The Outlier

Sometimes when you poor through some stats, you'll come across an outlier. Now, you may not know exactly how to define an outlier, but rest assured, you all know one when you see it. And you've all not only come across them in stats, but you've watched them being made.

OK class, listen up. Today's Algebra lesson is about "outliers." Pay attention, because there will be pop quizzes along the way.

First, let's define "outlier." It is not, I repeat, it is not what you'd be likely to get if you were to ask Omar Moreno or Rey Ordonez or Juan Pierre if they were/are effective offensive players. No, "outlier" is an algebraic term, not commonly a baseball term. It refers to an anomalous value in a set of numbers, a number that just doesn't fit in with the rest of the set. For instance, here's an example. Suppose you have a set of numbers like; 53, 44, 52, 56, 48, 2, 50, 57, 42. So what's the outlier? What's the number in the set that doesn't fit? That's right, it's 2. That's the outlier in the above set of numbers. The outlier can also be a number that doesn't fit in the other way… 13, 12, 20, 17, 21, 342, 16, 22. What's the outlier here? Ah, I see my old high school baseball coach knows math as well as history. Very good, Mr. Gratwick, it's 342.

Now, this is not an eighth grade Algebra class (if it was, I'd have my daughter Maggie write it), it's a baseball column. So what does this have to do with baseball? Simple, outliers happen all the time in baseball in the context of a player having one really big year in the course of an otherwise ordinary career. Now sometimes these outlier years are pretty harmless, mere moments of curiosity and/or excitement in the course of an otherwise mundane list of years or numbers. For instance…

21, 13, 12, 16, 50, 18, 18, 24, 19

Mark from Earlham, either one of you... what's the outlier? Very good, it's the 50. Milosh Mamula would be proud. (Note to non-Earlham grads – that's a real name.) Now, who can tell me what these numbers represent? Ah, Mr. Hardy, you have your hand up. Excellent, it's the set of Brady Anderson's yearly home runs. That "50" stands out like a sore thumb, doesn't it? In 1996, Brady Anderson somehow hit 50 home runs in a single season for the Orioles, and he never hit more than 24 in any other season. Can you spell, "suspicious" class?

Or, how about this one, same page, different pew…

14, 28, 16, 39, 61, 33, 23, 26

C'mon, this one is easy. Correct, Ms. Dellinger... Roger Maris' yearly home run totals for his complete seasons. That "61" actually fooled a lot of Hall of Fame voters into thinking Maris was as worthy of induction as your grandfather, the consistently fine Edd Roush.

Another easy one, in the same vein (or maybe in the butt)…

25, 24, 19, 33, 25, 34, 46, 37, 33, 42, 40, 37, 34, 49, 73, 46, 45, 45, 26, 28

Yeah, let's a lot of home runs. But, you can't deny that one number in the set just jumps out. Did you realize that, prior to the 2001 season, and after the 2001 season, Barry Bonds never hit a 50th home run in a single season? Bonds had as many 50+ home run seasons as Brady Anderson, as Ryan Howard, as Prince Fielder (and his dad), as David Ortiz. In fact, Bonds only came close to reaching 50 the year before he hit 73. He only topped 45 four times. Now maybe that was because other teams developed an irrational fear of pitching to him after 2001. But, he had as many 40-49 home run seasons before 2001 as he did after 2001. Can you say "suspicious," class?

Outliers can also be found in other statistics, not just in raw numbers. Take this set of OPS+ numbers…

141, 201, 136, 136, 120, 147, 133, 128, 142, 127, 126, 149, 129, 127

The outlier is pretty obvious, isn't it? Do you know who put up these numbers? Clearly, he was a pretty good hitter. Even taking out the outlier, this guy was consistently an offensive power, regularly topping the average offensive output of his league by an average of 34 percent. Does it help to note that his offensive abilities haven't really been recognized, because he played largely in the offensively-challenged 1960s, and his raw numbers generally weren't overwhelming? How about if we say that this outlier was in the same year that Maris put up his outlier? That's correct, Mr. Ryczek, it's Norm Cash. In this case, it's pretty well established how he put up that .361 average and 41 home runs in 1961. As Cash himself was quoted as saying, "I owe it all to my corked bats." Well, that's one way.

Rarely, you'll see a player put up outlier numbers in unconnected categories. Wally Moses was one such oddball. His single season home run numbers were…

5, 7, 25, 8, 3, 9, 4, 7, 3, 3, 2, 6, 2, 2, 1, 2, 0

Not much doubt about the outlier here, is there? If it wasn't for the fact that this was 1937 and not 1997, we'd also have an easy explanation.

Then there were Moses' yearly stolen base numbers. They were just as extreme…

3, 12, 9, 15, 7, 6, 3, 16, 56, 21, 11, 4, 3, 5, 1, 0, 2

What was he up to in 1943 with the White Sox? Who knows…

You can play this game with pitchers as well, say…

14, 9, 16, 15, 13, 14, 7, 15, 17, 17, 27, 12, 9

And the outlier is? That's right, Larry, it's the 27. And this set is? Correct, Bob Welch's yearly win total for his seasons as a regular starter. Now, Welch was a good pitcher, won more than 200 games, but that 27 looks kind of odd, doesn't it? You have to think luck had a great deal to do with this outlier, since the years surrounding it were all with pretty good Oakland A's teams.

Three other pitchers come quickly to mind when discussing outliers. However, in all three of these cases, they managed to sucker one or more teams into believing that their outlier season wasn't a fluke, despite the fact they continued to flounder, or at least not excel, for some years afterwards. It's cases like this that seem to indicate teams historically don't always know their Algebra. Jack Coombs came up with Connie Mack's Philadelphia Athletics as a talented young all-around ballplayer (he was an excellent hitter who also played some in the field). It took him a while to get rolling, posting sort of average ERA+ marks of 109, 83, 128 and 103 in his first four seasons (1906 to 1909). Then, in 1910, and really only for one year, he was one of the great pitchers in baseball, winning 31 games, including 13 shutouts, with an ERA+ of 182. Although he won 28 and 21 games in 1911 and 1912, he never really had a year again like 1910. His ERA+ figures for his entire career (using only those seasons wherein he was a fairly regular starter) after 1910 were…

90, 94, 108, 100, 70, 73

His shutout figures for his entire career are also worth recounting, since Coombs had 13 of his 35 career shutouts in one year.

1, 2, 4, 6, 13, 1, 1, 2, 3, 0, 2

After he missed almost all of the 1913 and 1914 seasons, and Mack released him as part of his first salary dump, Coombs found another taker for his services in the Brooklyn Dodgers, for whom he was capable (the last four ERA+ numbers represent his work with the Bums), though hardly brilliant. Of course, Uncle Robbie, et al, traditionally did things on the cheap. Around this same time, they picked up another pitcher of questionable value, this time from the Giants, Rube Marquard, maybe on the theory that he would match his 1911 or 1912 season. Even though Coombs and Marquard both pitched for the Dodgers in the 1916 World Series, neither would ever reach the heights they had attained with their original teams.

Even so, the Dodgers did better with Coombs and Marquard than the employers of Gene Bearden did after the knuckleballer's one big year. Here are Bearden's yearly win totals for his entire major league career…

0, 20, 8, 4, 3, 7, 3

Talk about an outlier. After winning 20 in 1948 for the Indians, including the pennant-clinching playoff game, plus another game in the World Series, Bearden only won 25 more games in his entire career. The explanation here is pretty simple, at least if you've read Bill Veeck's "Veeck… as in Wreck," wherein the Indians owner in 1948 explained that Casey Stengel's return to major league baseball as manager of the Yankees in 1949 doomed Bearden's career. Casey had seen Bearden pitch in the Pacific Coast League, and knew that his knuckleball, in addition to being his only really effective pitch, usually ended up outside the strike zone when it crossed the plate. Casey told his Yankee batters to lay off the knuckler, and Bearden was toast. What's odd though is that after Bearden was waived by the Indians in 1950, he was signed by the Senators. And then signed by the Tigers after he was waived by the Senators in 1951. And then traded to the Browns (who were then owned by Veeck, who certainly should have known better) in 1952. And then picked up by the White Sox on waivers in 1953. In other words, half of the teams in the American League took a flyer on Bearden, apparently hoping he could repeat his outlier season, after they all apparently had the word not to swing at his knuckleball.

Bearden is far from the only major league pitcher to have a single, big win season. Another in this crew put up these win totals…

0, 5, 2, 20, 13, 2, 4, 6, 3

Alright class, who are we talking about this time? Matt, I think you just beat out brother Andrew with the answer on this one. It's Wayne Garland, who unexpectedly won 20 for the Orioles in 1976, signed a (for the time) huge free agent contract with the Indians in 1977, and then threw his arm out trying to justify same.

However, major league general managers never seem to learn, since this phenomenon of hoping for a repeat of an outlier season has continued into the 21st Century, even to this very day. These OPS+ figures were put up by a certain major league catcher, and earned him a huge payday in 2004…

117, 100, 129, 124, 104, 89, 74, 169, 127, 106

That's a pretty good hitting catcher, but that "169" is way out of line with the rest of his production. Anyone know who this is? That's right, Brian from Reading , it's Javy Lopez. He posted all of these OPS+ figures in seasons where he caught at least 100 games, but that 2004 season, just before he signed with the Orioles, was an outlier.

That brings us to a couple of puzzling free agent signing during the current off season. The next two sets of numbers belong to the same player, the first being his OPS+ figures as a regular, and the second his home runs. Caveat emptor…

101, 114, 91, 97, 88, 163, 93, 105, 112, 108

15, 20, 13, 21, 23, 48, 19, 25, 26, 25

What these numbers don't show is that this same player put up similar outlier numbers in runs, hits, RBIs, OPS and total bases in the same year, that being 2004. I'll bet the master of trivia, D. Bruce Brown, knows this one. That's right, it could only be the Red Sox' new third baseman, Adrian Beltre. While it is pretty well-established that Theo Epstein is a genius, you have to wonder how desperate the Sox were this winter for a third baseman. Maybe they didn't think Kevin Youklis could go back to third, or that Big Papi couldn't handle first, or that they couldn't find someone else who could handle first. Maybe, but this is a little hard to believe, at least if the Sox think they're getting a third baseman who can hit enough to get them past the Yankees in 2010. Beltre is a fine fielder, and he reached the majors at a very young age, but his entire career, except for his outlier 2004, screams that he is a barely above average major league hitter. This signing has brought very little critique, maybe because it's Theo Epstein doing the signing, but Beltre is one of the more extreme outlier cases on record.

Finally, to bring the Algebra/History lesson to a close, we give you the light comedy troupe known as the Mets, who obviously haven't done their homework in either subject. In other words, they haven't learned from history, and they don't understand the concept of outlier. Their trade for Gary Matthews, Jr., is as equally off-the-wall as the Beltre deal, so there's no point even giving a quiz. Even if he hadn't been figured for HGH, his 2006 numbers say, "stay away!"


84, 111, 82, 103, 94, 121, 93, 77, 83


.705, .780, .675, .811, .756, .866, .742, .675, .697

Batting Average

.227, .275, .248, .275, .255, .313, .252, .242, .250

The outlier in every case is the 2006 season. And, he's been awful every year since 2006. Other than that one year, he isn't even an average offensive player. Even if he's a gift, and he wasn't since the Mets still have to pay him $2 million and they gave up a serviceable middle reliever, he's not someone a contender needs to give meaningful playing time. But, then again, are the Mets a contender for anything other than third place?

Philly Baseball Insider Top Stories