View Full Version : Stat I'd like to have.
06-26-2006, 12:20 PM
There has been a lot of talk on this board concerning the value of a advancing runners that are already on base, and the relative value of a productive out vs a hit vs a walk. This got me thinking about a different way to measure a batters effectiveness in these situations towards the goal of advancing other runners. I'm really not sure is there is a way to get these stats publicy but the formula I am thinking about would have these variables.
X = total runners on base when a batter was at the plate
Y = total bases acquired by those runners as a function of the batters event (walk,hit, out etc)
Z = total number of outs created in these situations
Is there a way to play with these variable to come up with a way of measuring a players effectiveness at advancing other runners?
Could it be as simple as Y-X-Z or would it need to be more complex?
06-27-2006, 01:28 AM
I'd like to see that stat as well, but I'd have no clue where to start...
06-27-2006, 01:34 AM
There is a table below to show the effectiveness of advancing runners. At first I misread your post and posted this response, sorry if it is off topic, but I'll leave it up so others can see.
06-27-2006, 01:37 AM
The table is a little off. But here is the explanation.
Explanation of these tables: For example, in innings where the team at bat has a man on second with no one out, they add an average of 1.051 (EFR) runs in that inning. According to this study, an attempt to steal second with no outs risks a loss of .576 runs (.815 - .239) for a gain of .236 runs (1.051 - .815). This means that stealing second with no outs must be successful more than 70.9% of the time to increase a team's expected runs. The formula is:
(Start EFR - Fail EFR) divided by (Success EFR - Fail EFR)
Table on left is total number of runs scored in those situations, table on right is percentage of times a team scores atleast one run in those situations.
Entry from February 1999 issue of By The Numbers, a publication by The Society for American Baseball Research
Stolen Base Strategies Revisited
EFR = Expected Future Runs
Average runs scored in these situations
Percentage of times a team scores at least one run
Men on Number of Outs Men on Number of Outs
FST 0 1 2 FST 0 1 2
- - - .452 .239 .091 - - - .261 .148 .061
x - - .815 .486 .210 x - - .424 .267 .124
- x - 1.051 .653 .313 - x - .608 .400 .216
x x - 1.384 .852 .404 x x - .617 .413 .219
- - x 1.278 .912 .358 - - x .819 .651 .269
x - x 1.638 1.131 .471 x - x .843 .646 .275
- x x 1.884 1.313 .576 - x x .841 .667 .274
x x x 2.176 1.481 .718 x x x .855 .664 .316
I can't remember the exact time frame right now, but these number were taken from about 10 seasons worth of games, in the early 90's IIRC. So things have changed since then.
Here's what we can see: Man on 1st, with no outs, a team scored atleast a run 42.4% of the time. In situations with a man on 2nd and 1 out (the result of a sac in many scenarios) the percentage actually drops to 40%. So judging by these numbers, a sacrifice is not beneficial.
06-27-2006, 10:25 AM
Just to reiterate a point: sacrificing a runner to second actually decreases a team's chances of scoring.
06-27-2006, 10:45 AM
I'd love to see this stat too Pedro, the problem is in the data collection. Even then, it would be complicated by things like certain situations (defensive indifference) and the speed of the runners ahead of you.
I think the best way to do it is too come up with your basic assumptions about what "should" happen given a certain event in a given situation (ie. all runners advance 2 bases given a double). Then simply look at how many times a player performances each event in each situation and do the math. It would be unfair to penalize Austin Kearns for Scott Hatteberg being unable to score from second on a single...
06-27-2006, 10:49 AM
Slightly different angle, but here's a chart that shows the Reds are second in the majors in "not losing runners." That is, not grounding into double plays and not getting caught stealing.
vBulletin® v3.8.7, Copyright ©2000-2013, vBulletin Solutions, Inc.