01 baseball analyst 1982 06

8/3/2019 01 Baseball Analyst 1982 06

1/20


2/20


3/20

Ballpark Effects

!l.!as ahle to compile 10 team/years warth of data. This art icle"stJr.1marizes my findinlJs on National League ballparks for inf ie lderrors and double plays. The s ta t i s t i cs compiled are from thefollowing yearsl 1. For the years 1973, 1 9 7 6 ~ 1979 and 1980I compiled comparative data for each team, home and away, andfbI' i t s ' o p p b n e n t ~ p l a y i n ~ in the original team's park- and athome; 2. For the years 1972 and 1978 I compiled comparatives t a t s . for only fbr each t e a ~ ' s performance at home and on theroarl. For example, for 1973 T have the Cubs' performaoce:atWrigley Field compared to on the road, and the performaoceC u b ~ ' opponents at Wiigley Field compared to the t o t a l s " n f : C ~ h s ' opponents playing at home in Cubs' road games. For 1t;172, Ihave only the Cuhs at ~ t i g l e y Field compared to the Cubs enthe roari.

The s ta t i s t i cs which are compiled here are inf ie ld er rorS ;outf ie ld errors ( t h ~ r e s u 1 t s ' o f which s ~ e m i n t b n c 1 u ~ i v e ' a n d not worth much further discussion) and double plays. Thenumbers used for comparative purposes are inf ie ld ei rors , outfiEdErrors and double plays per thousand plays. Plays are figuredby multiplying Innings Pitched times three and s u h t r a c t ~ n g s t r ikeouts . The next step in m a ~ i n g the compariso0 o performanceis to divide the "home-park" s ta t i s t i cs by t h e " r o a d - p a r ~ ~ s t a t s .The resul t i s a ranking of ballparks by the i r amount o.f yariat ionfrom the "average" of a l l other ballparks other than t h ~ ~ o m e park. The idea is to measure the difference in production,for example, of inf ie ld errors in Wrigley Field in games betweenthe Cubs and thei r opposents, as opposed to the p r o d u c i i o ~ ofinf ie ld erros in Cubs' games on the road.

-. 'i .......

"'

... :: ..,


4/20

I will again use my favorite ba11team. the Cubs, as anexample of how I produced the comparative s ta t i s t i cs ;10Ye ar I ( I P X 3 )Team Totals K DP EOF ((IPX3)-K)LUrig1eyRoad

: : 53017717

2191621740:: 29.91

41994346

806745

Road EIF1000

M14

484:: 17394

124123

17717

= 27.8329.9127.83

The resul t of the comparison is that the probabil i ty ofan inf ield error .in a Cubs v. Opponent game is about 7.5% greateri f the game is in Wrigley Field than in the " ~ ~ e r a g e opponent'shome stadium." The complete f inal data follows:Team

CHI*mON-JMON-ONYPHILPITTSLATLCINHOULAso"llSF-ASF -tJ

EIF29.9132.1727.4927.9124.0828,4327.6134.q123.9027.0628.2932.3R35.1635.32

Home

7.007.928.647.566.028.896.928.036.165.145.056.077.2610.03

OP45.4947.4438.4941.3940.8840.3046.3543.2838.5841. 3942.0740.6038.0339.56

(per 1000)

27.8331.5731.0629.1627.9228.3329,8429.7329.3028.9429.8230.0934.4829.95

RoadEOF

7.07-6.157.095.817.448.166.687.406.197.476.227.068.146.28

DP

42.8343.4239.6242.7043.8345.0542.0139.3343.3240.5141.6544.2739.7138.72

* Stats from 1972, 1973 and 1976 are for old Parc Jarry;the MON-O s ta t s represent those for Olympic Stadium. years1978, 1979. 1980.Candlestick Park h ~ d an as t rotur f surface in 1972 and 1973,before the ci ty fathers decided to restore the natural surface.

4

:: 1.07


5/20

Ballpark Effects

The remaining s ta t i s t ics will be l is ted in order of rank.For purposes of the ast ro/natural argument, an asterisk will

" .

follow the astroturf park figures. The comparative figuresfollollJ: Home IAway Rat iosRank ,Park ErF Park EOF Park OP

1 crN .816* HOU .688* 5L 1.103*'2 PHIL .862* PHIL .809* ATL 1.1003 ffON-O .884* LA .B12 mON-J 1.0934 SL .925* SO .860 CHI 1.0625 HOU .935* SF-A .892* SF -N 1.0226 l.A .'=349 mON-O .937* HoU 1. 022*7 NY .957 CHI .990 LA 1.0108 PITT 1.004* cnJ .995* mON-O .971*9 MN-J '1. 019 SL 1. 036* NY .96910 SF-A 1. 020* ATL 1.OB5 SF-A .95B*11 CHI 1.075 PITT 1.089* PHIL .933*12 SO 1.076 mN-J 1.288 SO .91713 ATL 1.174 NY 1. 301 PITT .895*14 SF -N l . l7g SF -N 1.597 C ~ J " .891*'

l}Avera'1es:Natural 1.061 1.133 1.025Astroturf .921 . 921 .968

The averages are simply the means of the figures in the abovetables and are not weightpn by ei ther plays. or, in the casesof montreal and San Francisco, by the number of team/years ofdata. They are put out simply for purposes of discussion, andthe extra work of going beyond the simple averages did not seemworth the f for t considering the fact that they are not beingused for any purposes other than approximation for discussionpurposes.DISCUSSION-CONCLUSIONS:

The most obvious conclusion is that the statement aboutfielding percentages being higher on astroturf is clearlyjus t i f ied by the evidence. However, i t is also apparent thatthere is some substantial variation among the ballparks in thedifferent classes. Atlanta and San Francisco appear to be thebanes of National League inf ielders , and any i n f i ~ l d e r who

5


6/20

. . . . . . . ..:1 a l l ["lark C::ffects

expects to win a gold Q10ve while paying half of his gamesin e i ther place. I}Jhether just i . f ied or not, inf ie ld gold glovestend to he Bwardad mainly on f ielding percentage (other factorsseem to include hi t t ing, range, and o ~ e r a l l subject ive evaluat ion).I can remember no Giant or Brave in f ie lder winning a gold gloveor Sporting News Fielding award in the past ten years, in fact ,~ a v e Johnson went from perennial American Gold Glove second basemanin al . imoreto a 30 error man in Atlanta, resul t ing in his endinQ his careeras a f i r s t baseman (although his deteriorRting range was probablyas important a factor in tha t move).

Although the difference between natural and astroll lrffor outf ield errors is l?rger than the difference for inf ielderrors , I am less ready to make the obvinus conclusions thatthe tu r f i t s e l f is responsible for the difference. The absolutenumbers of errors are smaller for the outf ie ld errors , so thedeviation is actually less signif icant . IntUit ively, thetrueness of a bounce seems less s ign i f ican t af ter the ba l lhag t ravel led to the outf ie ld . More l ike ly , the differencein surface comes into play more on throws, since i t is usuallythe player making the throw wh6 is charged with the error whenthE'! hn] 1 takes C3. bad hop. For It!hat it 's wort.h, it's also t ruethat fiplding percentaQe dOGS not corre l la te well with theC3.lUard ing of gal d n loves for out f ie 1ders-Lui tness Dave Parker.

Finally, the differences noted in double plays are smaller ,leadinQ to the l ikely conclusion that ,whatever the effect thatastroturf has on the production of double plays, the effect isl e ~ s s i g ~ i f i c a n t then i t is for the production of errnrs . Thefact that the second baseman and shortstop must play deeper onas t ro tur f Luill prohah 1 y tenr! t C1 reduce the pr obab i l ty of a

6


7/20

Flallpark Effects

double play; on the other hand, the fact that the ground bal lt ravels fas ter will tend to increase the probabil i ty of adouble play. These offsett ing factors seem to about cancel eachei ther out . The size of the difference in average and thedis t r ibut ion of resu l ts is such that I am not willing to drawany conclusions about the net effects based on th is amount ofdata.

~ a l 1 p a r k effects have been shown to be s ignif icant onhi t t e r s ' s ta t i s t i cs ; i t is sensible to expect that the sameis true for fielding s ta t i s t i cs . Certainly, playing surfacEsare one important difference a m o n ~ the 'var ious bal lparks .Others may also he important , fof instances, the swirlingwinds at Candlestick Park may be as responsible for producinoinf ie ld errors as the grQund ( reputed to be fraught with tinypphhlES ann clay clumps), although dist inguishing among thecausal factnrs is probabJy impossible with only numbers a v a i 1 3 h l ~ . Subjective comments from readers k n o w l ~ d g a b l e about your lncalb ~ l ] p a r k would be appreciated, i f there are any readers nf th ismaries tart cle. Should thE Analyst ' ~ e t off thF. ground, r hopeto have .!\mer ican League ba l l parI- rlata summar i ze ri for thE m ~ x tr:dition.

7


8/20

THE DISTRIBUTION OF RUNS SCORED( .by Dallas Adams ..

How often will a team be shutout? How often will i t score exactlyone run? Or ten runs, or twenty? What, in short , is the distributionof runs scored in a game? Obviously the distr ibut ion wil l vary betweenhigh scoring and low scoring teams; and i t will vary, partly due to the"shortness" of the 162 game season, between equivalently scoring teams,although this difference will be less than that between high and lowscoring teams .The approach uti l ized for this study was to group together equivalently scoring teams in order to determine, for each group, an averagedistribution of runs scored. Since 1968, the Official Baseball Guidehas l i s ted, for each team, the -scores of a l l that team's games theprevious season. This furnishes ten seasons, 1967 through 1976, ofscoring data; a to tal of 232 team-seasons.First9 each team was grouped with equivalently scoring teams. thegrouping being made on the basis of average number of runs scored pergame. In al l , there were 11 different groups; each having i t s owndist inct range of average runs per game, as shown in Table 1. 'TABLE 1

Group Range ot Number of Group averagenumber Average Runs teams in runs per gamei I per game for the groupthe seasonI 1 2.75 - 2.99 6 2 . 912I 2 3.00 - 3.24 14 3.183! 3 3.25 - 3.49 23 3.3714 3.50 - 3.74 26 3 .612i 5 3.75 - 3.99 49 3.864I 6 4 .. 00 - 4.24 40 4.122 7 4.25 - 4.49 34 4.3488 4 .. 50 - 4.74 24 4 . 5999 4 .. 75 - 4.99 13 4.86710 5.00 - 5.24 2 5.15711 5.25 - 5.49 1 5.290

For example, the 1967 Cincinnati Reds averaged3.73 runs per game t hencethey belong to Group 4.The next step was to tabulate, from the Guides, the number of timeseach team scored each specific number of runs. Once th is was completedfor a l l teams, the totals within each group Were converted to percentsThat is , for each group the percentage of times i t was shutout wascomputed, likewise the percentage of times i t scored exactly one run,etc. These percentages are shown in Table 2; this table presents thecomplete grouped information and forms the basic data set tor theremainder of th is study .The inherent behavior of the basic data can be easily seen on Figur1 which presents for low scoring, average scoring and high scoring teathe probabil i t ies of such teams scoring an exact number of runs. Notethat the solid l ine on this figure represents the average run scoring

8


9/20

EXACT GROUPNUMBER . 1OF' RUNS0 14.521 18.912 17.033 16.414 10.97

5 9.096 5.027 2.728 1.789 1 . 3610 1.2511 0.3112 0.3113 0.2114 0.0015 0.0016 0.1017 0.0018 0.0019, 0.00more

T.HE DISTRIBUTION OF RUNS SCOREDTABLE 2

THE PERCENTAGE OF GAMES IN WHICH EXACTLYTHE GIVEN NUMBER OF RUNS ARE SCORED(ARRANGED BY GROUPS)GROUP GROUP GROUP GROUP GROUP GROUP GROUP GROUP

2 3 4 5 6 7 8 9.11.42 9.79 9.98 8.11 7.21 6 . 49 5.93 4 . 7516 .. 90 15 . 35 13.72 12.90 11.54 10.32 10.00 7 .6017 . 64 17 . 79 16.28 14.53 14.29 13.37 12 . 10 12.1115.81 16.79 15.13 15.73 14.69 14.58 13.34 14.6812.22 12.99 12.28 13 . 78 13.93 13.39 13.11 13.219.41 9.22 10.61 11.05 11.09 11.26 11.66 11.786.13 5 . 72 8 . 00 7.82 8.40 9.15 9 . 66 9 . 644.95 4 . 72 5.12 5.76 6 . 34 7 . 00 7.69 7 . 552.15 3.53 3.85 3.72 4.02 5.04 5.49 5.651.53 1.68 1.94 2.54 3.14 3.34 3.68 3.520.83 0.92 1.27 1 . 58 2.21 2 . 21 2.36 3.520.53 0.81 0.86 0.95 1.25 1 . 43 1.81 2.190.22 0.33 0.36 0.58 0.68 1.15 1.32 1.240.18 0.16 0.24 0.35 0.48 0 . 53 0.73 0.710.00 0.08 0.17 0.25 0.23 0.31 0.54 0.95. 0.04 0.08 0.05 0.11 0.20 0.18 .0.21 0.430.04 0.03 0.02 0.13 0.09 0.05 0.18 0.050.00 0.00 0 . 02 0.05 0.11 0 . 04 0.08 0.10OoOOi 0.00 0.05 0 . 01 0.05 0.05 0.08 0.190.00 0.00 0.05 0.04 0.05 0.09 0.05 0.14

I ,FIGURE 1

GROUP GROUP10 11

4 . 01 4 .. 947.41 8.0211.42 6.1713.58 11.1111.11 17.9012 . 04 12.3510 .80 9.268 . 95 6 . 175 . 86 4 . 943.40 7.41

2.78 2.473.70 3.701.85 1.851.54 2.470.31 1.240 . 31 0.000 . 62 0.000.00 0.000.31 0.000.00 0.00

PROBABILITYOF SCORING EXACTLY A GIVEN NUMBER OF RUNS(COMPOSITE MAJOR LEAGUE DATA, 1967-1976).20 _. .....- - - ' -- ' , ... !" --- ! _ ~ = l t e a m ~ ~ ~ O r 1 ~ - less ithanJ/ 7 " , ::! 3.74 runs per game.16 - 7--'; : -- . - r - . I- a l l teams I

PI........ \ tI' ..R .".- i .............. teams scoring more thano .12 7 ' 4.25 runs per gameB tIA I IB iiiI .08 - :L /ITY 04 ~ - - - ! - - - _ t _ _ - - t - ..- . . .. .-+-----1

.ooL-__ 1 - ~ ~ __ __ __~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ o 2 4 6 . 8 10 12 16 18 20EXACT NUMBER OF RUNS9


10/20

"THE DISTRIBUTION OF RUNS SCOREDdil t r ibut ion for a l l major league teams in the ten year period 1967through 1976. On Figure I t the abscissa represents the exact number of""runs in a game, while the ordinate represents the probability of scoringexactly that number of runs. .The general trends are apparent. As scoring increases (scoring beingmeasured by the average number of runs scored per game over a ful l seasoor other period of t ime), the peak percentage decreases and shif ts to ther ight . After the peak, the tai l ing-off portion of the curves show thatthe lower ~ c o r i n g the team, the faster the t a i l -of f .The regular behavior of the curves makes i t a simple matter to obtainempirical equations expressing, for a team with a known average rate ofscoring (R/G), the probability of occurance of each individual run to ta l .I f N i s the number of runs and PN i s the probabil i ty of scoring exactly Nruns in a game, then:

( l) When N is S runs or fewer: PN =A + B(N) + C(N)2where A = 38498 - .10839 (R/G) + .00800SS(R/G)2B = .0010306 + .024139 (R/G) - .002943(R/G)2C = -.01879 + .002Sl4(R/G) - .0000lS06(R/G)2(2) When N is 6 runs or more: PN : D(EN)

where D =6.6210 - 2.496S(R/G) + .27Sl8(R/G)2E : .058479 + .24022(R/G) - .02229l(R/G)2As mentioned, R/G i s the average number of runs scored per game over tht.,ful l season or other period of time.Possible applications of the runs scoring distribution described abovinclude:

( l )(2 )

The determination of whether a given pitcher on a team receivedabove average or below average batting support.The determination of whether the batters on a team had a tendencyto get hot simultaneously (and/or to slump Simultaneously). Fori f so, the number of high (and/or low) scoring games for the teamought to be more than theoret ical ly expected. Whereas i f thebat ters on the team got hott or slumped, independently of eachother, then some would be hot while others were slumping, thustending to cancel out the effects on the team's overall offense;the probable resul t being that the number of high and low scoringgames would be close to the theoretically expected values.

(3) An examination of. the records of the teams involved in closepennant races might reveal that the ultimate winners wereconsistently closer to (or perhaps consistently farther away from)their theoretically expected runs distr ibut ion than were thenon-winning teams.(4) The data from Table 2 can be uti l ized to obtain an equationrelating won/lOSS percentage as a function of runs scored andopponents' runs. (This item will be the subject of a futurereport) .


11/20

NOLAN RYAN'S FIFTH NO-HITTEROn September 26, 1981, Nolan Ryan pitched his fifth career no-hitter,as the Astros defeated the Dodgers 5-0. While these comments are not analytical,they include a few things that readers may not have found in news articles.Ryan had a lo t of control trouble in the first three innings and wasconstantly behind the batters. He threw so many pitches that I commented to afriend that i t would be tough for Ryan to get a complete game. We estimated thathe had thrown 70-75 pitches through the third; i t turned out he threw 65 pitchesthrough 3, another 16 in the fourth, and a total of 129 for the game. He threw77 strikes and 52 balls, for 3 walks and 11 strikeouts.The control problem was blamed on overstriding; if so, a conference in vthe fourth with the pitching coach straightened that out. "r recorded Ryan1smotion (1981 Abstract) the first two innings asACGHL PS UY 150 -+ C FHM QThis is a composite of a number of pitches. In the sixth or seventh inning rrecorded his motion asACGIL PS TY 150 -- B FKM Q .At this time I noticed that his .motion, particularly the follow-through, variedfrom pitch to pitch (e.g., FHM). I couldn't determine the pitches t but expecthe crosses over more on the fast ball.An A ~ c o u n t Form box score (1979 Abstract) summarizes the game.

LOS ANGELES HOUSTON4LOPES 63 k k 9PUHL 3u 3 s 43 WSMITH k 4GARNER ws sc *1 k s'8LANOREAUX k w 43 3u 8SCOTT k E6 13 '4 97BAKER 43 43 k 53 7CRUZ w 0 ss 13 s'a3GARVEY WSWK l3 2ASHBY 4 S" 7 w 75GUERRERO k 9 63 5HOWE b3 9 63 *22SCIOSCIA k k 9 3SPILLMAN 43 89ROENICKE 5 9 7' PITTMAN s6THOMAS ws 53 43 3WALLING Ww1 OWER k 6REYNOLOS t 43 63 0'1GOLTZ 1 YAN w s K sacfPERCONTE k1 ORSTER *1 FChoice 54JOHNSTONE 43 *2 FChoice 561 TEWART1 OWE

9 9 9 3 RYANPOWER 9 9 1GOLTZ 2FORSTER 6 6STEWART 3 2HOWE 3LOS ANGELES o 0 0 000 000 T- 2:46HOUSTON 002 o 0 0 o 3 x A- 32115

---- Tom Jones

11


12/20

-;

WINS AND IJOSSES FOR ALL PLAYERSMark D. Pankin

With the exception of wins, losses , and saves for pi tchers ,the baseball s ta t i s t i cs which are kept do not provide informationabout whether a player ' s contributions helped his team win games.Instead they focus on some aspect of play which is presumed to bepar t of the winning or losing of games. An attempt i s made hereto provide a method for evaluating player ef for t s with regard towhether they contributed to a victory or defeat .The basic idea i s t ha t when a team wins a game, many of i t splayers made a s igni f icant contr ibution to t ha t win, and when ateam loses , most players can share the blame. Four basic "credits"can be awarded: a Team Pitching Win(TPW) which can be given toindividual pitchers on the winning team, a Team Batting Win(TBW)which may be earned by some bat ters on the winning team, a TeamPitching Loss (TPL) , and a Team Batting Loss{TBL). Correspondingcategories for f ie lding are not established because it is usuallyalmost impossible to assess whether or not bet ter fielding couldhave turned a loss into a win or the other way around.Team Pitching Win(TPW)Every pi tcher on the w i n n i ~ g team receives a TPW except those who

(a ) pi tch "brief ly and ineffect ively" in the scorer 's opinion;or (b) enter the game when the i r team i s so fa r shead it "can ' tpossibly lose . II

Both ca l l for the scorer 's judgement, but speci f ic standards couldbe established. The "brief ly ' and ineffect ively" provision is usedwhen awarding a (regular) win to a pi tcher , and it has not causedproblems to leave th i s to the judgement of the off ic ia l scorer . Aminimum number of innings might be required of the s tar t ing pi tcher(I suggest two) unless he leaves due to injury. Standards forbeing ahead would depend on the inning and the size of the lead, butare not discussed here for the sake of brevi ty . However, anypi tcher with a t l eas t two innings should get a TPW.Team Pitching Loss {TPL}Every pi tcher on the losing team receives a TPL except those who

(a) have no runs (earned or unearned) charged to them in theusual senseior (b) enter the game when the i r team i s "hopelessly behind."

The usual way of charging runs (to the pitcher who put the men onbase) i s used because it is as good as any and easy to use. Unearnedruns are counted because it i s rare t ha t a pi tcher makes no contr i bution to an unearned run. Judgement i s needed for IIhopelesslybehind" and this could depend on the future scoring in the game.For example, i f a pi tcher comes in t ra i l ing 10-2 and gives up tworuns, i f the f ina l is 12-5 he would not get a TPL, b ~ t i f the f ina lscore i s 12-11,12-10,12-9 (or even 12-8 depending on how early thelosers scored the eighth run) then he would get a TPL. Alternatively,specif ic standards could be established.12


13/20

Team Batting Win(TBW)This i s somewhat more complicated and involves calcula t ing the"Run Participationll(RP) of each player . For each bat ter in aninning in which his team scores

(a) i f the ba t te r gets on base, except when a runner is out ona f ie lder ' s choice, his RP i s the number of runs scored inthe par t of the inning beginning with his plate appearance(except for runs which score on a wild pitch , passed ba l lor balk while he i s bat t ing) ;but (b) the bat ter makes an out (including when he reaches baseon a f i e l d e r ~ s choice out play and a l l sacr i f ices ) , his RPi s the sum of (i) the number of runs t ha t score on the playwhen he makes his out, and (i i) the number of runners whoadvance on the play and score l a te r in tha t inning.

The RP of a player is adjusted for being caught s teal ing or beingout in a double play af te r reaching base by using (b) instead of(a) to f igure his RP. For example, i f the f i r s t ba t te r singles ,the second hi t s into a DP, and then some runs score, the f i r s t man'sRP is o. Another example to i l lus t ra te the general process: Asingles to s t a r t the inning, B walks, C forces A a t th i rd , D forcesC a t second while advancing B to th i rd , E singles scoring B, F hi tsa three run homer, G also homers, and H makes the thi rd out for ato ta l of five runs. The RP i s A-5 , B-S, c-o (his out didn ' taccomplish anything), D-I (he advanced a runner who scored), E-S,F-4 (one run had already scored when he came up), G-I, H-O. Eventhough F had the "big blow" in the inning, A, B, and E have higherRP because by not making outs they kept the inning going and madea l l the scoring possible.A b a ~ t e r who hi t s more than once in an inning i s assigned RPaccording to (a) i f he reaches his f i r s t time up regardless ofwhat he does on his second appearance. I f he makes an out hisf i r s t time up, he i s assigned RP according to (b) fo r tha t time upand according to ei ther (a ) or (b) for his second appearancedepending on whether or not he makes another out.A player ' s RP for the game i s obtained by adding the RP's for eachscoring inning.I f the team's margin of victory i s three runs or less , then a l lbat ters with a t l eas t one RP receive a TBW. I f the margin i sfour or more runs, then a TBW is given to a l l bat ters with an RPof a t l eas t the margin minus two (so in a 9-3 win, an RP of a tl eas t 4 i s needed to get a TBW). The idea i s t ha t af te r a cer tainpoint extra runs are not v i t a l or even part icular ly helpful . SinceRP's "come in bunches" in big innings, it is l ikely tha t even in a" laugher" several players wi l l get TBW's.Team Batting Loss (TBL)Every ba t te r on the losing team receives a TBL except those who

(a) reach base safely, excepting f ie lder ' s choice outs , a teach plate appearance (have a "perfect day");or (b) enter the game when the i r team i s "hopelessly behind."I f a player makes no outs , then he has done nothing to diminishhis team's scoring chances and should not be given a TBL whichwould indicate tha t he contributed to h is team's offense beinginsuff icient .

13


14/20

I t may seem unfair tha t in a 12-11 game some players wouldget TPW's and others would get TBL's. The fact is , however, theobject is to score more than the other team, and a pitching s ta f fwhich allows the opposition fewer runs than i t s own team scores isdeserving of some credi t while the bat ters who did not produceenough runs deserve some blame. This points out tha t theses ta t i s t ics are very much team dependent as thei r names suggest,and as such they must be used with care. They could be used to compareplayers, both hi t t e r s and pi tchers , on the same team or on teamswith similar records. However, it would be unfair to use th isdevice tq compare a player on a good team with one on a bad team.By using a winning percentage or wins minus losses divided bygames played, it might be possible to get some insight for MVPevaluation (assuming the players in contention were on top teams).The Guidry vs. Rice debate would be a perfect set t ing.The major drawback to th is system is that TBW and, in somecases, TBL can ' t be determined from an ordinary box score. Aplay-by-play description, an account form box score (1979 BaseballAbstract, p. 52), or something similar is needed. For th is reasonI can ' t provide any nice l i s t s or comparisons. My dream is thatsomething along these l ines would become an "offic ia l" s ta t i s t i cwould could easi ly be kept by the off ic ia l scorers and league off ices .What appears here is essent ial ly a proposal in need of refiningand more precise defini t ion. The few standards given are open todiscussion, and i f any circumstances have been omitted, they shouldbe able to be handled using the general guidlines presented here.AII comments and suggestions are welcomed.

14


15/20

Home Runs - - A Matter of AttitudeRobert H. Kingsley, August 1979

Atlanta is an easy home run stadium - - Busch Memorial in St. Louis is a pitcher 'spark - - L.A. Stadium is not easy for home runs. Fans, players, sports writers andannouncers, a l l have viewpoints on how the stadiums compare for home runs. But thereare ways to put a yardstick on these stadiums. Let 's divide the factors affecting homeruns in three broad categories - - f i r s t , the physical dimensions of the stadium; second,the weather or climate factors; and third, the att i tude factor - - what do hi t ters andmanagers think about the stadium for home runs. The f i rst two factors are generallyaccepted and will be treated briefly; emphasis given to the att i tude factor.

Distances from home plate and height of the fences a l l around the outfield are themost important physical dimension factors. But slope of the field, frequently abouttwo feet downward from home plate to the outfield acts just like a lower fence height.Further, foul bal l areas make a difference as more pop-ups are caught in the largerfoul bal l areas affecting both batting average and home runs.For the weather or climate factors sometimes referred to as "carry of the ball" ,

a ir density is perhaps the most important. The thinner, or less dense the air , thelesser the resistance to the fl ight of the bal l . Altitude above sea level and temperature of the a ir determine a ir density. Dry a ir is just slightly more dense than a irwith water vapor in i t , so we shouldn't expect high humidity a ir (short of rain, hai l ,etc .) to slow down the fl ight of the bal l . Winds, even on a season average basis mustbe evaluated everywhere except in domed stadiums. The famed winds at Candlestick madehome runs tough for r ight handers and easy for le f t handers, but the stadium encirclement completed a few years ago has greatly reduced the impact. Winds have been a majornegative factor in hit t ing home runs a t Arlington Stadium. Resiliency of the baseballis affected by both temperature and absolute humidity (water vapor in th e air ) . Testsshow that baseballs that ' a r ~ dry, or warm go farther than those that are cold or damp.Resiliency is not easy to harness because of many variables, including baseball storageand handling practices. Baseballs, relat ively dry in storage tend to gradually absorbmoisture from damp a ir with the result that the bal l gains weight, the cover becomessofter and the ball just won't go so far . Among many stadiums, the impact of weatherfactors is minor, but i f Shea Stadium in New York were exactly the same as RoyalsStadium in Kansas City, weather factors would cause 35 percent more home runs inKansas City.Attitude, viewpoint, or emphasis on hitt ing home runs by the individual hi t ters isthe third, but a different kind of factor affecting stadium home runs. Since intent to

hit home runs is a human and emotional factor, i t does not lend i t se l f to the engineering and scientific approach necessary to handle the other broad factors of physicaldimensions and weather factors. Perhaps the best way to come to grips with this factoris by an actual example. Atlanta Stadium has been considered an easy home run stadiumand Busch Memorial in St. Louis a diff icul t home run stadium. Let 's take a look at a l lthree broad factors in both places by the l i t t l e table below - - (all numbers in feetunless otherwise marked):

Physical DimensionsFence distancesFoul l inesPower alleysCenter fieldFence heightSlope - home plateFoul bal l area

Atlanta

3303854026to fence 11% easier for HRs

Busch Memorial

33038641410 1/2

1


16/20

Atlanta Busch Memot:ial... Weather FactorsStadium elevationTemperature - game timeHumidity - game timeWinds - game time

96077 .565%same

at airportat airport

Bat RightBat Left 2 1/2% easier for HRs

4607661%same--

Using a l l the data above and our standard method of determining stadium home runAtlanta Stadium comes out 33% easier for home runs than Busch Memorial fore average Major League home run hit ter . All of the differences in the table can beindividually to reach the 33% figure. Busch Memorial is longer by 1 foot at theower zone and 12 feet at centerfield, but the 12 foot difference over 400 feet inis not very significant; accordinglYt Atlanta ,is only 3% easier because ofdistances. The 4 1/2 foot difference in fence height a l l along the fence is muchlore significant and results in 11% more home runs at Atlanta Stadium. Including th efor foul ball areas, the to ta l difference for physical dimensions is 15%. For

or climate factors, the altitude of Atlanta Stadium of 960 feet above sea level.s the most significant. Using th e 500 foot difference and 1.5 of temperature, thedensity factor favors Atlanta by 13%. The resiliency factor appears to favort. Louis but the fact that Busch Memorial is so near th e Mississippi River, higherhumidity exists compared to th e airport at Lambert. Field. Considering th eigher temperature a t Atlanta as well, Atlanta is considered to be favored by approxi2%. Atlanta is favored by 1% over Busch Memorial for winds primarily because ofin-blowing winds from r ight field a t Busch Memorial, mostly affecting lef t handed>tters. Combining individual factors, Atlanta is favored by 16% for weather factors.1.16 x 1.15 = 1.33 or 33% more home runs at Atlanta. This 33% advantageassumes a l l of the hi t ters have the same att i tude toward hit t ing home runs atusch Memorial as at Atlanta, which of course is not the case as we will see l a ~ e r on.Actual home runs hit can be used in a couple of ways to actually express how thewo stadiums have compared for home runs, considering a l l factors. Comparing the BravesIt home and away is one method; another is relating how the Braves' opponents compared

t Atlanta versus how they did at home against Atlanta. These approaches assume consis-in home and away pitching by visitors and Braves pitching respectively. The sameethods can be used fo r the Cardinals and the past three years should be enough for acomparison. Here are th e home run data:Braves Braves OpponentsHome Away At Atlanta At home against Braves

1976 43 39 1976 56 301977 97 42 1977 111 581978 87 36 1978 89 43227 117 256 131

227 256 ... 1.96117 ... 1.94 Avg ... 1.95 131


17/20

Cardinals Cardinal Opponents( .Home Away At Busch Mem. At home against Cardin

1976 27 361977 41 551978 29 5097 141

97 =141 .69

197619771978

Avg = .66

405332125

125199 = .63

518662199

Because "away" for the Cardinals includes Atlanta and not Busch Memorial, and because"away" for the Braves includes Busch Memorial and not Atlanta, th e averages of .66 and1.95 should be adjusted on a comparable basis. Also, adjustment needs to be made forthe temporary fence at Busch Memorial for 1976 which was removed for 1977. The surprisingresults are that 34% less home runs were h it a t Busch Memorial, and 80% more home runsh it a t Atlanta stadium when compared to the average National League stadium for1976, 1977 and 1978.Earlier, we arrived a t a factor of 33% more home runs at Atlanta because of factorsof physical dimensions and weather or climate factors. Relating the 33% difference toth e 1976, 1977, and 1978 average park, Atlanta is 14% above this average and Busch Memoria

is 14% below the average. (1.14 .86 = 1.33) We can now solve for the att i tude, oremphasis factors by dividing the percentage value for a l l three factors by th e combinedphysical dimensions and weather factors, i . e . , for Atlanta 1.80 1.14 = 1.62. Resultsare summarized below:Atlanta Stadium compared for home runs to 1976-1978 National League Average Park

All factors - 80% easierPhysical dimension andweather data factor - 14% easierAttitude 7 or emphasisfactor - 62% easier

Busch Memorial Stadium compared for home runs to 1976-1978 National League Average PaAll factors - 34% more difficultPhysical dimension andweather data factor - 14% more diff icul tAttitude, or emphasisfactor - 23% more diff icul t

Surely, the startl ing att i tude factors are much larger than might be expected.How can attitude or emphasis be broken down or described. Characteristically, longbal l hi t ters have a peaking distance from home plate where their long drives land morefrequently, falling off somewhat closer in , and of course falling off farther out.


18/20

mgest hi t ters could be peaking at 430 feet, while sho.rt hi t ters could be peaking atfeet. Generally, long hi t ters peaking beyond most of the outfield fence are think-home run most of the time anyway. However, average home run hi t ters , and particularlyhome run hi t ters are faced with choices. Hitters peaking a t , or short of the fence1 their most usual hitt ing directions can help the team more by forgetting about homemeeting the bal l , hitt ing the bal l where i t is pitched and otherwise get on base

i s negative home run emphasis. Hitters peaking a t. or just beyond the fence couldhitt ing 2 to 3 times as many home runs if they take a positive rather than a negativeon hitting home runs; the positive attitude means more strike outs, pop-ups, andfly balls , but a lo t more home runs, too. I t is not just a case of individual judg

~ n t about home run emphasis. The fans, manager and hitt ing coach a l l have something toabout home runs.The home team's attitude about home runs is usually adopted by visit ing teams.artainly, that has been the case at both Atlanta and St. Louis as the earl ier homen totals have indicated. Could be a lo t of reasons for this - - but l ikely a major onethat th e visit ing team feels that the home team would have adopted the most appropriateJme run emphasis game in view of the current hi t ters and distances to fences - - so theysame kind of home run emphasis game. Very few exceptions occur to this sameof game strategy.Stadiums have a way of gaining a reputation by one way or another. Atlanta Stadiumas a reputation as the easiest home run stadium - - thousands of people have been therethe balls "just flew r ight out of there ll That can happen on a warm, dry day, withof positive home run emphasis and usually a fast bal l pitcher. After al l , theNational League home runs have been h it in Atlanta almost every year. On the otherBusch Memorial is known as a pitcher 's park - - tough on home runs. A common view-Jnt is that the fences seem' a long way out and the bal l doesn't carry very well. Theyre farther out, but only in centerfield: but another factor is the damp a ir moving int night as a breeze from the South, or right field direction;without much home runi t t ing prowess in recent years, and plenty of negative home run emphasisJ few homehave been hit in Busch Memorial up to 1979.One or two big hi t ters on a team has a way of motivating others on a team towardhome run emphasis ( if he can do i t , why can' t I7). This kind of emphasisearns the stadium. a reputation. Certainly,this was the case with Henry AaronAtlanta; also big hi t ters at Veterans and 3 Rivers stadiums to a lesser extent. Weonsider Fenway not so easy for home runs as the reputation i t has gained over the years.Attitude about home runs, or home run emphasis i s very ephemeral. The manager byirection to the hi t ters can change overnight the way the game is played. 50% more homeor 50% less home runs are obtainable in many instances.On the way to Spring training, the past two years we have gone by Atlanta to look

the scene of a l l these home runs. Our thought that the fences might be 15 to 20shorter than the indicated distances was met by considerable brist les from the veryRobert Johnson, grounds superintendent. He assured us the fences were right withinmatter of inches and we w o u l ~ be welcome to check the measurements. He mentioned thatdays the balls did seem to f ly out of there, but a lo t of days they'didntt, but oneanother the stadium was looked on as an easy home run stadium. Further, with JoeStadium Operations Director, he mentioned that th e many home runs had' causedbut he also wanted me to know the fences were r ight a l l along. In discussingt i tude and emphasis, he fe l t that the outlook of the Braves toward hitting home runs"lad to do with the big season totals.

18


19/20

At the Braves training camp at West Palm Beach, talks with Paul S n y d e r ~ Minor LeagueAdministrator, confirmed the outlook toward hitt ing home runs."I think att i tude is th e (biggest factor in home run hitting," he said, with Atlanta primarily in mind. With the \power zone distances pretty much the same in the National League, and the emphasis figureswe developed for Atlanta Stadium in this art ic le , he couldn't be more right.There was a chance to contact Henry Aaron, now Vice President, Player Development.

I t would have been s i l ly to ask him i f he emphasized home runs - - instead he was graciousto le t us take his picture. From our viewpoint, he was hit t ing the bal l frequently(3 t 771 hits) and'hard (755 home runs) and was emphasizing the home run everywhere - - hislong hit drives peaking beyond th e fences almost everywhere.This year the Cardinals broke from the start ing gate hitt ing many more home runsthan previous years. Simmons, the star catcher, has taken a more positive view abouthome runs. He has adapted his swing to pull th e bal l down the line toward the shorterfence in Busch Memorial Stadium. The Cards and visi tors alike were hitt ing more homeruns and hitt ing them well up in the stands. Last year the Cardinals h it 79 homersand only 29 were h it at Busch Memorial. This year the Cards have h it a tota l of 75through Friday August 17. The Cards have tailed off on home run hit t ing since Simmonswas hurt and haven't got going again since he returned. A recent check with Jim Toomey,Public Relations Director, indicates "hit t ing the bal l better" accounts for quite a lo tmore home runs, as th e Cardinals continue to lead the league in hi t t ing. For a long time,in fact since Musial ret i red , the Cardinals haven't had much home run hit t ing power so thelimit.ed negative emphasis which they have been doing, has made sense in relatively largeBusch Memorial . However, there are indications the home fans would l ike to see more homeruns.Apart, of course, from the varying hitt ing capabil i t ies and pitching capabilitiesagainst the home run: of Major League teams, we have named three broad factors affectingstadium home runs: 1) physical dimensions; 2) weather or climate factors; and 3) home runatt i tude or emphasis. But perhaps there is even a fourth, plus improvements in the threefactors we have cited. There is room for improvement in anything that has a scientif ic,

engineering side to i t so we never know precisely that everything is properly indicated.However, in our 25 years ofexperience in determining stadium home run potentials, wehave discarded such factors as variations in baseball bats, sun angle, night games versusday games, travel and some others, as usually impractical and of relatively small effect .I t could be asked why we prepared this article by some who are familiar with ourwork. But there have been many who have thought our home run potential values, embracingphysical dimensions and weather factors, for the average Major League team, were much toolow for Atlanta Stadium and somewhat high for Busch Memorial. But since the measurementsto the fences and the weather data are so much alike, i t would be very unlikely therecould be a greatly different home run potential. The really significant differences arethe 4 1/2 foot higher fence, and the 500 foot h i g ~ e r alt i tude at Atlanta.Further, th e att i tude or emphasis factor wil l be recognized as the same emphasisfactor we developed ear l ier and was introduced in Sports Il lustrated in July 1973.Since then, we have annually furnished baseball act ivi t ies emphasis factors solved formathematically, for both th e home team and the visi t ing team a t each Major League stadium.The Atlanta-Busch Memorial stadium comparison was a good opportunity to relate stadiumswhere the home run potential could not be vastly different but where there was a vastdifference in tota l home runs and would highlight the emphasis factor.

19


20/20

. , . " ' 1 " ' - ' - ' - " , : ~ .. .:;.ru >. ....-: - ......._:- , . _ ~ . ~ - ' - ~ ~ n z . ' , , " , - : ' . ' " ' - . : ._,., v- " . - ~ -. -

Through the courtesy of the Elias Sports Bureau in New York and the Weather Servicewe have accumulated the number of home runs and the. temperature and humidityor night home runs were hit at Atlanta last year. We plan to determine toextent more home runs are hit on warm, dry days. Perhaps there will be examplesat will i l lustrate the emphasis that will show how the Atlanta Stadium earned i tsa home run stadium.In the way of affi l iated information. Atlanta is 9% easier and Busch Memorial Stadium18% more difficult in home run potential than the average Major League stadium for(1.09 .82 =1.33). We have noted no changes in outfield fence distances andin the Major Leagues from 1978 to 1979. So home run potentials for Majorstadiums remain the same as provided in our let ter of August 17 , 1978. In theLeague, batting right handed in Wrigley Field and batting le f t handed in(very slightly) are considered easier home run situations than Atlantar the average Major League home run hi t ter . Similarly, the Astrodome and Olympicare more difficult than Busch Memorial t where batting le f t handed is morethan right handed. Similarly, counting either batting right or batting lef ta specific example,our information showed seven instances where home runs in theare easier than Atlanta, and eight instances in the American Leaguehome runs are more difficult than Busch Memorial Stadium.

However, as this article meant to prove, i t '"is not just home r u ~ potentials thate important for stadium home runs; with home run emphasis a very big factor at Atlantaothers, too, i t is "the way they play the game".

01 baseball analyst 1982 06

Documents