Tag Archives: losing streaks

Losing Consecutively

The Twins starting the season with a nine-game losing streak was not good. Good teams don't lose a lot of games and even more so don't lose them in long streaks. That got me wondering about how teams with losing streaks fared overall.

It was pointed out here that the Twins streak was more noteworthy because it started the season so it looks even worse. Baseball-Reference only allows searching from the beginning of the season if you want to include all teams over all years so that's the best quick investigations can do. In order to search throughout the season, some work needed to be done.

I processed every season in the modern era through last year and grouped all games together by their streaks. A stretch that went WLLWWW would be a one-game winning streak, a two-game losing streak, and a three-game winning streak. Some brief spot checking of last year's data showed the streaks were properly classified. I merged that with the season results for those seasons in order to associate how teams did given a streak of a certain length.

To start, I decided to look at the range of results for teams with at least a losing streak of every length. The full dataset contains every season since 1900. Here are the number of seasons that contained a losing streak of this length from 1900 through 2015.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 23
2117 2116 2102 1896 1469 1014 605 377 210 130 73 43 27 19 9 2 4 3 4 4 1 1

First thing that popped out to me was the 2117 seasons with a one-game losing streak but 2116 seasons with a two-game losing streak. I checked the data to find the season, looked it up and determined it was right. If you want to know which team it is, the answer is at the bottom.

I did not want to have to normalize the seasons to match 162-game seasons nor have to deal with seasons shortened by a strike so I processed only the 162-game seasons. Converting that to a box plot for a pretty picture gives us this.

Boxplot of seasons with a losing streak of a given length

The widths of the box plots represent the number of seasons. The total number of seasons in the data set are 1234. In tabular form it looks like the following.

Losing streak length Seasons with that streak 95th percentile 75th percentile Median 25th percentile 5th percentile
1 1234 109 89 81.0 73.0 50
2 1234 109 89 81.0 73.0 50
3 1232 109 89 81.0 72.5 50
4 1121 109 88 80.0 72.0 50
5 882 109 86 78.0 70.0 50
6 605 108 84 76.0 68.0 50
7 361 102 82 75.0 67.0 51
8 229 103 82 73.0 66.0 43
9 121 95 77 71.0 64.0 51
10 66 92 76 68.0 64.0 50
11 37 89 76 68.0 62.0 43
12 23 91 73 67.0 59.0 53
13 12 77 70 67.0 63.0 60
14 7 70 66 63.0 60.5 60
15 4 65 60 53.0 51.0 51
17 3 67 64 61.0 50.5 40
19 2 57 57 56.5 56.0 56
20 1 52 52 52.0 52.0 52
21 1 54 54 54.0 54.0 54

Nothing knowing except the Twins having a nine-game losing streak, we would expect them to finish with 71 wins. We do know more than that and the preseason projections were not as kind to project the Twins for 81 wins. Many of them were closer to the high 70s so somewhere between the median and 25th percentile marks seems a more likely scenario.

I did another check on what happens if the losing streak happens in the first quarter of the season but the results are very similar. The sample size also becomes a problem as you get into the longer streaks. So I am skipping that and moving on to postseason probabilities.

It was already known at the beginning of the season that reaching the postseason would be unlikely. Well, losing nine straight hurts that a lot.

Probability of reaching postseason

Note the "0" does not mean did not lose at all but instead is a placeholder to mean the average for all seasons.

Good thing they avoided losing ten games instead of nine, unless they managed to stretch it to eleven games. Not much to say here other than don't lose a lot of games if you want to make the postseason.

Spoiler: Postseason probability data SelectShow

In conclusion, don't expect good things this year. The team started out with marginal chances for reaching 81 wins and now they're trying to do it in 153 games.




Spoiler: Team that never had a two-game losing streak SelectShow

Game 9 Recap: Minnesota 3, Texas 4

Jackie Robinson Day. Jack Roosevelt Robinson broke the color barrier in baseball when he joined the Brooklyn Dodgers on April 15, 1947. He played 9 years in the majors and was inducted into the Baseball Hall of Fame in 1962. I could go on, but if you watched baseball today, you already know everything there is to know about #42. Sh*t, even if you didn't watch baseball today you probably already know all about Jackie. He is that significant to the Civil Rights era and the eventual improvement of race-relations in America; a true icon.  His inclusion and eventual debut in the majors was carefully orchestrated by Branch Rickey (to me, one of the more interesting characters in MLB history). In case you're still wondering, this game came nowhere near the historic magnitude of Jackie Robinson or even Branch Rickey.

The game ended poorly for Glen Perkins and he took the loss. Robbie Ross took over in the 6th and pitched 2 scoreless inning for the Rangers to notch his 2nd win. Through the early innings, it was the Clete Thomas and Liam Hendriks show. In their season debut for the Local 9, the former Twins draft pick (and recent Tiger) jacked a 2-run homer and had a nice (sort-of?) outfield assist, and the Aussie pitched 6 innings of 1 run ball. After completing 7, the Twins had a 3-1 lead and the top of the order due up. It should have been enough to get the Twins back in the win column, ready to salvage the rubber match of the series against the Dallas, Texas Rangers of Arlington.  Except it wasn't... Continue reading Game 9 Recap: Minnesota 3, Texas 4