Announcement

Collapse
No announcement yet.

Swinging For the Fences

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Swinging For the Fences

    Swinging for the Fences
    Does Missing Matter?

    by Will Carroll and Mike Carminati

    In 1999, Sammy Sosa and Mark McGwire battled for the National League home run title, just one year after their record-setting chase that ended with both breaking Roger Maris’ single season home run record. On September 18, 1999, Sosa drilled a Jason Bere offering over the ivy at Wrigley for his sixtieth home run of the season, becoming the first man to hit sixty roundtrippers in two separate seasons. McGwire became the second man to accomplish this feat eight days later and again beat Sosa out for the home run crown, 65 to 63. Though both again bested Maris' old record, their 1999 campaign is now a footnote to the history of the home run chase that re-energized baseball in the late 90’s.

    Just as an out-of-the-blue bolt of plate discipline presaged Sosa’s assent, his decline might have been predicted by his tendency to swing and miss that haunted him even in his stellar 1999 season. Sosa swung at and missed 475 pitches in his record-setting 1999 campaign. This is the highest total for any major-league batter over the last five seasons and isn’t the “swing and a miss!” call of the announcer the cruelest fate in baseball? But what does it mean in the greater scheme?

    Does having a tendency to swing and miss more than most impair a batter's productivity as we have been told since Little League? Do batters with better batting eyes tend to be more productive than the average batter? Is it better to be patient at the plate or go for the first pitch you can hit? Does this data tell us anything new and could that be used to help build a better team or find successful players?

    First, let's take at the Sosa-inspired SAM (Swing And Miss) batter. As you can see from this list of the leaders in swinging strikes over the past five seasons, the concept is aptly named:

    Code:
                 Swinging
    Player     Year   Strikes  Home Runs Strikeouts
    ---------------------------------------------------------
    Sammy Sosa   1999    475    63    171
    Sammy Sosa   2001    431    64    153
    Sammy Sosa   2000    422    50    168
    Alfonso Soriano 2002    406    39    157
    Mo Vaughn    2000    405    36    181
    Jim Thome    2003    403    47    182
    Preston Wilson 2000    394    31    187
    Richie Sexson  2001    389    45    178
    Jim Thome    2000    387    37    171
    Sammy Sosa   2002    385    49    144
    Dean Palmer   1999    383    38    153
    Jacque Jones  2002    374    27    129
    Richie Sexson  2000    372    30    159
    Craig Wilson  2004    371    29    169
    Jose Hernandez 2002    365    24    188
    If this is any indication, then a large number of swinging strikes does not appear to be a detriment to one's ability to produce. All of the batters listed had some degree of offensive success--the lowest home run total is 24. However, they struck out a ton. Does power hitting require this type of all or nothing approach?

    Perhaps a glimpse at the other end of the spectrum will shed some light. Here are the batters who collected the lowest number of swinging strikes while batting enough to qualify for a batting title:

    Code:
                 Swinging
    Player      Year  Strikes Home Runs Strikeouts
    --------------------------------------------------------
    Brian Roberts   2003    54    5    58
    Placido Polanco  2003    57    14    38
    Juan Pierre    2003    58    1    35
    David Eckstein  2003    59    3    45
    Juan Pierre    2001    63    2    29
    Juan Pierre    2004    63    3    35
    Placido Polanco  2004    63    17    39
    Eric Young    1999    64    2    26
    Luis Castillo   2001    67    2    90
    Mark Grace    2001    68    15    36
    Jason Kendall   2002    68    3    29
    Luis Castillo   2003    70    6    60
    Fernando Vina   2002    71    1    36
    Scott Hatteberg  2003    71    12    53
    Scott Hatteberg  2004    72    15    48
    Mark Grace    2000    73    11    28
    Brian Giles    2002    74    38    74
    There is not a lot of power there. You should note the exception is Brian Giles and his 38 dingers in 2002. Clearly, there seems to be some relationship between high swinging strike totals and high home run totals, but it is far from absolute.

    Let's take a look at a few examples to see if swinging strikes have plagued those batters' careers:
    Code:
    Sammy Sosa:
    
    
         Strikes   Total
    Yr    Swinging  Pitches    HR   Strikeouts   BA    OPS
    -------------------------------------------------------------------------
    1999    475    2889    63    171    .288    1.002
    2000    422    2956    50    168    .320    1.040
    2001    431    2867    64    153    .328    1.174
    2002    385    2746    49    144    .288     .993
    2003    349    2363    40    143    .279     .911
    2004    319    2158    35    133    .253     .849
    
    Adam Dunn, who set the new single-season strikeout record last year:
    
    
         Strikes   Total
    Yr    Swinging  Pitches    HR   Strikeouts   BA    OPS
    -------------------------------------------------------------------------
    2001    120    1185    19     74    .262    .948
    2002    300    2945    26    170    .249    .854
    2003    213    2032    27    126    .215    .819
    2004    301    2895    46    195    .266    .956
    
    Jose Hernandez, who came close to breaking the record a couple of times:
    
    
         Strikes   Total
    Yr    Swinging  Pitches    HR   Strikeouts   BA    OPS
    -------------------------------------------------------------------------
    1999    261    2311    19    145    .266    .764
    2000    240    2005    11    125    .244    .687
    2001    348    2367    25    185    .249    .743
    2002    365    2384    24    188    .288    .834
    2003    343    2253    13    177    .225    .634
    2004    117     930    13     61    .289    .910
    
    Barry Bonds, who has had a remarkably low number of swinging strikes especially considering his home run totals:
    
    
         Strikes   Total
    Yr    Swinging  Pitches    HR   Strikeouts   BA    OPS
    -------------------------------------------------------------------------
    1999    111    1700    34    62     .262    1.006
    2000    141    2381    49    77     .306    1.127
    2001    141    2678    73    93     .328    1.379
    2002    104    2397    46    47     .370    1.381
    2003    123    2185    45    58     .341    1.278
    2004     85    2425    45    41     .362    1.422
    Do you have to swing big to connect big? There seems to be a bias towards the “three true outcomes”--walk, strikeout, homerun--that typify the careers of Rob Deer and Dave Kingman. The insights from the MLB.com database are plentiful. At the heart of this, what we want to know is how often a batter sees a hittable pitch but misses. Total number of pitches does not capture that. Certain types of pitches such as called balls must be ignored--credit cannot be given for not swinging at those.

    We must first define what is and is not a hittable pitch. We can say with confidence that every pitch that a batter swung at, successfully or unsuccessfully, was hittable. Watching too many games with Vladimir Guerrero or Alfonso Soriano can shake this assumption. SAM will then be the number of times a batter swung and missed (including bunts) divided by the hittable pitches that a batter saw: called strikes, swinging strikes, fouls, foul tips, and balls in play (including bunt attempts). Balls called, intentional balls, hit by a pitch, and all pitchout attempts are not considered hittable. The assumption is that a batter is never required to swing at a ball.

    Now, here are the highest SAM totals for all batters who qualified for the batting title, 1999-2004:
    Code:
    Name        Yr  SAM  HR  SO  BB  BA  OBP  SLUG  OPS
    -------------------------------------------------------------------------
    Sammy Sosa    1999 .283  63  171  78 .288 .367  .635  1.002
    Sammy Sosa    2001 .272  64  153  116 .328 .437  .737  1.174
    Jacque Jones   2002 .254  27  129  37 .300 .341  .511  .852
    Dean Palmer   1999 .252  38  153  57 .263 .339  .518  .857
    Richie Sexson  2000 .251  30  159  59 .272 .349  .499  .848
    Jose Hernandez  2002 .249  24  188  52 .288 .356  .478  .834
    Andres Galarraga 2000 .249  28  126  36 .302 .369  .526  .895
    Sammy Sosa    2000 .249  50  168  91 .320 .406  .634  1.040
    Jim Thome    2003 .248  47  182  111 .266 .385  .573  .958
    Greg Vaughn   1999 .248  45  137  85 .245 .347  .535  .881
    Sammy Sosa    2004 .248  35  133  56 .253 .332  .517  .849
    Sammy Sosa    2002 .247  49  144  103 .288 .399  .594  .993
    Sammy Sosa    2003 .247  40  143  62 .279 .358  .553  .911
    Craig Wilson   2004 .247  29  169  50 .264 .354  .499  .853
    Mo Vaughn    2000 .245  36  181  79 .272 .365  .498  .864
    Jeromy Burnitz  2004 .244  37  124  58 .283 .356  .559  .916
    Jose Hernandez  2003 .243  13  177  46 .225 .287  .347  .634
    Richie Sexson  2001 .239  45  178  60 .271 .342  .547  .889
    Jim Thome    2000 .239  37  171  118 .269 .398  .531  .929
    Mo Vaughn    1999 .238  33  127  54 .281 .358  .508  .866
    Geoff Jenkins  2003 .237  28  120  58 .296 .375  .538  .913
    Greg Vaughn   2001 .235  24  130  71 .233 .333  .433  .766
    Preston Wilson  2000 .234  31  187  55 .264 .331  .486  .817
    Mark McGwire   1999 .233  65  141  133 .278 .424  .697  1.120
    Jose Hernandez  2001 .232  25  185  39 .249 .300  .443  .743
    Jacque Jones   2004 .232  24  117  40 .254 .315  .427  .742
    Vinny Castilla  2001 .232  25  108  35 .260 .308  .467  .775
    SAMmy, indeed! Sosa himself is the archetype for the free-swinging, powerful slugger (or he was before 2005...). Only Jose Hernandez really defies the typecasting, though Jacque Jones is no lumbering basher. The low of 13 stands out due to the 30, 40 and even 50 and 60 home run seasons represented. Perhaps it is more telling to note that the OPS is skewed more towards the exceptional.

    Now here are the lowest:
    Code:
    Name        Yr  SAM  HR  SO  BB  BA  OBP  SLUG  OPS
    Juan Pierre   2003 .035   1  35  55  .305 .361  .373  .734
    Juan Pierre   2004 .036   3  35  45  .326 .374  .407  .781
    Brian Roberts  2003 .043   5  58  46  .270 .337  .367  .704
    Luis Castillo  2001 .044   2  90  67  .263 .344  .341  .685
    Juan Pierre   2001 .044   2  29  41  .327 .378  .415  .793
    Jason Kendall  2004 .044   3  41  60  .319 .399  .390  .789
    David Eckstein  2003 .045   3  45  36  .252 .325  .325  .651
    Luis Castillo  2003 .045   6  60  63  .314 .381  .397  .778
    David Eckstein  2004 .048   2  49  42  .276 .339  .332  .671
    Jason Kendall  2002 .048   3  29  49  .283 .350  .356  .706
    Chuck Knoblauch 1999 .048  18  57  83  .292 .393  .454  .848
    Omar Vizquel   1999 .048   5  50  65  .333 .397  .436  .833
    Placido Polanco 2003 .049  14  38  42  .289 .352  .447  .799
    Scott Hatteberg 2004 .049  15  48  72  .284 .367  .420  .787
    Luis Castillo  2004 .049   2  68  75  .291 .373  .348  .720
    David Eckstein  2002 .049   8  44  45  .293 .363  .388  .752
    David Eckstein  2001 .049   4  60  43  .285 .355  .357  .712
    Scott Hatteberg 2003 .049  12  53  66  .253 .342  .383  .725
    Fernando Vina  2002 .050   1  36  44  .270 .333  .338  .670
    Jason Kendall  2003 .050   6  40  49  .325 .399  .416  .815
    Placido Polanco 2004 .050  17  39  27  .298 .345  .441  .786
    Paul Lo Duca   2002 .051  10  31  34  .281 .330  .402  .731
    Luis Castillo  2000 .052   2  86  78  .334 .418  .388  .806
    Omar Vizquel   2002 .053  14  64  56  .275 .341  .418  .759
    Eric Young    2000 .053   6  39  63  .297 .367  .399  .766
    Mark Loretta   2004 .053  16  45  58  .335 .391  .495  .886
    Luis Castillo  2002 .053   2  76  55  .305 .364  .361  .726
    Maybe we should have called it the Juan! As a group, the high SAM batters are clearly more valuable. Again, the highest OPS on the low SAM groups barely make register when compared to the high SAM group.

    Using the data for all batters who met the batting title eligibility requirements, the data was checked for correlations between SAM and a number of conventional baseball stats. The only thing that seems to be affected by the free-swinging high SAM scores are strikeouts. This is not much of a news flash.

    The surprising part is that it does not seem to lead to more conclusions. A high SAM does seem to lead to more home runs and slightly more power. However, one would expect that free swingers walk less frequently. That's the conventional wisdom, but the facts don't bear that out. Actually, walks per plate appearance ever so slightly tend to increase with a higher SAM. One would also expect that all the hacking would lead to fewer appearances on the basepaths. Again, it just isn't so. On-Base Percentages are barely affected.

    When someone swings so freely, does it affect the kinds of pitches he sees? Does he get fewer hittable pitches? Hittable pitches in the strike zone? Does he end up doing the pitcher a favor by going down on strikes on fewer pitches? Are free swingers continually behind in the count, are at least more so than the average batter? Do young batters swing more freely than veterans?

    Actually, the answer to all these questions is "no." None of the associated statistics has anything whatsoever to do with SAM.

    In the end, high SAM scores do not really help us to define a hitter. Yes, they strike out a lot. Yes, the big swings do lead to slightly more power. But aside from those factors, free swingers are not a homogeneous group. Some run deep counts. Some don't. Some get on base often. Some draw walks more often. Some get pitched around more often. But in all these cases, some don't. Some are young. Some are old. Also, however you want to evaluate batters overall, there's no way to say that high SAMs lead to more or less effective batters.

    Again, conventional wisdom dictates that batters who swing and miss more often are a drag on the batting order. However, nothing seems to indicate that. This begs a question. A favorite stathead of ours asks “If we know that a high SAM rate does not necessarily lead to a low walk rate, what does that mean about these players? Is it their superior batting eye that enables them to cut loose more often? Or does it suggest that if they took shorter, more controlled swings, they might miss less and maybe hit for a higher average, but at the expense of the slugging ability that is the key to their offensive value?” It’s something we hope we can answer soon.

    But perhaps we're being too hasty. SAM may not equate to certain characteristics in analysis of all players. However, for an individual player, could a change in SAM portend a change in other characteristics for that player? If he swings and misses more often, does his on-base average drop? If he cuts down on his SAM, do his homers go up?

    We looked again to the data, comparing the statistics of all of the players who batted enough to qualify for a batting title over the last five years from one season to the next. For each of these two-year pairings, the average player's SAM went from .132 in year one to .130 in year two. As a whole, SAM had very little variance from year to year, supporting our fledgling stat SAM as a true measure of a consistent characteristic of a player.

    Given that, what are the largest one-year swings and what has that meant for the given player's performance? Let's take a look:

    Code:
                Prev Yr
    Yr   Name        SAM  SAM  SAM Diff Prev Yr OPS  OPS  OPS Diff
    2002  Alfonso Soriano  .156  .220   .064   .736    .880   .143
    2004  Ron Belliard   .076  .138   .062   .760    .774   .014
    2001  Andruw Jones   .144  .199   .055   .907    .772  -.135
    2002  Jim Edmonds    .139  .194   .054   .974    .981   .007
    2002  Jacque Jones   .199  .254   .054   .751    .852   .101
    2002  Corey Koskie   .155  .203   .049   .850    .815  -.035
    2004  Bret Boone    .141  .188   .047   .902    .740  -.162
    2000  Vladimir Guerrero .163  .208   .044   .978   1.074   .096
    2004  Jose Valentin   .154  .197   .043   .776    .760  -.016
    2003  Jim Thome     .205  .248   .043   1.122    .958  -.164
    2002  Aramis Ramirez  .147  .187   .040   .885    .666  -.219
    2001  Manny Ramirez   .165  .202   .036   1.154   1.014  -.140
    2004  Adam Kennedy   .111  .146   .035   .743    .757   .014
    2004  Carlos Pena    .174  .206   .033   .772    .810   .038
    2001  Jeromy Burnitz  .184  .216   .032   .811    .851   .039
    2000  Richie Sexson   .219  .251   .032   .818   .848   .030
    
    Now here are the largest one-year declines in SAM:
    
    
                Prev Yr
    Yr   Name        SAM  SAM  SAM Diff Prev Yr OPS  OPS  OPS Diff
    2002  Manny Ramirez   .202  .117  -.085   1.014   1.097   .083
    2000  Damion Easley   .180  .115  -.065   .779    .766  -.013
    2002  Vinny Castilla  .232  .176  -.056   .775    .616  -.158
    2003  Corey Koskie   .203  .150  -.053   .815    .845   .030
    2004  Shea Hillenbrand .118  .070  -.048   .782    .812   .030
    2000  Dean Palmer    .252  .204  -.048   .857    .809  -.048
    2004  Pat Burrell    .186  .139  -.047   .713    .821   .107
    2000  Carlos Delgado  .192  .146  -.046   .948   1.134   .186
    2000  Juan Encarnacion .206  .161  -.046   .736    .764   .027
    2001  Jason Kendall   .109  .063  -.045   .882    .693  -.189
    2002  Scott Spiezio   .125  .081  -.044   .764    .807   .044
    2001  Steve Finley   .132  .089  -.043   .904    .767  -.137
    2002  Albert Pujols   .149  .108  -.041   1.013    .955  -.058
    2003  Derrek Lee    .189  .149  -.040   .872    .888   .016
    2000  Alex Rodriguez  .195  .155  -.040   .943   1.026   .083
    2004  Brad Wilkerson  .173  .133  -.040   .844    .872   .028
    2004  Mark Kotsay    .137  .098  -.040   .726    .829   .102
    Then a change in a batter's SAM doesn't correspond to a change in his value as a batter, if OPS is any indication--that is, at least not for the extremes. What about the relationship overall for all players?

    The verdict is that OPS has nothing whatsoever to do with changes in a player's SAM. Neither do OBP, Slugging percentage, isolated power, batting average, walks per plate appearance, home runs per plate appearance, pitches per plate appearance, hittable pitches per plate appearance, or balls per plate appearance. Strikeouts per plate appearance were just about the only stat that increased with an increase in SAM.

    The next time that the announcer’s voice drips with disgust at a swinging strike, remember that overall, the big whiffs are meaningless. It won’t help when that last strike ends the game, leaving the bases loaded in the ninth, but it might end the bias against the free swingers that baseball’s held for decades. A strikeout, it turns out, is just another out. Take your cuts, men.

  • #2
    Would you rather read what these guys wrote, or a random post by me?
    No hedging.
    v


    Comment


    • #3
      Originally posted by kjoe@Sep 5 2005, 01:14 PM
      Would you rather read what these guys wrote, or a random post by me?
      No hedging.
      Considering I pay to read what these guys write, and no slight against you kjoe, but I wouldn't do the same for one of your random posts. The answer should be rather obvious.

      Comment


      • #4
        Originally posted by Fishbone@Sep 5 2005, 12:34 PM
        A strikeout, it turns out, is just another out. Take your cuts, men.
        I'm not sure that's the conclusion they reached here. What the numbers say is that guys who swing and miss a lot can still be productive, and guys who make a ton of contact aren't necessarily productive. But that's entirely different from arriving at the conclusion that a strikeout is "just another out."

        Interesting stuff, but they put the wrong summation on it, if you ask me.

        Comment

        Working...
        X