More tales from the team splits junk drawer

A few weeks ago I wrote an article on some weird and fun things I found out thanks to the batting splits available for every team from 1957-2006 (available at Baseball Reference). Having put big heaping gobs of data into Excel, I can now look up and find some interesting odds and ends about the most extreme and interesting team achievements of the last half-century.

Essentially, these articles involve me diving into a junk drawer that I’ve put together, and coming back with whatever knick-knacks catch my fancy.

The weird world of Sam Mele

At the end of the last article, I looked at team stolen bases, finding that some teams never stole when the game was a blow out. This was especially true of teams run by Felipe Alou and Sparky Anderson.

I want to look at the opposite extreme: Were there teams more prone to run when it didn’t matter?

I’m going to divide all stolen opportunities into two categories: blowouts (when one team was ahead by five or more runs) and close games. I have a basic formula to determine how frequently teams tried to steal: (SB+CS)/(H+W+HB+ROE-HR-2B-3B). For those unclear, “ROE” is reached on error. It ain’t perfect but it gives a quick ‘n dirty sense of how often teams ran.

When the game is close, teams try to steal one in every 10 opportunities (10.03 to be exact). When it’s out of reach, they go for it about once in every 30 occasions (29.02). Since I got the data, here’s exactly how often teams try to steal over the last half-century in each circumstance:

Situation           Steal Frequency
Tie Game    Once Every 8.87 Chances
1 Run Dif.  Once Every 9.49 Chances
2 Run Dif. Once Every 11.18 Chances
3 Run Dif. Once Every 11.86 Chances
4 Run Dif. Once Every 12.89 Chances
5 or More  Once Every 29.02 Chances

Certainly makes sense. Stealing bases is a way to play for one run. The bigger the gap, the less that one means. But in a sample size of 1,240, you can always find some screwy teams.

Sure enough, 28 teams stole more frequently when the game was out of reach. Here are the most extreme:

Team      Increase
1962 MIN    76.32%
1960 BAL    63.45%
1959 CIN    40.70%
1966 NYY    37.80%
1972 DET    36.00%

Few fun things to note based on that. The king of them all was the 1962 Twins. They didn’t run that often in blowouts actually—only once every 19 chances. But they only ran once every 34 opportunities when it was closer.

And that was part of a broader trend for the Twins. In his six years at the helm, the Twins ran more often in shellackings four times, and only a little less often five times. This doesn’t make much sense. Why play for one run when one run doesn’t matter? Sam Mele had a lousy reputation as a manager. When he fired pitching coach Johnny Sain after 1965, staff ace Jim Kaat wrote a public letter saying the team would be better off getting a new manager instead. The mid-1960s are an historic what-if; a team that didn’t seem to achieve as much as they could’ve. After Calvin Griffith axed him, no team ever hired him to manage again, despite being the only man to take the franchise to the World Series between 1933 and 1987. That he’d be he king of this screwy base running strategy amplifies what a lousy idea it was.

In his defense, he was also the product of his era. All but one of these squads came from 1958-72. At least one team a year from 1958-1967 ran more frequently in blowouts. It peaked in 1963 and 1965, when four teams each stole more when it mattered least. Since 1972, only the 1989 Mets have stepped up their larceny when the team had enough insurance to cover it. Making it even stranger, that’s when baseball’s offense reached its post-deadball era nadir. If ever there was an era where one couldn’t overcome a five run deficit, it was then. Maybe that’s why they tried chipping away. After all, they had less chance to wait on the long ball.

Step away from the lineup card, Lou Piniella

Here’s another bit of junk in the drawer: Which teams have done the worst jobs choosing their leadoff hitters? Baseball Reference has splits by batting order, so how ’bout we look into that.

One longstanding bit of conventional wisdom (both in and out of stathead circles) is that you need a leadoff man who can get on base. The problems are that some look at speed or batting average rather than on-base percentage. So which teams got the worst OBP out of their leadoff slot?

Turns out the 1969 Padres take the cake, with a .237 OBP (and .196 batting average!) setting the table. The 1981 Blue Jays are a close second with .239, and no one else approaches them. The best, if you’re wondering were the 1958 Phillies with Richie Ashburn posting a .437 OBP.

In the partial defense of the Pads, that wasn’t so much a failure to find a leadoff hitter as much as it was a failure to find an offense. They had one of the worst team-wide OBPs of the last half-century.

Which teams did the worst with what they had. I’ll divide leadoff OBP by team OBP. For once, I’ll begin with the best, and save the worst for later:

Team         #1 OBP  Team OBP          Hitter
1960 CHC      0.404     0.313      Richie Asburn
1958 PHI      0.437     0.339      Richie Asburn
1981 OAK      0.406     0.317   Rickey Henderson
1963 CAL      0.395     0.309      Albie Pearson
1985 MTL      0.397      0.31         Tim Raines

Gives you a whole new appreciation for Richie Ashburn, doesn’t it? The common denominators are: good hitter with out machines behind him. Makes sense.

Now for the fun part: Which teams did the worst possible job filling out the top of their lineup card:

Team        #1 OBP  Team OBP       Hitter
1994 SEA      0.27     0.335      Rich Amaral
1999 SEA     0.279     0.343     Brian Hunter
1981 TOR     0.238     0.291  Alfredo Griffin
1960 NYY     0.272     0.329       Tony Kubek
1969 SDP     0.237     0.285       Jose Arcia

Remember that nice little defense of the Padres? Screw that. Bad offense or not, Pedro Gomez did a terrible job figuring out his lineup. Not only are they on the list, but they and the 1960 Yanks are ones that should be the most embarrassed. At least the top three played in a DH-league. The 1960s teams had such wildly unproductive table setters even when pitchers gobbled up many at-bats.

The really striking bit is that the same damn team led by the same guy has the top two slots, the Lou Piniella’s Mariners. He has an idea of who should lead off: a fast guy. If he gets a fast guy who gets on base, like Ichiro!, all is well with the world. If he can’t, well, any ol’ speedster will do. This year he has Alfonso Soriano leading off. He’s definitely more a power guy than anything else, and his OBP is really suffering in the dog days of summer this year (sub-.300 OBP since July 1). He’s speedy, so he leads off.

If you’re curious, 797 of the 1,240 teams have better OBPs from their top slot than their team. That may sound low, but normally your best overall hitter has the best or near-best OBP. He’ll be the guy in the heart of the lineup. And he’ll have some similar players around him. Besides, of the teams that have lower leadoff OBPs, many are close to their team averages. Of the 343 teams with lower leadoff OBPs, 121 score at least 98% their team’s mark. Only 44 get less than 90% of their team-wide OBP from the first hitter.

That goes for you, too, Clint Hurdle

There’s another, even nastier way to look at this: have any teams actually received better OBP from each of the other lineup slots? How often do teams put their single most OBP-challenged man at the top of the line up? I really need to make one qualifier: for leagues without a DH, I’ll ignore the pitcher’s slot.

Well, as it happens, there have been 79 teams where the worst OBP came at the top of the order. Seventy-nine! Normally, I think sabermetric criticisms of managers are overblown nitpicks, but … my God. There’s no excuse for any team to do this, let alone one or two teams every year doing it.

Stupid lineups peaked in 1960, when four of the 16 teams—the Giants, White Sox, Yankees, and Orioles—all committed this sin. Incredibly, they had three extremely well-regarded managers in Paul Richards and Hall-of-Famers Casey Stengel and Al Lopez. Bill Rigney lasted forever as well.

From a broader perspective, it reached a sustained prime from 1967-1978 when one-tenth of all teams (28/284) made this same damn flaw. Most notable was the 1971 Red Sox, who somehow, some way, got a worse OBP from their leadoff man than their No. 9 slot. This was before the AL adopted a DH. The mind boggles.

Subsequently, this problem dropped dramatically, and from 1979-93 it only happened 14 times to 362 teams, once for every 26 squads. Despite the rise of sabermetic numeric analysis uber alles, bad lineups have made a comeback. Heck from 2002-06, in the same period Moneyball made OBP a household term, an even dozen teams have committed this sin. In fact, team Moneyball itself narrowly missed making the list. The year the book came out, the 2003 A’s had better OBP from their 2-8 slots than their top man.

The five managers with the most seasons under their belts: Tony La Russa, Anderson, Gene Mauch, Bobby Cox, and Joe Torre, have combined to do it once in 125 seasons (Mauch in his sophomore season, 1961).

A couple guys, though, have really distinguished themselves for their inability to put the best player second worst player up on top: Ralph Houk, Harry Walker, and Clint Hurdle. All had three teams made the Unwanted 79.

At least Houk could blame it on a long career. With 19 seasons, he filled out more lineup cards than the other two combined. He did it twice early in his career, following the lead of Casey Stengel, his mentor, placing All-American out machine Bobby Richardson at the top of the lineup card. Even still, in over half his seasons his leadoff OBP was lower than the team wide OBP.

Harry Walker did it three times in only eight years. This was a man who had Joe Morgan on his squad for much of his career! He ensured the Astros had bleak starts to their lineups in 1968, 1971, and 1972 (the last after Morgan had gone to Cincinnati).

Clint Hurdle makes Harry Walker look like some kind of super-genius. He’s only managed five complete seasons, and yet pulled off this feat three consecutive years from 2003-05. His leadoff OBP was below team average in 2002 as well. Last year, it was ever so slightly better than the team’s OBP: .343 to .341. Take out the pitchers, and I’ll bet his leadoff hitters are still worse than the rest of the position players. This year, his leadoff OBP in Colorado is worse than slots two through six, and the team-wide mark.

Much to my chagrin, I’m getting long-winded in my old age, and that’s all the junk I can dig out for this week. But there’s plenty more slop in this drawer.