STRAT RESEARCH ASSOCIATES (SRA)

SRA is an organization devoted to the quantitative study of Strat-O-Matic (SOM) baseball. SRA has developed several SOM computer-based managing aids. If you are interested in taking the guesswork out of playing SOM baseball, you have come to the right place !!

 Interested in SOM BB tournament play ? The STAR TOUR provides a national clearinghouse for conducting your very own approved tournament !

There will be Excel and (probably) Lotus 2000 card set versions of LINEUP out sometime in the February/March 2000 timeframe.  A MSWorks version is also possible, stay tuned ...

 Questions/comments email to: velten@dnaco.net.

 Vince Velten, President of SRA

Table of Contents

The LINEUP Spreadsheet

Draft Analysis

Team Management Analysis - Team Strength and Managing Skill

 

Runs Created in SOM

General Analysis - When is it Luck ?

 Other SOM Web sites

The SOM Game Co. can be contracted directly via email at som1@ix.netcom.com. Do SRA a favor and tell them you got the info from this web page !

 


The LINEUP spreadsheet

  

SRA has developed a spreadsheet to address perhaps the central question in SOM player selection - when to trade-off glove for stick or 'stack' against a backwards/forwards pitcher ? Probably the easiest way to explain LINEUP is to show an example output using the 1994 card set: (NB: Set your fixed font preference to Courier or Courier New or another fixed pitch font in order for the example views to align correctly)

 

OB% TB% BA  CBA Runs  trnmnt:         0               ninn: 9              Pos Rng E
289 364 212 221 3.99  Gaussian W%:    Batting Average 233     daultond.phn 2   2   2
326 444 276 241 5.8       479         On Base %       274     bagwellj.hon 3   1   14
212 255 212 216 1.86                  Total Base %    377     lindj___.kca 4   1   10
334 313 264 218 4.23                  Def OB (P card) 10.9    boggsw__.nya 5   2   19
288 299 224 232 3.27                  Def TB (P card) 59.1    larkinb_.cin 6   1   16
312 440 239 198 5.39  Pythagorean W%:                         bondsb__.sfn 7   1   5
287 461 255 197 5         487         Runs Created    3.85    griffeyk.sea 8   1   7
344 374 325 248 5.31      482         Gaussian RC     4.03    gwynnt__.sdn 9   2   5
97  171 85  85  0.5                   s               1.82    ashbya__.sdn 1   2   0

OB% TB% BA  CBA Runs
251 448 229 294 4.05  Gaussian W%:    Batting Average 208     piazzam_.lan 2   3   2
365 496 242 235 7.69      521         On Base %       256     thomasf_.cha 3   4   12
259 263 204 188 2.48                  Total Base %    425     reedj___.mla 4   3   5
199 440 175 174 2.96                  Def OB (P card) 55.2    williamm.sfn 5   1   18
241 291 164 183 2.49                  Def TB (P card) 84.4    bellj___.pin 6   2   23
302 540 260 234 6.32  Pythagorean W%:                         bellea__.cla 7   3   9
328 459 304 279 6.04      513         Runs Created    3.95    loftonk_.cla 8   1   3
273 380 202 201 3.86      518         Gaussian RC     4.17    salmont_.caa 9   2   13
115 186 94  94  0.66                  s               2.01    madduxg_.atn 1   3   30
LA  8   5   5   5     copyright 1995 SRA.  All rights reserved.
Prk LSI RSI LHR RHR

LINEUP is based on the well-known sabermetric runs created (per out) formula 'discovered' by Bill James (RC=OBP*TB/(1-OBP)*27). In it's current version all major game effects are included such as W/N power, ballparks, clutch, double plays, 1W pitcher hitting for tournaments, even the impact of baserunning speed is included (note that base stealing is NOT included).

 Explanation of LINEUP display

 As an example of the use of LINEUP, consider the following question: Would it be worthwhile to play Spike Owen over Matt Williams against Andy Ashby in LA ? Here is the result of that change:

 

OB% TB% BA  CBA Runs  trnmnt:         0               ninn: 9              Pos Rng E
294 367 218 227 4.12  Gaussian W%:    Batting Average 238     daultond.phn 2   2   2
331 448 281 246 5.97      468         On Base %       279     bagwellj.hon 3   1   14
217 258 217 221 1.93                  Total Base %    381     lindj___.kca 4   1   10
339 316 269 223 4.37                  Def OB (P card) 10.9    boggsw__.nya 5   2   19
293 302 229 238 3.38                  Def TB (P card) 59.1    larkinb_.cin 6   1   16
317 443 244 203 5.55  Pythagorean W%:                         bondsb__.sfn 7   1   5
291 464 260 202 5.16      476         Runs Created    3.97    griffeyk.sea 8   1   7
349 378 330 253 5.47      474         Gaussian RC     4.16    gwynnt__.sdn 9   2   5
103 174 90  90  0.54                  s               1.86    ashbya__.sdn 1   2   0

OB% TB% BA  CBA Runs
251 448 229 299 4.05  Gaussian W%:    Batting Average 218     piazzam_.lan 2   3   2
365 496 242 240 7.69      532         On Base %       272     thomasf_.cha 3   4   12
259 263 204 193 2.48                  Total Base %    413     reedj___.mla 4   3   5
341 339 262 324 4.73                  Def OB (P card) 64.7    owens___.caa 5   4   20
241 291 164 188 2.49                  Def TB (P card) 90.8    bellj___.pin 6   2   23
302 540 260 239 6.32  Pythagorean W%:                         bellea__.cla 7   3   9
328 459 304 284 6.04      524         Runs Created    4.17    loftonk_.cla 8   1   3
273 380 202 206 3.86      526         Gaussian RC     4.38    salmont_.caa 9   2   13
115 186 94  94  0.66                  s               2       madduxg_.atn 1   3   30
LA  8   5   5   5     copyright 1995 SRA.  All rights reserved.
Prk LSI RSI LHR RHR

Spike improves Pythagorean W% by 11 points ! This is a significant result, as experiments with LINEUP have shown that it is accurate to within roughly 5 points in Pythagorean W%. Note that were this analysis to be performed in San Diego, Williams comes out about 23 points better than Owen.

 LINEUP runs under windows and has been ported to Excel 5.0 & 7.0, 123 for Windows 1.0, or Quattro Pro 5.0 format. A Mac Excel version is also available.

Demo versions of LINEUP

Demonstration copies of the 1994 season card set version of LINEUP include just the carded players for Boston and Cincinatti. Note that these are .zip files (v2.04g), so you will have to unzip them yourself and then manually launch the appropriate spreadsheet application and open the file. MAC Users: The Excel version is compatible with the Mac 6.0 version of Excel (don't know about earlier Mac Excel versions) - there are decompression programs available on the WWW that are compatible with pkzipped files.

 Excel 5.0 (Windows) 69K
Lotus 1-2-3 1.0 (Windows) 70K
Quatro Pro 5.0 (Windows) 76K
Instructions on using the demo version 16K

Ordering LINEUP

As mentioned above, LINEUP is available in 2 popular spreadsheet formats (Lotus 123 97, Excel V5.0 or 7.0 [Win 95]). Simply print out the order form, fill it in, and mail with $15 + $3 (Cash, US Bank Check, or Money order - US Funds only) postage and handling to:

Vince Velten c/o Strat Research Associates

 2019 Upper Bellbrook Rd

 Xenia OH 45385-9339

 Make the check/money order out to Vincent Velten.

 NOTE: All of the spreadsheet formats require Windows or Windows95. Mac users - LINEUP can be ported to the Mac version of Excel for an additional $2.

Draft Analysis

SRA has also developed a drafting program (called DRAFT) that has been successfully employed at over a dozen SOM tournaments, producing a combined winning % in excess of 580 over 350+ TBA games. It runs under LOTUS 123 V2.4 (for DOS) and is still 'under development'. However, if you are really interested, come to a TBA event in the Midwest and you can check out the drafting program in person. SRA anticipates eventually making the drafting program available publicly in 2-3 years. For an indication of the methodology behind DRAFT, take a look at -

Runs Created in SOM

SRA has produced an extensive analysis of how Runs Created can be adapted to SOM, and plans to submit several articles to STRAT FAN over the next year. We have submitted an article to STRAT FAN on base stealing in SOM, unfortunately it was rejected as "too technical". Well, now you can just judge for yourself: Stealing: A Runs Created Analysis.

For those of you who favor the simplicity of OPS for player evaluation we have some bad news - read The Case Against OPS.

If you are mathematically oriented, we have developed a statistical model for using runs created in the context of a simulation game - Multinomial Runs Created.

We are continually in the process of converting our research results to HTML files; watch this page for future updates.

 

Other SOM Web sites

(Unofficial) SOM Home Page : Adam C Wodon's collection of SOM History, game comments, league pages, etc. - a must see.

 (Official) SOM Home Page: The game company's official WWW page.

 


10/04/1999 01:48 PM