The 2015 DMB Annual Season Database

2015: Rookies, Upstarts, and Royalty

by Steve Ehresman

The 2015 Major League Baseball season provided fans with brilliant rookies, wild post-season scrambles, and an electrifying world champion.  Our venerable National Pastime is alive and well in the early twenty-first century.       

The Kansas City Royals, still smarting from their loss to the San Francisco Giants in the 2014 World Series, came out of the gate with attitude, establishing themselves as a team to be reckoned with, despite their being underrated by the pundits.  Time and time again, the Royals came from behind, winning the World Series and forcing the baseball world to recognize them as a powerhouse.

In addition to the compelling season-long saga of the Royals, 2015 featured a dynamic rookie class, many of whom found themselves in the thick of pennant races.  Among this group were American League Rookie of the Year, Carlos Correa; National League Rookie of the Year, Kris Bryant; and Noah Syndergaard of the New York Mets, the surprise National League Champions.  Few predicted that the Astros, the Cubs, and the Mets, propelled by their young talent, would push their way into the elite post-season crowd and provide their fans with October baseball. 

The 2015 baseball season featured Josh Donaldson’s establishing his bona fides as a super star, while capturing the American League MVP, and Bryce Harper’s compiling a season for the ages, while wrapping up the National League MVP.  On the mound, American League Cy Young winner, Dallas Keuchel, and National League Cy Young winner, Jake Arrieta put their upstart franchises on the baseball map.  All-time great performances by Mike Trout and Zack Greinke underscored the depth of talent Major League Baseball enjoys.             

Diamond Mind Baseball is proud to present the memorable 2015 baseball season in our outstanding version 11 format, complete with all you need for a realistic and exciting replay.  An era of gifted youngsters and brash new teams is dawning.  Be part of the excellence. 

The 2015 Deluxe Past Season database contains everything you need to play games using teams and players from the 2015 season -- a full set of ratings and statistics for every player who appeared in the big leagues that year, plus team rosters, manager profiles, ballpark ratings and league schedules. Statistics include official batting, pitching and fielding totals with left/right splits for all batters and pitchers.

Also included is a complete set of real-life player transactions -- trades, disabled list moves, promotions, demotions, suspensions, and more -- plus the actual starting lineups for every regular season game played.

Note: This season database is a companion product for the Diamond Mind Baseball version 11 game. To use this database, you must also have Diamond Mind Baseball version 11. The game software provides you with all of the tools you need to play simulated games, make roster moves, produce dozens of statistical reports, generate league schedules, and more.

Imagine Sports Birthday Sale!

Believe it or not, Imagine Sports is 10 years old.  We started working with Tom Tippett to build the online version of Diamond Mind in the summer of 2005, and it was 10 years ago this week that we fired up the system for the first time.  
10 years is a long time, and we're happy and honored that you have taken this journey with us.  To celebrate 10 years, we're having a $10 off anniversary special for the rest of November:
$10 off anything and everything in the Diamond Mind store.  For anything already $10 or less, it's free!
Enter this code at checkout for your discount: ImagineSports10

***The fine print:  This offer expires at 11:59pm EST on November 30, 2015.  No returns or refunds are allowed on sale items.***  
In the meantime, rest assured that work is well underway on the 2015 Season Download (how 'bout them Royals???) as well as a bug-fix patch for DMB version 11.  We will let you know when each will be ready as the release date approaches.
Play ball and...
Happy Holidays from the Imagine Sports Team!

Do Batters Learn During a Game?

Editor's note: This article was written by David Smith, founder and president of Retrosheet, a non-profit organization dedicated to the collection and computerization of play-by-play scoresheets for games played prior to 1984. We have edited this material only to the extent of formatting it for publication on the Web.

David W. Smith
June 7, 1996

It is common to hear players, both batters and pitchers, comment on the value of being able to "make adjustments" during a game. For example, pitchers speak of "setting a batter up" by a certain sequence of pitches, which may take several at bats to accomplish. Similarly, batters often remark that they "look for" a certain type of pitch or in a certain location after considering what the pitcher has thrown before. Although it makes sense that a player will alter his mental approach as a result of earlier success or failure, I decided to go beyond the anecdotal interviews and ask if there were any tangible evidence indicating that this learning actually takes place.

I analyzed every play of every game from 1984 through 1995, which is 24823 games and more than 1.69 million at bats. The play by play information, which comes from the Baseball Workshop in Philadelphia, is publicly available. In the very near future similar data will be available for earlier seasons from Retrosheet, the organization of which I am proud to be President. The analysis here is limited to matchups between starting batters and starting pitchers, thereby allowing the study of the maximum number of sequential encounters in a given game. Given the realities of modern relief pitcher usage, it is very uncommon for a batter to face the same relief pitcher more than once in a game, and therefore the relievers were excluded. The batting performance of pitchers was also removed.

The next question is how to evaluate performance so that we can make the comparisons in a meaningful way. I chose to calculate the three standard aggregate measures: batting average, on base average and slugging average. Sabermetric studies in the last two decades have made it clear that these three reflect different aspects of batter performance and I therefore suspected that they might not all show the same pattern of learning during a game. The following table presents the results for all games from 1984 to 1995.

Batting by Number of Appearances.

All games, both Leagues, 1984-1995 

          PA    BA   OBA    SA 

1st   419870  .259  .327  .391 

2nd   401917  .268  .331  .413 

3rd   313880  .272  .334  .422 

4th    90994  .276  .338  .422 

In addition to noting how uncommon it is for a starting batter to face a starting pitcher four times in a game, we see clear patterns of improvement, or learning, in all three values as the game progresses. However, the three averages do not increase at the same rate. On base average rises slowly, only 3.4% from the first to fourth time at bat, while batting average and slugging average go up much more rapidly, 6.6% and 7.9% respectively.

In the 1950s Branch Rickey and Allan Roth developed a measurement called isolated power to examine extra base hits separately from singles. Isolated power is simply the difference between slugging average and batting average. For all at bats over the 12 years studied the isolated power is .134 (batting average of .260 and slugging average of .394). The isolated power values for the four times at bat are .132, .145, .150, and .146.

My interpretation is:

  1. the first time up batters are more concerned with making contact than hitting with power and;
  2. the second and subsequent times up they are adjusting with the result that they are able to swing more confidently and with greater power.

Of course, we can't lose sight of the possibility that pitchers are learning during these successive at bats as well. However, the increases indicate that in a relative sense the batters are ahead of the pitchers in their adjustments.

In an attempt to look for other factors controlling these numbers, I divided the games by league; these results are presented in the following table.

AL and NL Batting by Number of Appearances, 1984-1995 

American League 

            PA    BA   OBA    SA 

 1st    234152  .259  .328  .394 

 2nd    222576  .268  .330  .415 

 3rd    173057  .270  .332  .423 

 4th     53727  .274  .339  .424 

 National League 

            PA    BA   OBA    SA 

 1st    185718  .259  .325  .388 

 2nd    179341  .269  .331  .410 

 3rd    140823  .274  .336  .422 

 4th     37267  .278  .338  .418 

A quick glance shows that the two leagues are very similar in all three values, perhaps more so than might be expected, given the reputation of the American League as the more offensively minded. I would like to address this point with the small digression shown in the next table.

Correction for Effect of DH and pitchers, 1984-1995. 

                BA   OBA    SA 

 AL           .263  .331  .403 

 NL           .256  .321  .383 

 Total        .260  .326  .394 

 AL DH        .257  .335  .419 

 NL P         .142  .176  .178 

 AL All-DH    .264  .331  .401 

 NL All-P     .263  .330  .397 

The top portion of this table shows that the AL has more offense by all three measures, confirming the conventional wisdom. The middle portion presents the data for the DH and for pitchers, with the expected huge difference. The bottom part of the table was derived by subtracting the DH from the AL values and the pitchers from the NL data. The results show very close agreement, perhaps surprisingly so, between the two leagues. In fact, comparing the NL with pitchers removed to the entire AL with the DH included gives even closer agreement. My conclusion on this point is that essentially all the difference in offense between the two leagues is accounted for by the DH.

The next idea I had for subdividing the results was by home and road team, as presented in the next table. To my surprise, there are rather large differences between the two, both in absolute value of the numbers and in the pattern of changes. The home team has an overall seven point superiority in all three of the measures used here. However,

Home and Road Batting by Number of Appearances, 1984-1995 

 Home       PA    BA   OBA    SA 

 1st    209837  .265  .335  .401 

 2nd    200459  .272  .336  .421 

 3rd    153111  .276  .340  .431 

 4th     40051  .276  .341  .424 

 Road       PA    BA   OBA    SA 

 1st    210033  .253  .318  .382 

 2nd    201458  .265  .325  .404 

 3rd    160769  .268  .328  .414 

 4th     50943  .275  .337  .420 

 All              BA   OBA    SA 

 Home           .263  .333  .401 

 Road           .256  .320  .386 

 All            .260  .326  .394 

the greatest differences are in the pattern of the changes. In all three parameters, the rates of increase are steeper for players on the visiting team than they are for those who are playing at home. In fact, slugging average for the home players actually drops from the third to fourth time at bat. By the fourth time at bat, the home and road players are almost identical. This pattern is initially surprising, since it is not obvious why the road team batters should display so much more learning than the home team batters.

However, we must remember that there are two sides to each matchup and consider the pitchers as well. One of the great differences usually identified between different parks is the mound and it is common to hear visiting pitchers comment that it takes time to adjust. Therefore, it is reasonable to consider that there are two kinds of learning going on. The first is the mental part of the pitcher-batter confrontation, which we have seen to favor the batter, and the second is the physical adjustment by the pitcher to the mound. Presumably the home team pitchers are more familiar with the mound than the road team pitchers are and they should have less of this adjustment to do.

Let us consider the home vs road differences again, remembering that the home and road batters end up with very similar performances. By this argument, the learning displayed by the road team batters would therefore result mostly from the mental aspects, since the home team pitchers are not affected as much by the mound. On the other hand, the road team pitchers are starting the game at a relative disadvantage as they deal with the idiosyncrasies of that particular mound. Therefore, the performance by home team batters starts off at a higher level, but does not increase as rapidly, because there is less room for improvement before they reach the maximum in the fourth time up. However, it must be true that the road team pitchers have been successful in their adjustments, or else one would expect that the performance by home team batters would continue beyond what is actually observed.

There is one additional factor which might affect the batters, and that is the nature of the hitting background. Although the center field background does vary between parks, there is much less variation here than there is in the mound. One way to examine the effect of the hitting background would be to compare the performance of road team batters in the first game of each series to the later games in the series. If the background were a significant factor, then one would expect the first game performance to be different. I did not subdivide the results in this way, so this possibility remains unexplored.

Until last Saturday, these were all the ways I had thought of dissecting the data. However, that afternoon I received a call from Jerry Crasnick, who is a sportswriter with the Denver Post. Jerry was calling to discuss my research on the effect of artificial surface, which I presented at last year's SABR meeting, but during our conversation I mentioned my topic for this year. Jerry told me he had discussed this very subject with Craig Biggio of the Houston Astros and that Craig was firmly convinced that the physical demands of the position cause catchers to pay a huge offensive price.

The essence of Craig's comment was that he was "toast" his last time up. Of course, I was immediately inspired to check out this assertion and I reran my programs to record the batting events of catchers separately from those non-catchers. The results are in the next table, where even a quick glance shows that the patterns for catchers are quite different. The overall totals are lower, which is no surprise,

Effect of Being the Catcher on Batting by Number of Appearances 

 All Batters Except Catchers 

            PA    BA   OBA    SA 

 1st    395352  .259  .327  .392 

 2nd    355861  .271  .333  .416 

 3rd    282489  .274  .336  .425 

 4th     86237  .277  .340  .423 


            PA    BA   OBA    SA 

 1st     24518  .253  .319  .378 

 2nd     46056  .250  .310  .384 

 3rd     31391  .256  .316  .398 

 4th      4757  .257  .317  .394 

but what strikes me is the very slight increases by the catchers during the game compared to the other batters. In fact there are some noticeable decreases in batting average and on base average between at bats one and two for the catchers. Over the four plate appearances, catcher batting averages rise only slightly and on base average actually declines. Slugging average shows the overall increase we have seen all along, but to a lesser degree.

Do these results support the idea that catchers are damaged later in the game? In an absolute sense, it appears the answer is no, but compared to other players the answer is clearly yes.

What do these results have to do with learning? Assuming that catchers can make adjustments as well as other players, then it appears that their learning as batters is largely overcome by the physical demands of playing behind the plate.

Before I give my conclusion, there is one more point that must be made, which is to note that I presented no information for individual teams or players. It is always true in a study such as this that the results get less clear as the sample size gets smaller. I therefore made the large divisions of league, home vs road, and fielding position. When the results are divided more finely, to single teams or single batters, there will inevitably be many exceptions that cloud the issue, largely because of their statistical unreliability. I have chosen to avoid this confusion.

In conclusion I note that I began this study with the question that is the title of the presentation: Do Batters Learn During a Game? It is clear that the general answer is: yes, they do. However, it is also clear that the situation is a little more complicated than that and that a better understanding can be obtained by considering other factors, the two biggest of which are the effect of playing at home vs on the road and of being the catcher. So the next time you hear a batter say that he improved his performance by making adjustments during a game, there is a good chance you should believe it. On the other hand, if you hear a pitcher say it, then you might be a little suspicious.

Viewing or Printing the DMB Help File


The DMB Help File is located in the DMB game folder and is called dmb.chm.  If you want to use the Help file outside the DMB game itself, make a new folder and copy the dmb.chm file into the new folder.  From there you can open the file with the program of your choosing, such as PDF, Word, Notepad, etc. You might find it easier to use another program to help search for key word or topics.

2015 Playoff Database Now Available


The 2015 Playoff Database is now available which completes the trio of DMB Projection Database entries. Included in this database are the 2015 Toronto Blue Jays, New York Yankees, Houston Astros, Kansas City Royals, Texas Rangers, New York Mets, Pittsburgh Pirates, Chicago Cubs, St. Louis Cardinals, and the Los Angeles Dodgers.  Here's your chance to manage all these teams in postseason play to determine the 2015 Worlds Champion!

The rosters and stats are as of beginning of play on Friday October 2, 2015 .