The wacky, wacky west

Every baseball season is filled with stories of team and individual glory and disappointment. So far this year, the American League West has been perhaps the most storied division in baseball. Let’s take a look team by team.

The Angels

At this stage of the game, the Angels are beating their projected win/loss* record by seven games, which is pretty impressive.

*When I say projected win/loss record, I’m referring to the number of games we’d expect them to win based on the total number of runs they scored and allowed. That’s typically called the pythagorean record.

Let’s take a closer look at how they’re doing it:

Team    Bat   Start   Pen
LAA    -0.41   5.76  3.14

The Angels have scored 20 fewer runs than the average team, which would normally result in a WPA of -2. So the Angel bats actually deserve some credit for delivering in the clutch. Maicer Izturis, in particular, has answered the call. He’s fifth in the majors in delivering with the game on the line, according to Fangraphs’ clutch score.

At the same time, the Angels have given up about 25 fewer runs than average, which would normally result in a positive WPA of 2.5 games. However, the combined WPA of the starters and bullpen (and fielding) is almost nine. Big advantage, arms. In particular, the starters have stood out: second in major-league ERA at 3.75 (Oakland is first) and first in the majors in WPA (the Indians are second).

Here’s the key, however. The Angel bullpen is 20th in the majors in ERA (4.14) yet third in bullpen WPA (the Rays are first and the Phillies are second). The Angels are beating their pythagorean record the old-fashioned way: giving up lots of runs in blowouts and winning the close games (32-13 in games decided by one or two runs). It’s the classic “how to beat your pythagorean record” story.

Thing is, the Angels are really good at it. The record for the highest positive pythagorean variance at the end of June is 8.5 games, set by the St. Louis Cardinals in the strike-shortened season of 1994. Ironic footnote: The Cards wound up with only a four-game variance by the time the season ended in early August. Pythagoras giveth and he taketh away.

The record for the highest full-season pythagorean variance is 13, set by the Tigers in 1905. Last year’s Diamondback team had the ninth-highest variance ever, 11 games. Two years ago, they had a 12-game variance in their favor, the second-best ever. And the year before that, the Yankees had the third-best variance ever, 11 games. What’s going on here? Here’s a little graph of the year-by-year maximum pythagorean variances since 1960:

There isn’t much of a pattern here, other than the big bars in three of the last four years. Is it part of a larger trend? Well, it could be that the increased size, flexibility and strategic use of bullpens has led us to a “new era” of pythagorean variances: still very difficult to predict, but powerful when they pop up.

The Athletics

The Yankees must be Athletic gold and green with envy. Here the Yankees are, getting skewered in the press for relying too heavily on young pitching, while the Athletics, who are supposed to be in a rebuilding mode, are thriving on, um, young pitching.

In fact, the A’s are already coming out ahead in the Dan Haren deal with the Diamondbacks, the supposed deal for the future. Taking a look at some of my favorite pitching value stats:

                PRC    WPA    WSAB
Dan Haren        55   1.20       8
Smith/Eveland    73   1.82       6

Eveland and Smith have combined for more Pitching Runs Created and WPA than Haren, though Haren has the lead in Win Shares Above Bench. Let’s call it a tie.

What’s more, Justin Duchscherer’s move into the rotation has obviously been successful and rates up there with Ryan Dempster’s similar conversion for team impact. His pitch breakouts show that he’s changed his approach as a starter: relying more often on his fastball and less on his breaking stuff. Dempster’s pattern change has been similar, but not as extreme.

As you may know, the Internet is alive with pitching analysis. You can find out more about each of these starters here:
{exp:list_maker}Dana Eveland
Greg Smith
Justin Duchscherer {/exp:list_maker}In a way, every general pitching comment is also a fielding comment, isn’t it? Stats like Pitching Runs Created and Win Shares attempt to isolate pitching from fielding, but how well do they do? I don’t know. And the Athletics might be the best-fielding team in the majors. At this stage of the season, they are 30 plays above average, according to our team batted ball summary. Only the Cardinals are higher.

According to John Dewan’s plus/minus system (available at Bill James Online, for a fee)…
{exp:list_maker}Mark Ellis is the second-best-fielding second baseman in the majors (+17).
Daric Barton has been a pleasant surprise at first (+10, 5th in the majors).
Jack Hannahan has also filled in admirably for Eric Chavez at third (+6, 8th).{/exp:list_maker}But the rest of the team is average or worse. In fact, Bobby Crosby is the worst-fielding shortstop in the majors. The overall story doesn’t quite seem to add up to me, but it works.

The Rangers

Remember the graph at the top of the article? The Rangers have hovered around that .500 line a long time. In fact, they’re probably the most consistently average team of the year. Which is weird, because their batting is the best in the league and their defense the worst, even when you adjust for that hitters’ paradise of theirs.

But the Rangers have been within one full game (translation: two wins) of .500 for 38 straight games. Not just playing .500 during that time, but starting at .500 too. So I wondered, is this some sort of record?

The answer is: not yet. Here’s a list of the most consecutive days a team has spent within one game of .500 within a single season:

Team	Year	Games	Comments
Red Sox	1980	56	Once you adjust for ballpark, an exceptionally average team.
Padres	2005	55	Finished on a .500 roll. And yet managed to win the division.
Indians	1967	55	Through mid-July the essence of average. Then the bottom fell out.
Mets	1970	52	The WS champs were decidedly average in 1970. Three years later they would go to WS on an average record.
Pirates	1993	47	The year after Bonds left. Less than average after early June.
Reds	1964	47	Average the first two months of the year, scintillating after. Finished second.
Athletics	1890	46	The Players’ League Philadelphia Athletics. Out of business a year later.
Indians	1947	46	This .500 streak carried through mid July. Won it all the next year.
Angels	2000	45	An average string that started the first day of the season. Glaus hit 47 HR for the year.
Expos	1974	45	Another one through early July. Remember Bob Bailey?
Dodgers	1964	45	Sandwiched between two World Series teams; this was a mid-year streak.
Cubs	1995	45	The Grace/Sosa/Dunston Cubs. And Frank Castillo

Is there a lesson for the Rangers here? Not really. But let’s see how close they come to the all-time record for consecutive averageness.

The Mariners

You can read the Fangraphs’ WPA graph from Monday’s game with the Mets, but that doesn’t tell the real story. This one does:

Don’t believe me? Read this game thread. (Caution: NSFW)

References & Resources
Okay, so I stole the italics idea from Joe Posnanski, too. Hey, it’s the Internet. Memes spread.