The most surprising .300 hittersby Jeff Sackmann
July 29, 2010
Whatever it really means, like it or not, there's magic in hitting .300. Baseball loves round numbers, and a hitter gains a whole lot of respect from fans and peers if he's able to cross that particular threshold. Even if his name is Yadier Molina.
Of course, some .300 hitters are different from others. For every Yadi, there's a Todd Helton or Joe Mauer, who disappoints when his batting average doesn't start with a three. For today, I'm interested in the Yadis of the past, the guys with proven track records of mediocre batting averages who nonetheless join the .300 club.
Finding the surprises
How do we identify which .300 hitters qualify as surprises? We could compare each .300 season to the player's previous campaign, but that might just as easily reveal a unusually bad earlier season. We need to take a broader view.
That's where Marcel projections come in. Usually we use projections to predict the future—after all, that's pretty much the meaning of the word. But in order to predict the future, any forecasting system is, at least in part, making an estimate of the present. In other words, when forecasters projected Yadi's 2008 season, they were looking at the past and making a conclusion as to his true ability level.
It is fascinating—maybe even valuable— to have an estimate of a player's true talent level given the information available at a certain time in the past. To find projections of a player's performance for recent years, you're just a Google search away. But what if you wanted to estimate, say, Ted Williams' true talent level given the data available before the 1943 season, which he missed due to the war?
For this project (and others to come), I ran the Marcel algorithm through baseball history back to 1901. If you'd rather find some cool stuff yourself rather than read about what I find, I've made the full spreadsheets available. Have fun.
Marcel takes a player's performance over the past three years, combines it with a dose of league average, and spits out an estimate of how he will perform the next year. Thus, for every .300 season (or any other single-season milestone you care to consider), you can compare the actual results to Marcel's preseason estimate.
When discrepancies arise, the speculation can begin. In the jarring differences we'll see in a moment, did the player just get lucky for a season? Did he "put things together?" Is Marcel less skillful when forecasting certain types of players? Whatever the answer, it surely isn't the same in every case.
From zero (or .225) to .300
1. Richie Scheinblum - 1972. In March and April of '72, Scheinblum must have impressed somebody. In early May, the Royals traded away Al Oliver, and they handed Scheinblum the right field job. You have to wonder why.
The previous year, he hit .143 in 57 plate appearances. Two years before that, he got 222 plate appearnces for the Indians, in which he hit .186. Sure, a disproportionate number of those were pinch-hitting appearances, in which it's harder to have success. Given under 300 plate appearances, Marcel doesn't have a lot to go on, but even when heavily regressing him up to league average, the result is a .222 average and a wOBA of only .278. Yikes.
Sure enough, given the opportunity, Richie hit an even .300. He even walked 11 percent of the time for good measure. And it wasn't a total fluke—as a part-timer, he did even better in 1973, hitting .307. But in '74, he posted a .183/.242/.200 line and never got another shot.
2. Bill McKechnie - 1914. From 1911 to 1913, future Hall of Fame manager McKechnie was a part-timer, but Marcel saw enough in his roughly 600 plate appearances to mark him down for a .225 average in 1914. What Marcel didn't know is that he would move from the AL and NL to the upstart Federal League.
The 1914-15 Federal League is a subject worth an article in its own right (look for that in August). For today's purposes, it's enough to know that by poaching as many major leaguers as it could, it put out a pretty decent level of baseball but ended up diluting the talent pool in all three leagues as a result.
This precursor of the expansion era isn't enough to explain McKechnie's jump from .227, .247, and .129 up to .304, but it does help. Unlike Scheinblum's, McKechnie's magic didn't last. He managed only a .251 average in the 1915 Federal League and never hit above .260 again. Surely his two World Series rings as manager made him feel better.
3. Jim Hickman - 1970. Now this is what you call a career year. In 13 major league seasons, he hit above .275 only once. His OBP topped .370 once. He slugged better than .465 only once. But all of those personal bests came in tandem. For the '70 Cubs, Hickman hit .315/.419/.582, leaving it to posterity to figure out where those 613 plate appearances came from.
There are minor explanations, but not enough to explain the difference between Marcel's forecasted .234 average and Hickman's actual .315. The National League average of .280 was higher in 1970 than any other year of his career. And a BABIP of .337 was far better (luckier?) than any other he posted. Maybe Hickman just liked goats.
4. Ray Powell - 1921. For the first time, we're looking at a guy who was a true regular for the two years before his surprise .300 season. Powell logged a lot of time for some bad Boston Braves teams but posted his best results on a decent one.
In 1919 Powell hit .236/.303/.326 and led the league in strikeouts for good measure. In 1920 he struck out even more often and fell back to a .225/.282/.314 line. (Maybe it was some consolation that George Kelly struck out even more often, keeping Powell out of first place in that department.) Had Marcels been published, there wouldn't have been many Braves fans taking issue with the forecast of a .238 average.
Except for Powell himself. He kept striking out in 1921, leading the league again. He also led the league in triples, homered a dozen times (no mean feat in the early 20s) and hit .306/.369/.462. Maybe he hated Rabbit Maranville; the Braves shipped him out in the offseason, and for the next three years, Powell hit much better than the future Hall of Fame shortstop.
5. Rickey Henderson - 1999. For the Padres, Angels, and A's in 1996-98, Rickey walked a ton. And it's a good thing, because he wasn't getting to first base very often via the base hit. Given those three years of data, Marcel didn't question his value—it projected a .337 wOBA—but there was no reason to expect a renaissance of singles. Marcel forecast a .238 average.
Mets fans know the rest of the story. As part of the Wild Card-winning club, Henderson kept on walking and complimented his famous baserunning and patience with a batting average of .315. As was the case with Hickman, his BABIP explains a lot. 36.3 percent of Rickey's balls in play became base hits, the second-highest rate of his career. It was one of only three times he topped 30 percent after his 30th birthday.
There have been close to 3,000 .300 hitters in the last hundred years, and had Marcel been around all that time, about 200 of them would have been projected to hit .260 or worse.
Click here to see the full spreadsheet of all those .300 hitters, together with their Marcel projections as well as their Marcels for the following year, to get an idea of how much the surprise season influenced an estimate of their true talent level.
Jeff Sackmann is the creator of MinorLeagueSplits.com. With Kent Bonham, he founded CollegeSplits.com. Jeff and Kent blog about college baseball and the draft, and you can follow them on Twitter for bite-sized snacks of minor league and college stats. Jeff also has an email address.