This set of posts will cover the 1919 World Series. It has been around a month since the last baseball game of the 2019 season. Since then much work has been done to shore up the historical dataset and scripts that support it. A league snapshot is taken at the end of every day. Daily snapshots must add up to match the official end of year tallies which can be a challenge and missing games introduce some error which is often acceptable.
Daily data is derived from retrosheet.org compilations of event box (1910 – 1949) and play-by-play (1950-2018). Although event box data goes back to 1910 and before, 1919 is the first year where dailies add up to be close enough to known final stats to include into the historical dataset. TC Sim relied on data from 1970-2018. Now we will go back to 1919 meaning simulation will draw from an entire century of baseball games.
Conveniently the 1919 World Series happened 100 years ago and something very exciting occurred. Eight White Sox players purportedly threw the series for money; handicappers/mob were tilting the odds the only way they knew how before computers. Since the proof of this data model rests on accurate handicapping of baseball games, let’s see how well it does for the 8 games where CIN beats CHA 5-3. World Series were 9 game series back then.
In Part 1 we’ll cover high level basics between the two teams and in subsequent parts show playoff horse race tables for August and end of year. Then we’ll handicap all games using the same tables shown here for the 2019 World Series. Since we’re from the future we can also show how these games end with box scores and post mortem analysis.
First let’s look at the top half of MLB. This model does not discriminate between AL and NL; all teams and players are ranked together.
Cincinnati had the best record in baseball with White Sox second best. Reds had great pitching, White Sox great hitting. Reds had better fielding according to Unearned Runs above average. Back then there were far more errors committed than in modern baseball. Unearned runs count the same as Earned runs in determining who wins a baseball game. For some reason they only played 140 games that year.
Top CIN Players 1919
Reds had 7 players ranked in top 50 which is equivalent to top 100 for a 30 team league. Five of those seven are pitchers as one would expect based upon their PITCH in team status above.
Top CHA Players 1919
White Sox also had 7 guys ranked around top 50 with 5 of those 7 hitters as one would expect based upon their BAT in team status. Highlighted in bold red are the 8 players banished from baseball. Swede Risberg was appended to the end of this list to round out the 8 players. Eddie Collins was their only top player not part of the fix and was inducted into HOF after a very long career.
Based upon end of year stats these two teams seem very evenly matched with CIN perhaps slightly ahead. Having the above 8 players not play to their potential moves the handicapping needle towards CIN. Nothing in handicapping can be a sure thing and according to many historians many of the above played to their potential. We’ll see about that in subsequent parts to this series.
To round out this high level overview let’s look at the top ten MLB players according to this data model 100 years ago at the end of the 1919 season.
Top MLB Players 1919
Usual suspects round out the top ten. Top ten in 1919 is like top 20 in 2019 because there were half the teams thus half the players. Cubs had two pitchers in the top ten and White Sox had a pitcher and a hitter. White Sox pitching was their weakness with Cicotte their one almost sure win every time he pitched. Did he throw down? We’ll see.
The year 1919 happens to be the base year for this data model because it’s the earliest with a workable complete set of dailies for the season. Pretty much all MLB playoff games have play by play dailies. Integration between the regular season historical data and post season historical data is a work in progress and the reason for these posts.