THE BOOK--Playing The Percentages In Baseball

Thanks, Tango.  Does the presence of other base runners have an appreciable impact on the numbers?  For instance, with two out and the bases loaded, what are the chances of scoring from third?  How does it compare to overall chances of scoring from third with two out?  Which comparisons like that make the biggest difference?

Tango, what do you mean “chance of scoring?” For example, with a runner on first, what constitutes “scoring?” What about a DP and then a HR?  What about a force out at second and the the new runner scores?  Etc.  Can youe more specific what the numbers mean?

The chance of that particular runner in scoring during that inning.

So, at the start of a PA, our runner is on 1B with 0 outs.  At the start of the next PA, he’s on 2B with 1 out.  By the end of the inning, he scores a run.  So, he gets a success and an opportunity for 1B, 0 outs, and a success and an opp for 2B, 1 out.

"that particular runner”

So if a runner gets erased on a force, DP, CS, etc., there can be no run scored for that opp?

Even if he is replaced by someone else, as in a force?

Right.

***

However, you can infer it from here, if you are interested in force plays, etc:
http://www.tangotiger.net/RE9902score.html

If there is a runner on 3B with 0 outs, there is a .136 chance of NOT scoring a run in that inning.  So, the chance of scoring at least one run is .864 (that’s for 1999-2002).

We know in my link at the top of this page shows:
0.842 as the chance of the runner from 3B with 0 outs to eventually score (that’s for 1993-2007).

Presuming those numbers would be the same for the same time periods, then the guy who was on 3B seems to have been thrown out in favor of someone else either taking his spot, or getting other runners/batter to move up extra bases.

So, you can use the frequency chart to figure out how often a runner would score, to also include all the force issues, etc.

Remember, just look at the base/out charts for the lead runner, and compare that to the numbers I provided.

Hope that makes sense…

And pinch runner issues, etc.

Oh yeah, pinch runners.  If a player is replaced by a pinch runner and the pinch runner scores, what is that considered?  I am assuming that is NOT considered a score?

It may seem obvious to you, what “that particular runner” means, but it is not obvious to me.

And certainly from an analysis viewpoint, we don’t really care about “that particular runner of the same name.” Especially with pinch runners.  If there is a runner on 1st, I want to know how often he scores, including any pinch runners for him.

I think a lot of people are going to assume you mean “a run is scored at all, regardless of who scores it.” Especially with regards to pinch runners and being replaced after a force out.

For example, when announcers talk about a “leadoff walk going on to score a run,” I assume that includes when he is replaced by another runner on a force play.

Tango,

Do you have the frequencies for the 12 states where you give the chance of scoring?

The reason I asked for the frequencies is that I found a spreadsheet that you created about 8 years ago, where you calculated linear weights based on the chance of scoring, frequency that the base was occupied, and the percentage of the time that an extra-base was taken on a single or double.

Anyway, here is that spreadsheet. I don’t know where I found it at, Internet Archive I think. You can export the spreadsheet with the formulas intact from this link:

http://spreadsheets.google.com/ccc?key=pzy9IhjJPqasjMeVtsRRxEw

How would you calculate the value of the Out?

The frequency that each base is occupied can be calculate from an RE table. I used Jarvis’ data for the frequency of taking the extra-base. I still can’t get the getting-on-base component to work out correctly.