There have been plenty of studies on baserunning, and plenty of controversies surrounding it. I myself fall strongly into the category of aggressive baserunning wins ballgames. That said, I decided to take a look at all of the potentially baserunning situations and determining where the value was.
As such I looked at both stolen bases and advancing on a hit.
To conduct this study I took advantage of Baseball Prospectus' amazing run expectancy charts, and utilized the 2011 and 2012 seasons.
In an effort to not bore you with a ton of numbers and to get to the point I managed to calculate the changes in run expectancy for every potential situation of a stolen base or a caught stealing (with the exception of stealing home, because well if you steal home, in my mind the run expectancy is 1...)
A double steal of second and third with one out has the highest positive change in run expectancy. In that situation, with runners on first and second and one out the run expectancy for the inning is .8936, however after the double steal that jumps to 1.29. If the lead runner is caught, you still leave a runner in scoring position, but the run expectancy of a runner on second with 2 outs is only .3137.
Now you are saying well, you just lowered the run expectancy if you were caught by more than you gained if you stole successfully. That is correct, however consider this, any time you are caught stealing the damage is worse than if you successfully steal. For example with no outs, stealing second adds .2131 to the total run expectancy for the inning but getting caught costs you .5918 runs.
That said, the cost of outs in general is relatively high. Take a look at the following:
Cost of Outs with a runner on first and NO Outs.
Batter Out: .3474
Ground Into Double Play: .7533
Caught Stealing: .5918
So yes, if you are stealing there is a risk, so if you NEVER steal a base you will not risk losing runs. And that is the argument that many make regarding not stealing. That said, I went through and calculated the percentage of successful stolen bases you'd need to justify stealing. That number is 73.5%. So if you attempt 19 steals, you better be safe 15 times or you are hurting your team.
Of the 30 teams in 2011, 13 were successful more than 73.5% of the time. 8 of those teams were in the top 13 of stolen bases on the season. But stolen base prowess didn't correlate to winning. San Diego was easily the most successful team, converting 79.4% of their attempts, whereas St. Louis was the least successful converting 59.4% of their attempts. In fact, San Diego was caught only 5 more times than St. Louis, and stole 113 more bases. If you just pretend that they only ever stole second with no outs, San Diego added 24.08 more runs than St. Louis did via the stolen base. That said, we all know how 2011 ended, St. Louis won the World Series, and San Diego finished in dead last in the NL West with 71 wins.
Now then, time to defend why I feel the stolen base is still undervalued... If you need to be successful 73.5% of the time it means that 26.5% percent of the time you are caught... 48 catchers in 2011 caught fewer than 26.5% of attempted basestealers. Sure of those catchers barely played, but 13 of them started more than 60 games on the season. Some of the "names" Jason Varitek, A.J. Pierzynski, Josh Thole, J.P. Arencibia, John Buck, Carlos Ruiz. If you are facing any of these catchers and have decent speed chances are you can benefit from running.
Now I don't believe in reckless theft. For example, Juan Pierre last year was caught 17 times, and successful only 27 times. That is bad baserunning. Jacoby Ellsbury, 39 of 54, is just below-average. Coco Crisp stole 49 bases last year, in only 59 attempts, 84.5% success rate but his team was below the success threshold. That is because Jemile Weeks and Cliff Pennington were caught 20 times and were safe only 36 times. Both finishing in the top 20 for caught stealing (technically Crisp did as well, but 49 successful steals is a plus).
On the other hand, Texas featured Elvis Andrus, 37 of 49. Ian Kinsler 30 of 34, and Craig Gentry 18 of 18. The three of them combined to account for 85 stolen bases and 16 caught stealing. An exception 84.2% success rate. The team as a whole was 143 for 188, and accounted for 59.4% of the teams total stolen bases, but only 35.6% of the teams caught stealings. In other words, those three guys made up the most powerful trio of runners last season.
The reality is, the stolen base is an extremely valuable weapon that needs to be used in appropriate situations. Ian Kinsler for example is not a fast runner, in fact many scouts consider him a 45-50 in the speed category (ML Average), but in 2011 he was successful on 30 steal attempts (88.2% of the time), he picked his battles and won and in the four games he was caught, the Rangers won all four times. 2011 wasn't an anomaly though, for his Major League career he has been successful 83.8% of the time.
Advancing on the Hit:
A completely different theory to baserunning is advancing on a hit. For example going first to third on a single or scoring on a double from first. According to the Bill James Baseball Handbook 2012, Texas was the best team in 2011 in gaining bases... the worst, St. Louis. So again, baseball baffles.
The key in moving bases is to advance successfully without being thrown out. In the 2011 season Texas took 179 bases, and were thrown out only 15 times. A 90.2% success rate. Conversely, Arizona, took 159 bases and were thrown out 45 times. A 77.9% rate. For comparisons sake, the MLB average in 2011 was 85.2%. Perhaps most interestingly here is the fact that Texas took the second most bases, but were caught the fewest amount of times.
For the sake of this article, and because I am already getting far too wordy, I decided to only look at a runner on first advancing to third on a single. To do this I again consulted the run expectancy chart and found the run expectancy for a runner on first with no outs, one out, and two outs. From there I placed runners on certain bases based on results. What I found is that on a single, a successful advance from first to third (as opposed to just going station to station for one base) added .2487 to the run expectancy when there were no outs. With one out it adds .2498 to the run expectancy, however with two outs it merely adds .0408. In other words, with two outs, it is relatively pointless to risk advancing to third on a single.
Speaking of risk, what is the cost associated with attempting to take third on a single.
Well if you are the first out (In other words, runner on first one out now instead of first and second and no outs) and you cost your team .9298 runs. If you are the second out .6762. And if you make the third out trying to advance, .4344, but the inning is over. Again, risk-reward. A little less precise on my calculations here, but if you are successful in trying to advance 11 of the 14 times you attempt to do so with no outs (78.89%) you are adding to your overall run expectancy and helping your team. In other words, 29 of the 30 MLB teams were actually helping themselves when attempting to advance. Only Arizona fell below the 78.89% success rate cut-line.
With one out you only need to be successful 73.01% of the time in order to increase your run expectancy over the span of a season. However with two outs, you better be successful 91.41% of the time to justify attempting to take that extra base with two outs.
So the old adage, never try to take an extra base with two outs, is certainly true.
What I've managed to ascertain through this glimpse at baserunning is that in general the risk-reward payout is higher with no outs (duh) but the lowest risk and most reward comes with one out in all situations. The biggest reason I can think of for this is advancing or stealing typically takes you out of a double play situation. As such by taking the extra base if the next batter hits a routine groundball to short instead of being an easy force out the base-runner now has options. With no outs, if a double play occurs the inning continues, but with one out the inning is over.
All in all baserunning is an aspect of the game that in the past was highly valued, then tapered off in the 90s and 2000s and is slowly returning to prominence. I believe that a player with plus baserunning skills (not necessarily speed) is one of the most valuable tools a player can have. Especially when it is paired with a high frequency of reaching base.
Thanks for the patience, I hope some of this made sense. Feel free to ask any questions.
No comments:
Post a Comment