Notice: Trying to get property 'display_name' of non-object in /var/www/html/wp-content/plugins/wordpress-seo/src/generators/schema/article.php on line 52
keyboard_arrow_uptop

“It was the best of times, it was the worst of times …”

-Charles Dickens, “A Tale of Two Cities”

If you paid even one whit of attention toward the American League West last season, you’re probably already aware of the major high and low points: The Angels juggernaut fails to win the division for only the second time in seven years, and accompanies the Athletics on a journey to a not-so-magical place called Five Hundred Land. Meanwhile, Texas earns its first post-season berth since 1999, battles valiantly all the way to the World Series, and finally gets rolled up in five games by San Francisco. Josh Hamilton wins the AL MVP. And somewhere along the way, a hopelessly out-of-contention Mariners squad allegedly pulls out of a preliminary agreement on a deal that would have sent Cliff Lee packing to the Bronx (much to the consternation of unnamed Yankees officials, some of whom accused Seattle of not bargaining in good faith), and instead moves him to Texas in exchange for a four-player package and cash.

On the morning of The Trade (July 9, to be exact), the Mariners found themselves mired in a 34-51 hole and an even 16 games behind first-place Texas, amounting to an approximate playoff odds percentage of 0.1%—but I want to spend less time focusing on that past state of hopelessness itself, and more on how Seattle ended up in such a deplorable state in the first place. A review of the tape indicates that the Mariners’ run prevention up to that date in the season was nothing worth writing home about (4.26 runs allowed per game), but was still a few ticks better than league average even after the inclusion of park factors. Phrased another way, it helped explain Seattle’s problems only to the extent that they were not a great team; owning a merely decent pitching staff and defensive corps may preclude a 95-100 win pace, but taken in isolation, it doesn’t explain how the Mariners ended up on a 60-65 win pace.

Where the root cause of the Mariners’ competitive ills could be found was in the other key variable in the Pythagorean record formula: run-scoring. By the morning of July 9, the Mariners had plated 291 runs in 85 games, good for a 554-run pace over the span of a 162-game schedule, or marginally better than the 10th-lowest scoring team in a full season (550 runs, by the ’76 Angels and ’88 Orioles) since the designated-hitter rule was enacted in 1973. By season’s end, Seattle held this most undesirable post-1973 record outright with a final 513-run showing, which can only be properly illustrated with this shocking reminder: Of the 12 Mariners hitters who amassed at least 200 plate appearances on the season, only two posted a final OPS+ of 100 or higher: Ichiro (113 OPS+) and Russell Branyan (123 OPS+). The next highest? Franklin Gutierrez, at 87. Only two regulars mustered an on-base percentage of .340 or higher. Only one mashed more than 12 home runs. High-salaried newcomers Chone Figgins and Milton Bradley crashed and burned in awe-inspiring fashion, hitting .259/.340/.306 and .205/.292/.348 in their respective Mariners debuts. The offensive collapse of the Seattle starting nine was swift and absolute.

Why am I devoting page space to what some would now regard as ancient history, and what others would probably rather bury than spend valuable time reminiscing over? My purpose is twofold: (1) plenty of observant and intelligent baseball fans are aware that the 2010 Mariners offense was bad, but I’m willing to bet that there are a considerable number of those who didn’t realize the full extent of the badness, and I feel like that’s worth revisiting; and (2) this makes for an appropriate segue into a forward-looking glance at how the Mariners’ offense—and all the others in the AL West—is projected to fare in 2011, but perhaps not in the conventional manner you’re expecting. The question I want to answer is this: How does each projected starter’s PECOTA projection compare against an objective definition of “good” or “bad” offensive production at their given position?

In July 2007, Kevin Goldstein (with a generous data-mining assist from William Burke) penned an article on major-league position differences, constructed from the offensive statistics of the 30 players who had amassed the most total starts at each defensive position over the last three seasons. In Kevin’s words, this approach allows for the “[elimination of] players who had jobs here and there, but are not purely defined as starters—guys in just because of injuries or bad planning—and instead lets us focus solely on guys who have jobs year in and year out.” Both the method and the findings struck me as insightful and fascinating back when I originally read them, and I always figured I’d put together an updated version when the time was right. That time is now.

In essence, I emulated the exact approach described in the original piece right down to the punctuation marks, with the only material difference being the use of 2008-10 statistics, which should better reflect the current composition of the major-league talent pool and run-scoring trends. And as Kevin did in his original piece, I split each 30-player set into four groups—three sets of 10 (Bad, Average, and Good), and then just the top five players at each position (Elite), all sorted by OPS. It’s not anything approaching perfect, but it’s close enough and it’s readily understandable for even the statistical novice. Before jumping into those groupings, however, here is the straight 30-player average at each position from 2008-10, scaled to 600 plate appearances at catcher and 675 plate appearances at every other position:

POS  AB  H  2B 3B HR BB  SO SB CS  AVG  OBP  SLG
C   533 141 29  1 16 52  97  4  2 .264 .334 .414
1B  583 163 35  2 28 79 121  3  2 .279 .368 .492
2B  604 168 34  3 15 55  92 13  4 .278 .342 .419
3B  600 161 34  2 22 61 126  8  3 .268 .338 .440
SS  610 168 31  4 12 49  89 15  6 .276 .332 .400
LF  598 166 33  4 23 63 117 14  5 .277 .350 .459
CF  604 161 32  6 17 58 122 26  7 .267 .335 .424
RF  599 165 34  4 21 65 123 12  5 .276 .349 .450

And what follows now is the exact same OPS chart for all four teams in the AL West, but with each team’s projected starter at a given position being identified by bolded and underlined font, according to which average their 2011 PECOTA forecast strikes the closest to:

Los Angeles Angels of Anaheim

POS   BAD  AVG  GOOD ELITE
C     670  732  834  861
1B    752  849  943  975
2B    691  740  828  853
3B    688  775  858  885
SS    661  725  797  840
LF    735  799  879  911
CF    693  765  811  827
RF    719  801  866  887

Oakland Athletics

POS   BAD  AVG  GOOD ELITE
C     670  732  834  861
1B    752  849  943  975
2B    691  740  828  853
3B    688  775  858  885
SS    661  725  797  840
LF    735  799  879  911
CF    693  765  811  827
RF    719  801  866  887

Seattle Mariners

POS   BAD  AVG  GOOD ELITE
C     670  732  834  861
1B    752  849  943  975
2B    691  740  828  853
3B    688  775  858  885
SS    661  725  797  840
LF    735  799  879  911
CF    693  765  811  827
RF    719  801  866  887

Texas Rangers

POS   BAD  AVG  GOOD ELITE
C     670  732  834  861
1B    752  849  943  975
2B    691  740  828  853
3B    688  775  858  885
SS    661  725  797  840
LF    735  799  879  911
CF    693  765  811  827
RF    719  801  866  887

 I must caution that in spite of the long string of “Bad” marks peppering every other team in the division besides Texas, there were a decent number of players—particularly in Anaheim—that sat in between “Bad” and “Average,” but simply didn’t garner an auspicious enough forecast to get bumped up into the next tier. It’s a very quick and dirty snapshot look at how each AL West offense compares positionally, and I hope to take a more nuanced view of the landscape before Opening Day. For the moment, however, this should give you some sense of the potential for another Rangers offensive feast running concurrent with another Mariners scoring famine in 2011. 

Thank you for reading

This is a free article. If you enjoyed it, consider subscribing to Baseball Prospectus. Subscriptions support ongoing public baseball research and analysis in an increasingly proprietary environment.

Subscribe now
You need to be logged in to comment. Login or Subscribe
brandonwarne52
3/10
Intriguing to say the least. Good stuff.
jhardman
3/10
One quick observation - is the lack of an underlined "elite" on the Texas side (which would represent Hamilton, the AL MVP) because of his position splits between LF and CF, with Borbon bringing the CF numbers down and Murphy bringing down the LF numbers? I was just curious how this breakdown works on the opposite end of the spectrum - finding the "elite" position players.
tbwhite
3/10
It's more likely a result of comparing actual results to PECOTA projections. If an elite season is a top 5 season at a position, those are likely to be composed of career/fluke seasons where the player outperformed his expectations. PECOTA projections regress towards the mean to account for the extra luck that created the "elite" seasons, but it doesn't put that luck back into the system so to speak.

Actually, I suppose that's where the percentile distribution comes in, you can get a feel for best case vs worst case scenarios, but when you distill PECOTA to a single number you remove good luck, and there are simply very few players whose established or "true" level would place them in the top 5 at a position every year.
jdeboer
3/10
Is it possible to get this type of breakdown for each division?
manoadano
3/10
There's something weird going on with the Oakland projections, probably with the park factor. PECOTA is projecting the A's to score their fewest runs since 1979 and allow their fewest runs since 1990. Many of the batters seem to be projected well below their established levels, while many of the pitchers are well above theirs.

Some examples of Oakland batters who outperformed their 2010 projections, yet have worse projections this year than last year (using TAv):

Willingham
2010 Proj: .299
2010 Actual: .311
2011 Proj: .288

Barton
2010 Proj: .277
2010 Actual: .298
2011 Proj: .268

Matsui
2010 Proj: .293
2010 Actual: .294
2011 Proj: .278

Ellis
2010 Proj: .264
2010 Actual: .273
2011 Proj: .243

Crisp
2010 Proj: .270
2010 Actual: .296
2011 Proj: .256

Pennington
2010 Proj: .246
2010 Actual: .263
2011 Proj: .244

Similarly, the pitching projections appear to be overly optimistic. Virtually everyone with a pulse (17 different pitchers) is projected for an ERA below 4.00.
tbwhite
3/10
I think your objective definitions of good and bad should be based not on historical data, but rather the 2011 PECOTA projections. Comparing historical data and PECOTA projections is a bit apples and oranges isn't it ? This historical data are influenced by all of the outliers, the career or fluke seasons, but the PECOTA projections don't/can't account for luck(at least in the mean projection), so you systematically are under-rating everybody for 2011.
conwell
3/12
I noticed that in the case of the Angels, the projections are significantly different than those published in the annual. I compared the annual to the spreadsheet and found significant differences and ones that matter quite a bit in this analysis. (I don't know why the discrepancy although it may have been mentioned somewhere else in the book or on the website). For example, the spreadsheet pegs Kendrick as a .708 OPS while the book has him at .775. Torii Hunter is pegged at .752 in the spreadsheet, but .819 in the annual.

Was there a change in the algorithm between the book and the spreadsheet or are they different numbers. I glanced at a few of the A's and noticed a similar discrepancy. While the relative differences in strengths of the team may remain, I would be curious to know what the ratings of Bad, Average, Good and Elite are based on the 2011 projections. I would be willing to bet that several of the positions above would move up a slot or two.