Math and history on Mets side

Posted on May 3, 2010 by

0


Well that did not go as planned.

After winning in a laugher in the opener of a three-game set versus the Phillies, the Mets got blown out in back-to-back days, including Sunday night when ace Johan Santana had the worst start of his career. Now more than ever, fans are wondering – How good are the 2010 Mets?

The Mets just ran off a streak of 10 wins in 11 games. Now in that streak there were several games where the opposing team’s inept play had just as much to do with the Mets winning as anything else. Still, how often does a bad team win 10 out of 11 games, regardless of the circumstances?

For a moment, let’s pretend that winning a baseball game is like flipping a coin, in that a team is just as likely to win as to lose (a .500 team) and that what has happened in the previous game(s) has no bearing on what happens in the next game.

In this scenario, there are 11 ways to win 10 out of 11 games and there are 2,048 potential outcomes. If you divide 11 by 2,048 you get 0.005371. But during a 162-game season, there are 152 different 11-game stretches for a team to win 10 games. So, if we multiply 0.005371*152 we get 0.8616392 – or mathematically we would expect a .500 team to win 10 out of 11 less than 1 time in a 162-game season.

However, we think that winning baseball games is not exactly like a flip of the coin, that things like chemistry and momentum may exist. So, let’s go back to the start of the 162-game schedule and look at every team to finish exactly .500 (excluding strike years of 1972, 1981, 2004 and 2005) and see how many won 10 out of 11 games during the season.

2008 Indians – yes
2005 Nationals – yes
2005 Brewers – no
2002 White Sox – no
1996 Reds – no
1993 Dodgers – yes
1992 Astros – no
1989 Expos – no
1989 Brewers – yes
1988 Expos – yes
87 A’s – no
85 Red Sox – yes
84 Angels – no
84 Twins – no
84 Phillies – yes
83 Padres – no
82 Padres – yes
77 Cubs – yes
77 Astros – yes
74 White Sox – no (team finished 80-80)
74 Astros – no
73 Cardinals – yes
73 Twins – no
69 Astros – yes
68 Braves – no
67 Pirates – no
66 Indians – yes

In this sample of 27 teams, we see that the number of times a .500 team won 10 out of 11 was 13, or less than that predicted by our mathematical equation. Now, this could be due to intangibles, injuries or normal fluctuations in the data. It should be noted that each team was treated as a “yes” or “no” and that several of the “yes” teams did it more than once in a year. For instance, the 2005 Nationals won 10 of 11 five times during the season (by using different end points during the streak when they won 15 out of 17) but were only counted once. If the multiple times that certain teams won 10 of 11 were included, our overall numbers would be different and possibly right where our raw mathematics would indicate.

Now, let’s look at it from a different point of view.

This time let’s take all of the teams from the past three years, see how many of them won 10 out of 11 games and check their record.

09 Phillies: 93-69 (.574)
09 Marlins: 87-75 (.537)
09 Rockies: 92-70 (.568)
09 Padres: 75-87 (.463)
09 Yankees: 103-59 (.636)
09 Red Sox: 95-67 (.586)
09 Twins: 87-76 (.534)
09 Angels: 97-65 (.599)
08 Rays: 97-65 (.599)
08 Red Sox: 95-67 (.586)
08 Yankees: 89-73 (.549)
08 Blue Jays: 86-76 (.531)
08 Twins: 88-75 (.540)
08 Indians: 81-81 (.500)
08 Royals: 75-87 (.463)
08 Phillies: 92-70 (.568)
08 Mets: 89-73 (.549)
08 Cubs: 97-64 (.602)
08 Astros: 86-75 (.534)
08 Dodgers: 84-78 (.519)
07 Cubs: 85-77 (.525)
07 Brewers: 83-79 (.512)
07 Diamondbacks: 90-72 (.556)
07 Rockies: 90-73 (.552)
07 Giants: 71-91 (.438)
07 Yankees: 94-68 (.580)
07 Indians: 96-66 (.593)
07 Tigers: 88-74 (.543)
07 Mariners: 88-74 (.543)

The mean is 88.7, the median is 89 and the mode is 88, 97. By wins it breaks down as follows:

< 80 – 3
81-85 – 4
86-90 – 11
91-95 – 6
96+ – 5

Assuming this is a large enough sample, there is a 90% chance that the 2010 Mets are at least a .500 team and an 86% chance they are better than .500 this year. Looking at the data another way, over the past three years, 41 teams finished below .500 and only three of those won 10 games out of 11. That works out to 7%.

There is nothing magical about winning 10 games out of 11 during the season. However, it is clearly something that bad to average teams do not do on a regular basis. With nothing else but this information to go on, one would conclude that the most-likely scenario is that the 2010 Mets are an 86-90 win team.

Advertisements
Posted in: Perspectives