More Insights on One-Run Games

My prior AZ Snake Pit article examined eight possible causes of losses in 1-run games. It concluded that the bullpen was a root cause and baserunning & defense were contributing causes.

Although that was a good start, let’s look at more amazing insights about one-run games!

The win-loss record provided insights.

In 31 one-run games, the Diamondbacks were 4.4 wins short of reaching the 36.8% winning percentage for games won or lost by more than one run (through 23 August).

“One run games can go either way, and most of the time they do.” — Lance Berkman

The good news is that bad luck likely accounts for most of the Diamondbacks’ results in one-run games. The next question is, “Will that bad luck will persist in the future?” About 20 years ago, Bill James had some interesting ideas which applied to teams that were five wins better than expected or 5 wins worse than expected in one-run games. The relevant idea was that when a team lost a lot of one-run games, the bad mojo continued in the following season.

Stupendously important was the one-run win on 17 August. That win kept the Diamondbacks out of the statistical group for which bad luck persists in the following season.

What happened in the ninth inning provided insights.

When the top of the ninth inning started, the Diamondbacks were:

• Ahead by 2 or more runs, their win-loss record was 1-1. The win was a home game.
• Ahead by 1 run, their win-loss record was 2-1. The wins were away games.
• Tied, their win loss record was 3-6. All 3 wins were home games.
• Behind by 1 run, their win-loss record was 1-15. The win was a home game; the game went into extra innings, which involved more luck than normal innings.
• Behind by 2 or more runs, their win loss record was 0-1.

Observations follow:

• When the Diamondbacks were behind by one run at the top of the ninth, they did not score in 14 of 16 one-run games. Two games went into extra innings – they won one and lost one. Scoring more runs in the ninth inning could help the Diamondbacks win games.
• Five of the seven wins were home games, of which one was in June and four were in July and August. Only one away win happened in July & August. Recently, home field advantage was important.
• In 17 of 31 one-run games, the Dbacks used a pinch hitter in the ninth inning. Win loss record in those games was: 4-13 (23.5% wins). When no pinch hitter was used, the win loss record was 3-11 (21.4% wins). The winning percentage was only very slightly better when a pinch hitter entered the game in the ninth inning. Perhaps a reliable power hitter on the bench could help the Diamondbacks win games.

For the NL West, Joe Posnanski’s rule is 80% accurate.

“As a rule, good teams lose more one-run games than normal, and bad teams win more.” — Joe Posnanski, June 2018

The first part of the rule. In games through 14 August, the teams with the best win-loss records (64.7% Giants, 60.3% Dodgers, and 55.9% Padres) all lost more 1-run games than normal as reflected by their lower winning percentages (58.8% Giants, 40.0% Dodgers, and 47.1% Padres). The first part of the rule predicted accurately.

The second part of the rule. In games through 14 August, the teams with the worst win-loss records were the 44.0% Rockies and 31.6% Diamondbacks. The second part of the rule:

• The Rockies won 45.7% of one-run games, which was more than normal as the rule predicts.
• The Diamondbacks won 20.7% of one-run games, which was less than normal (different than the rule predicts). Perhaps the Diamondbacks had bad luck, and they will tally more one-run wins in the rest of the season.

For NL West teams, three statistics are characteristic (instead of predictive) of teams that lose a lot of one-run games.

Runners Left on Base (LOB). Teams that lose a lot of 1-run games tend to have many baserunners who do not score.

Defensive Runs Saved (DRS) up the middle: second base, short stop, and center field. Teams that lose a lot of 1-run games tend to have poor defense up the middle of the field.

Wins Above Average (WAA) for the bullpen. Teams that lose a lot of 1-run games tend to have low WAAs for their bullpen. Although Rob Arthur showed a “fairly weak” correlation with fWAR over 29 seasons, this season in the NL West WAA is a much stronger characteristic.

Let’s look at these statistics for the NL West teams. The following table shows that Diamondbacks ranked second highest in runners left on base, had the worst Defensive Runs Saves (DRS) up the middle, and had the lowest Wins Above Average (WAA) for the bullpen. Those characteristics were consistent with the Diamondbacks having the lowest winning percentage in one-run games.

Games through 14 August. Data from Baseball Reference and The Fielding Bible.

Ideas for improving the Diamondbacks next season.

Having a reliable power hitter who could pinch hit in the ninth inning could help the Diamondbacks win more games. The Diamondbacks began the top of the ninth inning behind by one run in most (15 of 24) of their one-run losses.

Perhaps batters could win more one-run games if mentally they bat like it’s the bottom of the inning during away games. Most (5 of the 7) of the one-run wins were home games.

Excellent defenders at second base and center field could help the Diamondbacks win more games. A characteristic of teams that win a lot of one-run games is excellent defense up the middle of the field.

The latest 1-run loss (23 August to the Pirates) put an exclamation mark to the idea that better bullpen pitching could help the Diamondbacks win more games. Strongly supporting that conclusion are WAAs in the NL West and the previous AZ Snake Pit article that concluded that the bullpen was a root cause for 1-run losses.