Baseball for the Thinking Fan

Login | Register | Feedback

btf_logo
You are here > Home > Primate Studies > Discussion
Primate Studies
— Where BTF's Members Investigate the Grand Old Game

Tuesday, June 17, 2003

Return of YEPS: The 2003 YEPS 15

Riding into the Year-End Valley of Death

OK, kids, they finally gave me more than a day off at a time and that means that I could actually focus my eyes sufficiently to manipulate a bunch of tiny numbers in my YEPS (short for Year-End Projection System) spreadsheet.

Prior to that step, of course, was the selection of a group of batters whose performance in 2003 and/or some other element of “intrinsic interest” made them desirable candidates for what has become known as “The YEPS Fifteen” (any resemblance to the “Hollywood Ten,” “Chicago Eight,” “Fab Four”, or “The Magnificent Seven” is not only unlikely, it’s probably illegal).

This year’s list is a mixture of superstar, cult player, the heavily hyped, and the roundly panned. Let’s arrange ‘em into those categories, just for fun:

Superstars: Barry Bonds, Jason Giambi, Sammy Sosa
Cult player: Lance Berkman, Mike Lowell, Greg Myers
Heavily hyped: Adam Dunn, Erubiel Durazo, Jeremy Giambi, Andruw Jones, Aramis Ramirez, Alfonso Soriano, Miguel Tejada
Roundly panned: Garret Anderson, Alex Gonzalez, Aramis Ramirez, Alfonso Soriano.

Yes, a couple of players show up on both lists. No, YEPS doesn’t have a separate projection based on these categories (though it’s tempting).

Here are the stats for these fifteen players (appearing in alphabetical order) through June 12th:

as of 6/12 G AB R H D T HR RBI BB SO SB CS BA OBP SLG OPS
Garret Anderson 63 259 33 80 20 3 13 57 11 32 3 2 .309 .337 .560 .897
Lance Berkman 60 215 47 63 10 1 12 40 40 43 1 2 .293 .404 .516 .920
Barry Bonds 54 175 44 54 9 1 18 41 57 27 5 0 .309 .478 .680 1.158
Adam Dunn 62 206 41 41 4 0 20 43 37 78 4 2 .199 .321 .510 .831
Erubiel Durazo 61 215 36 61 16 0 8 38 36 41 0 0 .284 .386 .470 .856
Jason Giambi 63 227 36 53 8 0 15 42 44 58 1 0 .233 .358 .467 .825
Jeremy Giambi 39 102 13 18 5 0 5 13 22 30 0 0 .176 .323 .373 .695
Alex Gonzalez 59 216 26 72 18 5 10 43 10 32 0 4 .333 .363 .602 .965
Andruw Jones 61 235 42 71 10 1 18 51 26 54 1 1 .302 .372 .583 .955
Mike Lowell 66 259 43 77 19 1 18 51 24 42 2 1 .297 .357 .587 .944
Greg Myers 52 138 24 48 9 0 6 23 21 25 0 1 .348 .434 .543 .977
Aramis Ramirez 62 240 28 72 15 1 3 30 15 38 1 1 .300 .341 .408 .750
Alfonso Soriano 65 289 53 85 11 3 18 43 21 61 18 3 .294 .342 .540 .882
Sammy Sosa 45 162 32 47 13 0 6 27 29 55 0 1 .290 .398 .481 .879
Miguel Tejada 64 258 33 58 16 0 10 37 19 28 2 0 .225 .278 .403 .681

What YEPS does, for those of you who’ve either not been here before or have dutifully forgotten all about it, is to take these stats, combine them with the batter’s previous performance (usually the past two or three years of data), and use the aggregate performance level to produce a year-end projection.

It’s not a simple task to describe all of the steps involved in the projection (though they’ve been outlined several times in the past), but basically the system assumes that a player will perform close to the adjusted performance level for the balance of the year. Players like Jeremy Giambi and Alex Gonzalez, who are noticeably below or above their earlier performance levels, are expected to move in the direction that the earlier stats would suggest as their “true” performance level.

This is best demonstrated by showing the “straight projections” that the season-to-date numbers produce. These numbers assume that the player will keep the exact rate of performance being exhibited through the first ~60 games of the season over the entire year:

Straight Proj G AB R H D T HR RBI BB SO SB CS BA OBP SLG OPS
Garret Anderson 160 659 84 204 51 8 33 145 28 81 8 5 .309 .337 .560 .897
Lance Berkman 159 569 124 167 26 3 32 106 106 114 3 5 .293 .404 .516 .920
Barry Bonds 142 460 116 142 24 3 47 108 150 71 13 0 .309 .478 .680 1.158
Adam Dunn 159 528 105 105 10 0 51 110 95 200 10 5 .199 .321 .510 .831
Erubiel Durazo 149 525 88 149 39 0 20 93 88 100 0 0 .284 .386 .470 .856
Jason Giambi 157 565 90 132 20 0 37 105 110 144 2 0 .233 .358 .467 .825
Jeremy Giambi 108 281 36 50 14 0 14 36 61 83 0 0 .176 .323 .373 .695
Alex Gonzalez 151 551 66 184 46 13 26 110 26 82 0 10 .333 .363 .602 .965
Andruw Jones 160 616 110 186 26 3 47 134 68 142 3 3 .302 .372 .583 .955
Mike Lowell 154 605 100 180 44 2 42 119 56 98 5 2 .297 .357 .587 .944
Greg Myers 109 288 50 100 19 0 13 48 44 52 0 2 .348 .434 .543 .977
Aramis Ramirez 148 575 67 172 36 2 7 72 36 91 2 2 .300 .341 .408 .750
Alfonso Soriano 156 693 127 204 26 7 43 103 50 146 43 7 .294 .342 .540 .882
Sammy Sosa 129 465 92 135 37 0 17 77 83 158 0 3 .290 .398 .481 .879
Miguel Tejada 162 651 83 146 40 0 25 93 48 71 5 0 .225 .278 .403 .681

Looking at these numbers, you’d project that the long-ballyhooed Andruw Jones was finally going to have a season that would make his boosters clench their fists in triumph. On the other hand, Sammy Sosa looks like a major casualty. The Marlins’ Alex Gonzalez would figure to have an unparalleled “from out of nowhere” season, while Jeremy Giambi’s flop would be accompanied by the sounds made by the chronically flatulent.

Maybe the most amazing numbers here, however, are the ones associated with Adam Dunn. Instead of being the next Bonds or Thome, Dunn is looking like the missing link between Rob Deer and Dave Kingman.

YEPS makes a series of modifications to these projections based on elements of past performance, and by comparing the season-to-date performance to it. It’s predicated as much on the shape of the stats as the value of the stats, an area that doesn’t receive much attention from the current generation of number-crunchers.

As a result, YEPS comes up with some always interesting and surprisingly accurate year-end projections. Here’s what YEPS comes up with for this year’s fifteen:

YEPS projection G AB R H D T HR RBI BB SO SB CS BA OBP SLG OPS
Garret Anderson 160 658 85 197 48 5 32 134 27 85 8 5 .299 .327 .533 .860
Lance Berkman 159 575 122 177 35 3 35 117 106 117 5 6 .308 .416 .562 .979
Barry Bonds 142 451 118 144 26 3 50 110 154 71 12 1 .319 .492 .724 1.216
Adam Dunn 152 512 99 112 17 1 44 98 99 183 12 6 .219 .345 .510 .855
Erubiel Durazo 149 478 84 135 34 0 21 88 83 99 0 0 .282 .389 .485 .874
Jason Giambi 157 555 100 151 27 1 39 115 117 126 2 0 .272 .399 .534 .933
Jeremy Giambi 108 285 44 65 15 0 14 42 59 78 0 0 .227 .360 .428 .788
Alex Gonzalez 151 555 64 160 38 9 19 86 24 95 2 7 .288 .318 .493 .811
Andruw Jones 160 617 106 175 28 3 41 117 66 135 7 3 .283 .352 .536 .888
Mike Lowell 154 592 90 168 42 1 34 109 55 93 4 2 .283 .344 .530 .874
Greg Myers 109 284 42 85 15 0 11 43 39 54 0 1 .300 .385 .474 .859
Aramis Ramirez 148 568 68 166 35 2 14 79 36 93 3 2 .293 .335 .433 .768
Alfonso Soriano 156 671 117 193 31 6 38 96 41 144 43 9 .288 .330 .520 .849
Sammy Sosa 129 473 98 142 31 1 31 95 85 144 1 2 .301 .407 .569 .976
Miguel Tejada 162 643 96 165 38 1 29 108 50 80 6 1 .257 .311 .452 .763

You can compare the YEPS numbers to the straight projections and see how the raw numbers and the “shape” of the stats have been altered. We’ll look at several players in a bit more detail below, but basically what you see is that players who are performing above or below their established production levels will move in the direction of those levels, modified somewhat by what they’ve done so far this year. Compare Jason Giambi, Gonzalez, Jones, Mike Lowell, Sosa and Tejada’s YEPS projections to their straight projections and you’ll get a sense of this.

Over at the Big Bad Blog you’ll find some additional notes for some of the players on this list. (There are also a few comments that undoubtedly would not be deemed suitable for “Prime Time.”)

Let’s wrap up by looking at a question we pondered earlier in the year-what kind of season will Barry Bonds have after having the two greatest consecutive offensive seasons in baseball history?

What        G  AB   R   H  D T HR RBI  BB SO SB CS   BA  OBP  SLG   OPS
6/12       54 175  44  54  9 1 18  41  57 27  5  0 .309 .478 .680 1.158
Straight  142 460 116 142 24 3 47 108 150 71 13  0 .309 .478 .680 1.158
YEPS      142 451 118 144 26 3 50 110 154 71 12  1 .319 .492 .724 1.216
BoY           276  74  90 17 2 32  69  97 44  7  1 .326 .501 .750 1.251

YEPS projects Barry to gain ground in the second half, as the balance-of-year (BoY) figures, extracted from the spreadsheet and shown below the overall YEPS projection, indicate. I don’t know if I actually believe that, but there you have it.

So if YEPS is right, Barry will lose about 150 points of OPS; if he stays at his current pace, that figure will be a bit over 200 points. In any case, it’s amazing to think that anyone could decline that much and still be the best hitter in the game.

Don Malcolm Posted: June 17, 2003 at 06:00 AM | 5 comment(s) Login to Bookmark
  Related News:

Reader Comments and Retorts

Go to end of page

Statements posted here are those of our readers and do not represent the BaseballThinkFactory. Names are provided by the poster and are not verified. We ask that posters follow our submission policy. Please report any inappropriate comments.

   1. Jes Golbez Posted: June 17, 2003 at 02:18 AM (#611665)
Good article.
I'd love to see a YEPS projection for Javier Lopez.
He's had just a monster season so far with his HR explosion, and going into his walk year, I wonder what kind of offensive totals we can expect from Javi.
   2. Patriot Posted: June 17, 2003 at 02:19 AM (#611681)
What exactly is the definition of a "cult" player? Is there a Lance Berkman cult somewhere that worships him? He has been one of the best hitters in baseball for the last 3 years.
   3. Dan 'The Boy' Werr Posted: June 17, 2003 at 02:19 AM (#611682)
I was surprised that Bonds has about a 20-point lead over Pujols in EQA. Bonds an advantage of ~50 points in OBP, which I know is not trivial, but Pujols has 42 more total bases so far. Is Bonds' walks that much more valuable than Pujols' total bases, or does EQA overvalue OBP?

Pujols's extra total bases are largely the result of 54 extra AB. There's only a 21-point difference in their slugging, much smaller than the OBP difference. EQA is a rate stat, so playing time is not a consideration, nor is Bonds' 29-4 edge in IBB, which reduces his ability to rack up total bases.

When you add in the park adjustment, which will accentuate the difference between Bonds and Pujols, I don't think it's unreasonable at all.
   4. Patriot Posted: June 17, 2003 at 02:19 AM (#611687)
Actually, EQA does slightly overestimate the value of a walk. EQR is ESSENTIALLY a LW formula, and the LW it gives for a walk based on 1980-2000, is .36, whereas I expect that BsR would give more like .33.

This is not to say that Bonds does not deserve to rate ahead of Pujols, because he does, but the walk is slightly overvalued.
   5. Steve Posted: June 20, 2003 at 02:20 AM (#611767)
What does YEPS have to say about Paul Konerko?

You must be Registered and Logged In to post comments.

 

 

<< Back to main

BBTF Partner

Support BBTF

donate

Thanks to
Jim Wisinski
for his generous support.

Bookmarks

You must be logged in to view your Bookmarks.

Syndicate

Page rendered in 0.2882 seconds
47 querie(s) executed