in Part 2 of the 1919 World Series series series we’ll cover Game 1 using a new box score report. Eight players on the White Sox got paid to help their team lose a World Series. These posts look at each game in detail to see if there’s any blatant statistical evidence of foul play. First we need to see how these two teams match up against each other according to this data model.
Team status records show data collected over an entire season which may not reflect upon their current roster. The above shows CIN had great pitching according to PITCH derived from runs scored against. White Sox had the opposite, great hitting derived from runs scored above average. Reds had much better Unearned runs above average which means good fielding.
1919 Playoff Horse Race
Horse race tables for playoff season are based upon estimated team rosters using end of year WAA data. During the regular year horse race tables use estimated rosters applying beginning of day data — what you would know reading the paper in the morning. During current year these tables are made using rosters downloaded from a source and are not estimated. This could change next season as our source sometimes can lag or have errors.
In modern baseball there would be 8 teams starting off in divisional series. Back in the day the World Series was the only playoff series so this horse race table reduces to two teams. This table kind of aligns with team status showing CIN with a much better starting rotation and CHA with a much better set of hitters.
White Sox starting rotation in the above table is much better than their PITCH in team status. They have a below average relief staff but relief was not nearly a factor then as it is now. More on these differences later in off season simulation talk.
The above shows the inning line with the Reds winning 9-1 on a a 5 run 4th. The presentation of this is still a work in progress.
Let’s look further into this box score.
For now the paid off players will be highlighted in bold red. Since I have to do this manually it won’t be done in subsequent games as much. The above is a prototype display showing columns this data model views as important. Plate Appearance (PA) represents time. R and RBI represent runs. TB , H , and W represent hits.
I haven’t doubled or triple checked the accuracy but pretty sure the above is correct.
Latest Update: All tables have been updated to the correct standard. Code was still being written during the first couple of these posts and bugs were found after posting. Developing the code to build these tables was the purpose of this series of posts. Now all World Series can be perused.
Update: Lineup tier numbers are wrong in lineups for both teams in this post. Part 3 has tiers properly calculated. Since the code that calculates tiers wasn’t working I tried to calculate lineup tiers manually and didn’t do it right. CHA lineup tier is a maxed out tier 5.00 as shown in Part 3 which was calculated automatically.
End of both Updates
Jackson scores the only run for CHA and reached second on an error. Chick Gandil gives up an unearned run due to an error but that was the only unearned run. This model can assign and tally unearned runs against fielders. This isn’t very interesting for regular season but it can be interesting in post season. Players can affect an outcome giving up a run by not fielding just as easily as not scoring a run by not hitting.
The Pitcher table shows outs instead of innings pitched, Plate appearances pitched, Earned Runs , Runs, and Strikeouts. Cicotte gave up 6 runs in 11 outs or 3 2/3 IP. The first game was probably essential for the White Sox to lose in order to signal the fixers they weren’t being cheated. Cicotte made that happen and no Sox player got clever trying to score. All hits were singles according to TB column.
Tiering for lineups is based upon average and standard deviation of end of season lineups for all 16 MLB teams. This can only be calculated with a mostly full set of day by day box score data. A Tier 1.08 means Reds’ lineup is around 1/2 standard deviation above MLB average for the year 1919. CHA is at 2.41 which is much greater.
Lineups face pitchers however, not other lineups.
Ruether was their best pitcher ranked #8 and pitched a complete game like most starters did back then. He gave up no earned runs.
Ruether is around Tier 3 and Cicotte around Tier 5. Even though on paper CHA as a team has worse pitching, when CIcotte pitches they are top top tier and back then starters usually pitched the entire game. Off the top of my head looking at the Tier numbers White Sox would have been favorites for this game. That alone doesn’t prove or disprove anything however.
The code to make handicapping tables for historical playoff games like what was done during 2019 current year coverage is not completely finished but will be for Game 2 hopefully soon. Until then ….