Is offensive consistency better?

by Travis G on Jun 22, 2009 1:00 PM EDT in New York Yankees Analysis

We've heard about how inconsistent the Yankee offense is, but is it true? They went from scoring 15 runs against one of the best pitchers in baseball to scoring seven runs in three games against the god-awful Nationals.

Using Standard Deviation, I studied the offensive consistency of AL teams. The lower the STDEV, the more consistent the scoring is.

Through Sunday, June 21 -

From most consistent to least consistent

M's: 2.63

A's: 2.66

Angels: 2.82

Royals: 2.88

O's: 2.93

Yankees: 3.14

Red Sox: 3.20

Rangers: 3.24

White Sox: 3.35

Jays: 3.43

Tigers: 3.49

Rays: 3.68

Twins: 3.70

Indians: 3.74

As you can see, the better offensive teams are grouped toward the bottom, but does this correlate with actual runs scored?

Overall runs/game:

1. Rays (5.59)

Yanks

Red Sox

Indians

Jays

Angels

7. Rangers (4.91)

Tigers

Twins

O's

A's

White Sox

Royals

14. M's (3.78)

So while Seattle has the most consistent offense, it doesn't help them score. Conversely, Cleveland has the least consistent, yet one of the highest-scoring offenses.

Unfortunately, Standard Deviation doesn't tell the whole story. Because while Seattle or Chicago score consistently, it's often low (usually 1-5 runs), whereas teams like Cleveland or Tampa also have those stretches, but then they'll put up nine, 11, 15, even 22 on occasion, which throws their consistency off; but those high-run games are (needless to say) more valuable than low-run games.

What might be more helpful is scoring 5-6 runs consistently rather than 2 one day and 10 the next.

Stay tuned for Part II in the coming days.

(Couldn't be done without BRef and MS Excel)

Tweet Comment 16 comments | 0 recs |

Do you like this story?

Comments

I dont care how many runs we score

so long as its one more than the other team.

by ryanwk628 on Jun 22, 2009 1:50 PM EDT reply actions

this is a good idea and worth trying to measure

i don’t think SD is the way to go- we need binned histograms showing number of times in a row that the team scored 3, 4, 5, or 6 runs or more. Then we can compare the distributions.

by epoxypatch on Jun 22, 2009 1:51 PM EDT reply actions

the key question is how consistently the team scored more than lets say 4-5 runs every time. That team is going to dominate. Standard deviation won’t tell you that.

by epoxypatch on Jun 22, 2009 1:53 PM EDT up reply actions

Here you go...

3 or fewer runs: 19 times
4 or 5 runs: 23 times
6+ runs: 26 times

by New York Sports Jerk on Jun 22, 2009 1:58 PM EDT up reply actions

And I missed a game

Somewhere, but whatver.

by New York Sports Jerk on Jun 22, 2009 2:00 PM EDT up reply actions

Interesting

post. Just some points:
a)SD is only a good measure of variability if the distribution under question is Gaussian. Runs scored cannot be gaussian since it can only take on positive values. However it may approximate Gaussian near the center. It would be good to look at the full distribution.
b)It would be interesting to look at “consistency” vis deviation from expected runs score via Pythag.

by Buzzy on Jun 22, 2009 1:57 PM EDT reply actions

I would rather

Score 5 runs a game in a 3 game series than 13 runs in one game and 1 a piece the other two. Wouldn’t that average to be the same? 5 runs per game on paper? But the first example probably would give you a better winning % over a longer period than the latter.

by HappyLuckyGoldenDragonNumber1! on Jun 22, 2009 2:16 PM EDT reply actions

Hitting

I feel as if our team always starts slow and we always have that late 8th and 9th inning push but end up losing. I think its time for a change Girardi

by dan.altman89 on Jun 22, 2009 3:27 PM EDT reply actions

A win only counts as 1 win

and a loss as 1 loss, regardless of the margin of victory. Sure it’s more encouraging if you see your team has the firepower to post big innings and big game totals but it’s really immaterial. When the Indians beat us 22-4 the W-L column the next day looked the same as if it were 5-4. I’d like to see the statistics on STDEV from series to series on BA w/ RISP. I feel like the Yankees bat .750 with RISP one series and then .050 the next.

by ScottBrosiusWSMVP on Jun 22, 2009 3:52 PM EDT reply actions