Deep, deep sabermetrics

DMZ · May 2, 2008 at 9:45 pm · Filed Under Mariners 

The M’s have a bad record in games they score fewer than five runs. This is because the average AL team, in an average game, scores just under five runs (right now, it’s 4.48, but it’s early in the season… it’ll perk up).

If you score fewer than that in any given game and you have average pitching and defense, you’re likely to lose. You won’t lose every game, because sometimes the other guy scores three, or two, or one, but in general, you’re going to lose.

I know, shocking, isn’t it?

Here’s a 2005 Studeman article, “Runs Per Game” that helps illustrate this, and includes this handy table of runs, team win percentage, and how much that last run increased their chances

   RS    Win%   Diff
    0    .000
    1    .077   .077
    2    .208   .131
    3    .339   .131
    4    .471   .132
    5    .593   .122
    6    .686   .092
...

Going from no runs to one run is good, but it’s not nearly as good as runs 2-5, or even six. The importance of piling on only really goes down at run 8.

So here’s the M’s so far

M\'s Run Distribution, 2008 season to date

Or, as Studeman put it,

For instance, if your league averaged five runs a game, and your team scored exactly five runs in every game, it would typically have a .600 winning percentage instead of .500, even though it had scored the average number of runs. That is the power of looking at distributions instead of averages.

Yup. Of course, as I ranted recently, you can’t build a consistent offense like that. Power-and-walk teams slump, contact-and-steal teams slump. All offenses are inconsistent.

But to return to the meandering point — if a team only scores four runs consistently with average pitching, you can expect them to win only about 45% of their games (and that’s me taking a guess looking at the win percentages).

Moreover, though, looking at Studeman’s research, and others, the thing that really sticks out is not that the fourth or the fifth run is particularly important — it’s that scoring that next run is always important. If you’ve got an anemic offense, scraping out that extra run a game however you can do it, be that player upgrades, an improved lineup, or whatever you want, makes a huge difference in how often your team will win. You don’t have to go to being an offensive powerhouse.

Hopefully the M’s can juggle Clement and the rest of the lineup and start winning games with the bats.

Comments

7 Responses to “Deep, deep sabermetrics”

  1. Sentinel on May 2nd, 2008 9:50 pm

    Interesting way to look at it. I’ve never seen it broken down that way.

    So, what kind of improvements can they make to eke out one or two more runs per game? I’m sure having a little more team speed wouldn’t hurt. Maybe a little more patience at the plate, as a team. Everyone seems to think they have Ichiro’s ability to put the bat to the ball.

  2. teacherrefpoet on May 2nd, 2008 10:20 pm

    I can’t believe a team wins as many as 7.7% of games where they score only 1 run. Feels high to me, but hey, the numbers don’t lie yatta yatta.

    That graph leaves off the M’s one time being shut out, no?

  3. Colm on May 2nd, 2008 11:02 pm

    So that’s 19 games where they’ve scored four or fewer vs 10 where they’ve scored 5 or more. And after tonight’s game that’s 20:10.

    It seems were lucky this team’s even got a 13-17 record.

  4. DMZ on May 2nd, 2008 11:36 pm

    It doesn’t leave out the time they got shut out — it gets counted in the ‘1’ bin because of the way the Excel histogram’s built.

  5. Mat on May 3rd, 2008 1:52 am

    I can’t believe a team wins as many as 7.7% of games where they score only 1 run. Feels high to me, but hey, the numbers don’t lie yatta yatta.

    Probably one reason it seems high is that there ought to be some degree of correlation between the winning and losing run total. If the winner had a hard time scoring runs, it could be at least in part due to bad weather that day, or playing in a pitcher’s park–both of which would also make it hard for the losing team to score runs.

  6. HamNasty on May 3rd, 2008 7:51 am

    Awesome post DMZ, guessing it was partially inspired by my comment at the end of the game thread. Since they are 4-15 in games when they score 4 or less are was blaming it on luck? The offense? Or a sad combination of both where the offense needs to stop sucking and we need lady luck on our side?

  7. turkeycave on May 5th, 2008 11:24 am

    Does this information in any way change the leverage considerations of earlier inning situations? If one run can make such a significant change in winning expectancy, does it make sense to consider “small ball” strategies earlier in games than is normally acceptable to modern baseball strategists?

Leave a Reply

You must be logged in to post a comment.