In baseball, like all other sports, consistency is the name of the game. While it may be exciting for fans to watch their favorite team explode for 14 runs after scoring 1, 2, and 3 runs in the previous three games, it would probably have been better for the team record-wise to score 5 fiveruns in each game.
That being said, there have been a number of MLB teams so far this season that have an actual winning percentage which is a good deal better or worse than their Pythagorean winning percentage (a formula which calculates expected winning percentage based on runs scored and runs allowed). I chose to write about only a few of the more interesting teams in order to keep my effort level at a minimum.
To get a handle on how consistent the teams I picked were doing, I looked at the variance (standard deviation squared) of the runs they have scored and the runs they have allowed. There are probably other statistics one can look at to explain the question at hand, but I chose variance and the results I found were interesting enough for me to write them up.
New York Yankees
The Yankees' actual winning percentage is only .018 lower than its Pythagorean winning percentage, but it is always fun to write about the Yankees, whether it is to praise them or to bash them.
There does not seem to be a problem with the Yankees offense on the surface since they are averaging 5.8 runs per game, highest among the teams I looked at. Their ability to prevent runs is also fairly good as evidenced by the Yankees allowing an average of 4.8 runs per game.
However, the Yankees offense has a variance of 13.6, which is second highest among the teams I looked at. This variance indicates that there are some games where the Yankees will explode for a large number of runs and other games where they can barely scrape together any runs. While the Yankees have a high margin for error since they average +1 run per game, if their offense had been more consistent, they would right now be leading the AL East instead of being two games back of the lead.
The blame cannot be blamed on the team's defense because they have been relatively consistent with a variance of 9.4, a little lower variance than the average of the over-performing and under-performing teams.
Boston Red Sox
The team who the Yankees are trailing in the race for the AL East division lead is the Boston Red Sox, whose actual winning percentage right now is .052 higher than their Pythagorean winning percentage.
While Boston's average runs scored of 5.4 runs per game and average runs allowed of 4.9 runs per game are not better than the Yankees' averages, their consistency is much better.
The Red Sox offense has a variance of 9.5 (9.1 before their 11-3 triumph over the Washington Nationals) and the defense variance of 9.0 go a long way in explaining why the Red Sox have won more games than the Yankees although their run differential is worse. With this level of consistency, the Red Sox have a higher chance of putting up run totals in games that approach their averages instead of being all over the board. That ability has certainly helped them so far.
If you are a Cleveland Indians fan, then before you read this I suggest you take a shot of the hardest liquor you can find. In fact, you might be better off downing the whole bottle because of the 14 teams whose variance I calculated, the Indians had the highest variance for both runs scored and runs allowed and the second lowest differential between actual winning percentage and Pythagorean winning percentage (-.075).
Whereas most teams may have been inconsistent in only one department, Cleveland put up a variance of 13.8 for runs scored and 14.9 for runs allowed. Therefore, spectators really have no idea which Indians team will show up. They may play brilliantly on offense and defense, play brilliantly on offense only, play brilliantly on defense only, or play poorly on both sides. While the same is true for all teams, at least with most other teams there is a trend for one side or the other. With the Indians, there is no trend except their penchant for being inconsistent.
And their inconsistently certainly overrides their average run differential of +0.7 runs per game.
Right now, the A's are in a spot where they should not be, tied for the lead atop the AL West. By all rights, based on their run differential, the A's winning percentage should be .507 and not the .535 it actually is.
Even their average runs scored and runs allowed indicate the A's are closer to a .500 team (4.5 to 4.4). Their saving grace has been their consistency in scoring runs with a variance of 7.8 (highest among studied teams), surprisingly, since they have a reputation and a justified one at that for being a team with an anemic offense. Still, even if a team is only averaging 4.5 runs a game, the fact they can do so consistently always helps.
While the ability to prevent runs has not been as consistent (variance of 9.7) as the offense, it has been consistent enough to not hurt the offense so far and bring the team down to its Pythagorean winning percentage.
New York Mets
The New York Mets look to be the strongest team in a National League while playing in a division with no real competition and they are also out-performing their Pythagorean winning percentage by .029.
The way the Mets are out-performing their run differential may surprise those who have fallen in love with David Wright because they are doing it mostly based on the consistency of their ability to prevent runs (variance of 7.4). Their offense is not nearly as consistent with a variance of 10.8, but that is not looking like a problem for the Mets.
Combine their extremely consistent run prevention with their already high margin for error (+1 run per game) and then it becomes very apparent how the Mets have been out-perfoming themselves.
St. Louis Cardinals
St. Louis is another team that is out-performing their actual winning percentage, doing so by .049. They have done so largely by being supremely consistent in terms of both scoring runs and disallowing runs.
Before the White Sox game last night where the Cardinals lost 20-6, they were the most overall consistent team of the 14 teams I studied with a variance of 8.3 for runs scored and 7.1 for runs allowed. After that game, their variance in runs allowed jumped up to 10.6 while their variance for runs scored stayed the same, but this is not to say they are not still very consistent when it comes to preventing runs.
It will be interesting to see if the Cardinals can maintain their consistency in scoring runs when Pujols returns.
With an actual winning percentage .095 lower than their Pythagorean winning percentage, the Pittsburgh Pirates have the lowest differential between those two winning percentages in all of MLB. A dubious honor, to say the least.
Yet, the Pirates with a variance of 11.2 in runs scored and 8.9 in runs allowed are nowhere near as inconsistent as the Cleveland Indians and are even more consistent than the New York Yankees. So why are they so much lower than their teams in percentage differential?
The answer lies in what it means to be inconsistent. Inconsistency brings a team closer to .500. For an above .500 team, like the Yankees, it means that the team will do worse than expected. For a below .500 team, like the Pirates, inconsistency makes a team do better than expected since the team is being drawn closer to .500. Therefore, the Pirates would probably benefit from being more inconsistent.
Los Angeles Dodgers
The Dodgers should not be a game behind the San Diego Padres in the NL West and yet they are, due to under-performing their Pythagorean winning percentage by .049. The reason for their underwhelming performance so far is a combination of things.
With a variance of 9.8 in runs scored and 9.2 in runs allowed, the Dodgers are not greatly inconsistent in any one aspect, but they are also not consistent enough to override the fact they only have an average run differential of +0.6 runs per game. This low run differential leaves no room for any sort of inconsistency.
So far, the Dodgers are coming to learn that.