5.10 Value of Base Stealing

Is there a net benefit to attempting to steal a base?

## # A tibble: 2 × 3
##   event_cd            N   pct
##   <chr>           <int> <dbl>
## 1 Caught Stealing   713 0.244
## 2 Stolen Base      2213 0.756

What are common runners/outs situations for attempting a stolen base?

stealing |> 
  group_by(state) |> 
  summarize(N = n())
## # A tibble: 16 × 2
##    state     N
##    <chr> <int>
##  1 001 1     1
##  2 001 2     1
##  3 010 0    37
##  4 010 1   124
##  5 010 2   102
##  6 011 1     1
##  7 100 0   559
##  8 100 1   708
##  9 100 2   870
## 10 101 0    37
## 11 101 1    99
## 12 101 2   219
## 13 110 0    30
## 14 110 1    84
## 15 110 2    53
## 16 111 1     1

Most common: runner only on first

The corresponding run value reflects the success of the attempt.

## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

Focus on stolen base attempts

State of 100 1

## # A tibble: 2 × 3
##   event_cd            N   pct
##   <chr>           <int> <dbl>
## 1 Caught Stealing   210 0.297
## 2 Stolen Base       498 0.703

Runner stealing success = 70.3%

Broken down by New States:

## # A tibble: 4 × 3
##   new_state     N     pct
##   <chr>     <int>   <dbl>
## 1 000 1         1 0.00141
## 2 000 2       211 0.298  
## 3 001 1        39 0.0551 
## 4 010 1       457 0.645

  • 457 occurrences of successfully advancing to second

  • 39 occurrences of successfully advancing to third

Value of attempting stolen bases in this situation

stealing_1001 |> 
  summarize(Mean = mean(run_value))
## # A tibble: 1 × 1
##      Mean
##     <dbl>
## 1 0.00723

Stolen bases are worthwhile but overall value is 0.007 runs/attempt

BUT benefit really depends on success/failure and runners/outs situations