal that sent Dave Bancroft to play-manage the Braves, and in his playing days he was sort of a souped-up Randy Winn—solid complementary bat, vague leadoff skills, and a glove, the last being important both to McGraw and to playing center field in the Polo Grounds. But by the time Jennings took over, Southworth's line was .234/.344/.308—not what your third hitter ought to be producing. (Youngs was at .317/.415/.436, so it's not like he was dragging the lineup down.)

When Jennings took over, nothing changed initially. He found more starts for Hugh McQuillan and Mule Watson, which worked great (ERAs of 2.85 and 1.93, respectively), put Rosy Ryan permanently in the bullpen, and over 25.2 innings got a ridiculous 0.71 ERA out of reliever Claude Jonnard. But the move that saved the Giants season was one that would have repercussions on many National League seasons to come: He put rookie Hack Wilson in center field.

To be fair, Jennings didn't do this all on his own; based on usage patterns, it appears to be a response to a Southworth injury that I can't dig up. On June 11, Southworth started his last game of the Jennings period, allowing Wilson to take over center. If nothing else, Jennings knew potent bats, and what Wilson did in his time was striking. In chart form:

               AB  BB  R 2B 3B HR RBI  AVG  OBP  SLG
Start to 5/20   6   1  1  1  0  0   0 .333 .429 .500
5/21 to 7/6   111  11 20  7  8  2  28 .378 .434 .631
7/8 to finish 268  33 41 12  4  8  33 .257 .339 .422

In what was basically a month's worth of starts, and without even a week's worth of major league experience, Wilson had one of the best hitting stretches of his or anyone's career. That rookie month was better in his triple-slash line than any of his full seasons save 1930, and all three marks led the Giants lineup in the interim period (Youngs' .357/.424/.582 was second across the board). When McGraw came back, Wilson struggled mightily, whether for sample size, having a second time around the league, or chafing under McGraw, and with Wilson's 1925 season looking eerily like the last third of 1924 (even down to the same slugging percentage), McGraw sent him to Toledo for Earl Webb.

Under Joe McCarthy the next year, Wilson racked up numbers similar to his Jennings phase, and continued to do so through 1930. This split may say a number of things about Jennings, McGraw, and Wilson. We now know that McGraw did not give Wilson his initial shot, as the fourth outfielder slot prior to May 21 was second-year Jimmy O'Connell's (15 at-bats and a pinch-running appearance), not Wilson's. We also know that McGraw never saw firsthand what Jennings saw, that Wilson could carry a team if you gave him the chance.

Wilson wasn't going to be a great center fielder, and certainly not in the Polo Grounds. When Wilson slumped, McGraw sent him to left field for Irish Meusel and stuck O'Connell and normal first baseman George Kelly in center (letting rookie Bill Terry fill in at first). Given experience levels, player profiles, and subsequent history, my conclusion is that McGraw would have turned to O'Connell and not Wilson when Southworth got hurt. O'Connell started on June 14, so he was available for Jennings's decision making.

O'Connell also had an offensive profile resembling Southworth's, or at least more than Wilson's did, and O'Connell had been one of McGraw's center fielders for 1923. It doesn't seem to be a stretch to say that McGraw would have played O'Connell when Southworth went down and never given Wilson a chance. It was Jennings who gave that chance to Wilson, and it was only Jennings who saw what Wilson was capable of in his Giants stint.

Would the Cubs have Rule 5ed Wilson had his major league resume been pedestrian? I have no idea, but his run of dominance under Jennings may have given the Cubs enough sample to know that his hitting was truly major-league-caliber, and if that's what did it, then it's owed to Jennings and not McGraw. Besides Wilson's Hall of Fame career, the more immediate matter of a narrowly won pennant goes in part to Jennings' decision as well; it's safe to say that slugging .631 for any stretch was not part of the Jimmy O'Connell skill set, and although Southworth's usual stat line was one in the McGraw mold, he wouldn't have been this good either. Of the many things Hughie Jennings could have done after the Southworth injury, he chose the best one, and they won enough games off it to win McGraw's last pennant.

I don't know how many times the interim manager can be credited for something so momentous, but it seems safe to say that installing Hack Wilson was Jennings' choice, not McGraw's, and led to a pennant and a Hall of Fame career. Pretty heady stuff for a few weeks in 1924.

Click here to learn about THT's download subscriptions.

The League Award, 1970-2009

Brandon Isleib — 2010-01-14T10:40:15+00:00

I should have thought a little bit more about breaking up a series over the holidays. In any event, the continuation/conclusion of the League Award series, in which we look at the first eligible winner of MVPs if the AL's rules of the '20s had endured (no repeat winners and no player-managers eligible) is below. 1970 through 2009's in this article, whereas 1931 through 1969 was in the last one. There are 24 different awards this time, but a whopping nine of those involve either a B.L. Bonds or a J.A. Pujols, so the list is skewed toward the last few years. (In the NL of the 2000s, we only can keep three awards; the only other decade-league that low was the 1940s AL. I don't know if that's a sign of league illness, but it doesn't look good. Then again, the AL had no repeat winners from 1970 to 1989, and in the '70s it was viewed as the weaker league. Who let counterpoints into my argument?)

The aforementioned skewed list:

1970 AL Tony Oliva over Boog Powell
1972 NL Billy Williams over Johnny Bench
1976 NL George Foster over Joe Morgan
1977 NL Greg Luzinski over George Foster
1981 NL Andre Dawson over Mike Schmidt
1983 NL Pedro Guerrero over Dale Murphy/Andre Dawson/Mike Schmidt
1986 NL Glenn Davis over Mike Schmidt
1987 NL Ozzie Smith over Andre Dawson
1989 AL Ruben Sierra over Robin Yount
1991 AL Cecil Fielder over Cal Ripken
1992 NL Gary Sheffield over Barry Bonds/Terry Pendleton
1993 NL Lenny Dykstra over Barry Bonds
1994 AL Ken Griffey over Frank Thomas
1997 AL Tino Martinez over Ken Griffey
1998 AL Nomar Garciaparra over Juan Gonzalez
2001 NL Luis Gonzalez over Barry Bonds/Sammy Sosa
2002 NL Albert Pujols over Barry Bonds
2003 NL Jim Thome over Barry Bonds/Albert Pujols/Gary Sheffield
2004 NL Adrian Beltre over Barry Bonds
2005 AL David Ortiz over Alex Rodriguez
2005 NL Andruw Jones over Albert Pujols
2007 AL Magglio Ordonez over Alex Rodriguez
2008 NL Ryan Braun over Albert Pujols/Ryan Howard
2009 NL Hanley Ramirez over Albert Pujols

The NL sandwich atop the list is a series of shell games, with players getting awards early to give them back late (Foster and Dawson are involved in four of the awards one way or the other). That doesn't mean there aren't plenty of great quirks in here ... like these five:

1983 NL: Guerrero over three

Top five eligibles: Guerrero, Tim Raines, Jose Cruz, Dickie Thon, Bill Madlock

If, as Chris Jaffe noted this week, Fred McGriff's numbers are era-bound against him compared with slightly later peers, then those peers are to McGriff what McGriff is to Guerrero. Say what you will about the defense, but in his prime (1982-87) the top five hitters by OPS+ were himself, Mike Schmidt, Don Mattingly, Wade Boggs, and George Brett. (Jack Clark is sixth; wouldn't have guessed.) This was the only full year Guerrero played third base (sometimes you have to test "bat will play anywhere" scouting reports), and despite the failed experiment—range was passable, but hands weren't—he comes in the winner here amid a glut of Dawson, Murphy, and Schmidt seasons.

1986 NL: Davis over Schmidt

Top five eligibles: Davis, Gary Carter, Tim Raines, Kevin Bass, Von Hayes

What a year. The Red Sox lost the World Series but took home the MVP, Cy Young, and Manager of the Year awards, while the Mets won the Series, were 108-54 in the regular season, and got no awards ("Why isn't we getting no medals?"). In this hypothetical, Davis completes what Mike Scott and Hal Lanier started on the Astros award sweep.

Besides that a 108-54 team got locked out of the award voting with or without the hypothetical, the voting in '86 is instructive for two reasons. First, although the Astrodome needs some harsh park adjustments through most of its existence, it didn't harm its hitters' visibility per se; usually it just overrated its pitchers. Second, in the wake of the last Hall of Fame vote, Raines' solid showings in both '83 and '86 indicate that the argument on his candidacy shouldn't be about him being underrated in his prime. Even if a good argument exists there, the first fiddle might have been teammate Dawson rather than Rickey Henderson. Throughout the '80s, Raines drew consistent MVP voting, though generally down-ballot. Henderson's support was stronger in several years but not as consistent; Raines received votes in seven seasons and Henderson in eight (Dawson in nine). I can't conclude that the writers of the time ignored what Raines did.

What they may have ignored was Raines' White Sox career, which, while not as eye-popping as his Montreal days, was productive in its own right. Injuries obscured this somewhat, but in 1993 Raines was easily the most productive batter in that lineup after Frank Thomas, with a 138 OPS+ batting leadoff that was 15th among regulars and the fourth-best mark of his career. Raines posted 18 seasons of a 100 or greater OPS+; this is tied for 26th since 1901. The only Hall-eligible players with 18 or more who are not in are Harold Baines, Dwight Evans, Rusty Staub, Raines and Darrell Evans, and Raines had a much better peak than the others. I'm going to go against the grain and say that it was his White Sox days and not his Expo ones that have been forgotten, and these MVP results corroborate that idea.

Oh yeah, and something or other about Glenn Davis. Sorry, Glenn; I got into a Tim Raines rant.

1992 NL: Sheffield over two

Top five eligibles: Sheffield, Andy Van Slyke, Larry Walker, Darren Daulton, Fred McGriff

In years where Barry Bonds won an MVP, the debate primarily is over who should have gotten second, and this should have been Sheffield easily. The batting champion Sheffield hit 33 home runs, a combo that had only occurred twice in the quarter-century since Carl Yastrzemski's triple crown: Billy Williams in 1972 (who's on the list above) and Fred Lynn in 1979. But, although Bonds was better than Sheffield, why Sheffield was behind Terry Pendleton is anybody's guess, when Pendleton played the same position in the same division but with worse stats across the board. You can't get more directly comparable than Sheffield and Pendleton in that year, and Pendleton came out ahead. That just bugs me. ... Sheffield's year, on a team he barely played for and at a position from which he soon moved, seems to be one of the forgotten great seasons of the last 20 years. I don't know what else would be on that list, but this definitely is one of them.

1997 AL: Martinez over Griffey

Top five eligibles: Martinez, Randy Myers, David Justice, Jim Thome, Tim Salmon

Griffey was a unanimous choice for the award, and funny things seem to happen down-ballot when everybody knows who the winner is. There's no argument from me that Tino wouldn't have been the next logical choice. It's that Randy Myers pick below him. Myers' 1.51 ERA as closer for a wire-to-wire division winner was very shiny ... in his almost 60 innings of work. How this was infinitely more valuable than Roger Clemens' Triple Crown season, Randy Johnson's similar work, or three hitters whose OPS exceeded 1, I don't know. (Myers also placed fourth in Cy Young voting that year.) The next year would be Myers' last as a pitcher, plagued by ineffectiveness, injuries, and a strange waiver claim. All told, it was a very far cry from the season that somehow captivated award voters. The choice strikes me as illogical anyway, but regardless of how you feel, that was quite a precipitous fall.

2003 NL: Thome over three

Top five eligibles: Thome, Javy Lopez, Eric Gagne, Todd Helton, Mark Prior

Would Thome's Hall of Fame case be more impressive to the untrained voter if he had an MVP? Will National Leaguers of this era have their cases hurt by Bonds and Pujols capturing every MVP known to man? That Thome overtakes Sheffield in this exercise is even more intriguing than Thome's award. Thome and Sheffield might be two of the hitters most likely to have a no-MVP non-argument thrown at them when their candidacy comes up. Thome's season was plenty good, as was Sheffield's, and having two all-world performers in your league shouldn't mean they're the only two who get in the Hall. I know that you, the reader, know all these things, but as the past decade in the NL is in danger of historical reductionism, it's worth trying to salvage it now ... or something.

Conclusion

Some interesting names and forgotten seasons show up when you peel off the layers of repeat MVP winners. That doesn't mean the League Award was a good idea, but using its lens on history makes for fun argumentation.

Click here to learn about THT's download subscriptions.