For years now, decades even, the sacrifice bunt has been declining as a tool employed by Major League Baseball teams. The reasoning is pretty simple: when statisticians dug into the numbers, they discovered that outs are far more valuable, in general, than extra bases. FanGraphs has an excellent run expectancy chart<\/a> that we can use to visualize this. And, sure enough, a runner on second with one out has a lower run expectancy than a runner at first with no outs. A runner at third with one out has a lower run expectancy than a runner at second with no outs. On and on. There are no exceptions to this rule.<\/p>\n<\/div>\n You can also use this run expectancy chart to determine how often you have to successfully steal a base in order to \u201cbreak even\u201d. For example, a runner at first with no outs has a run expectancy of 0.87, a runner at second base with no outs has a run expectancy of 1.12, and no runners on with one out has a run expectancy of 0.25. A successful steal adds .35 expected runs, but a failure subtracts .62 runs. You need to succeed roughly two-thirds of the time in that scenario to gain as many runs as you lose in the attempts. You have to steal at an even higher rate in order to make it actually beneficial to your team. Of course, none of that takes into account the individual prowess of a given hitter. You can fudge the numbers a bit based on how likely the batter is to hit for extra bases or, indeed, hit at all. But it gives us a good starting point.<\/p>\n<\/div>\n However, even those numbers can be somewhat misleading as far as how much they can impact a baseball game.<\/p>\n<\/div>\n My podcast cohost, Jacob Milham, recently shared with me this article from 2021<\/a> that attempts to find what numbers best correlate with winning games. Run differential is the king, and it\u2019s not close, but if you look toward the bottom of that chart, you find the stats that have absolutely no correlation with winning a baseball game. Eighth-worst on that list? Stolen bases.<\/p>\n<\/div>\n That\u2019s right, increasing your stolen bases has almost no impact on how many games you will win.<\/p>\n<\/div>\n We can actually take this even a step further. Look a few steps higher and you see Speed listed. Now this is the FanGraphs Speed stat, which is perhaps a misnomer and indeed appears to have confused the writer of that paper because it doesn\u2019t actually correlate to a team or player\u2019s literal speed, but \u201can average of Stolen Base Percentage, Frequency of Stolen Base Attempts, Percentage of Triples, and Runs Scored Percentage.\u201d (link<\/a>). It also has almost no correlation to winning.<\/p>\n<\/div>\n And we can take this one further<\/em> step. We can use FanGraphs \u2018Baserunning stat. It\u2019s set so that 0 is average and measures how many additional runs a team gets from their baserunning prowess. Ten runs is generally correlated to a team victory, but the difference – entering play Sunday – between the best baserunning team and the worst is only about 24 runs. Meaning the best baserunning team in the sport only wins about 2.5 games more than the worst, with their baserunning. The difference between the best FanGraphs Offense team (which does include BsR) and the worst is about 292 runs, or 29 wins.<\/p>\n<\/div>\n Trying to steal more bases is simply a waste of time. If we go back to that 2021 paper, we see the biggest correlations to winning are almost all pitching stats, which explains how the Royals stuck in this season as well as they did despite their truly horrendous offense. But if you look at the offensive side, it\u2019s all about hard hit%, HRs, and OPS. Interestingly enough, unlike what we have known since Moneyball, OBP would not appear to be the most valuable rate stat; instead, that would be SLG – though ISO is even more valuable.<\/p>\n<\/div>\n That\u2019s right, if you\u2019re looking for an offensive piece to add to your team, you don\u2019t want a guy who gets on base anymore, you want one who hits it over the wall. And that<\/em> explains why Jonathan India has been such a huge disappointment despite getting on base at the fourth-best rate on the team. Without any ISO to his name, he just isn\u2019t actually contributing to wins in a meaningful way.<\/p>\n<\/div>\n