What Happened to All Those Steals of Third Base?

Athletes like Elly De La Cruz can skew our perception of reality. His powerful arm makes most shortstops look like they throw with a wet noodle. His 99th-percentile sprint speed makes most other baserunners look like they’re running on sand. His tall frame, which our website somehow lists at 6-foot-2, makes that guy on Hinge who claims he’s 6-foot-2 look like he’s actually 5-foot-8. Oh, and his 13 steals of third base this year might make you think steals of third are at an all-time high, which couldn’t be further from the truth.

As a fan of highly specific baseball stats – a bold statement to make on this website, I know – I like to check in on the stolen base rates at each bag. Practically speaking, that means I pay particularly close attention to steals of third, the oft-forgotten middle child of stolen bases. Steals of third are too common to receive the same amount of attention as steals of home; at the same time, they’re infrequent enough that they’ll always be overshadowed by the sheer number of second-base steals. Steals of home are almost guaranteed to make tomorrow morning’s highlight reel. Steals of second outnumber all others and thus dictate league-wide stolen base trends every year. Steals of third are stuck in the middle, and that’s especially true this season as their siblings are taking even more of the glory than usual.

The stolen base success rate at home (16-for-29, 55.2%) is the highest it’s been since at least 1969. Indeed, it’s above 50% for only the second time in that span. In addition, runners are on pace to steal home 36 times this year, which would rank second in the divisional era and well within shouting distance of first (38 SBH in 1998). Meanwhile, the overall stolen base rate (i.e. steals per game) is also on the rise, primarily driven by an increase in steals of second. The league is on pace to steal second base 166 more times in 2024 than it did last year, a 5.6% increase, as runners continue to test the limits of the New Rules™.

However, these more conspicuous developments are shrouding what might be the most interesting stolen base trend of the year. Steals of third are way down compared to last season, while the success rate has fallen to pre-rule change levels:

I first picked up on this peculiarity a couple of weeks into the season. Ben Clemens wrote about league-wide stolen base trends on April 11, noting that steals per game had not increased the way many presumed they would. Rather, baserunners were swiping bags a little less frequently than they had the year before. I was surprised to read this, so I went digging into the stolen base rates at each bag. That’s how I noticed that steals of second base (SB2) weren’t down, at least not by a meaningful amount. Stolen bases at third (SB3), on the other hand, were so few and far between that the lack of SB3 was making a dent in the overall numbers.

Through games on April 11, the league was on pace for 39 fewer steals of second (a 1.31% decrease) in 13 more attempts (a 0.3% increase) compared to last year. At third base, however, the league was on pace for 124 fewer steals (a 24.3% decrease) in 153 fewer attempts (a 25.4% decrease). Those were dramatic numbers, large enough to pique my interest even in such a small sample size. Still, I knew I needed to wait longer before drawing any conclusions.

Fast forward a month of baseball – from Spencer Strider’s devastating elbow surgery to Paul Skenes’s thrilling debut – and the overall stolen base rate was looking much healthier. Through games on May 12, the league was on pace to surpass its stolen base total from 2023 by more than 100 steals. Moreover, that all-encompassing number doesn’t do justice to the increase in stolen bases at second. Just past the quarter mark of the season, baserunners were on pace to steal second 225 more times than they had the year before. Yet, as you might have guessed, that meant the numbers at third weren’t much better than they had been a month prior; the league was still on pace for a 25% decrease in SB3. One thing had changed, however. Runners were trying to steal third a little more often; they just weren’t succeeding. Between April 11 and May 12, baserunners attempted to steal third 90 times in 421 games. That’s not so far off the pace from 2023 (605 attempts in 2,430 games). Unfortunately for those runners, they were successful only 66 times (73.3% success rate). That’s a shockingly low conversion rate. The last time the SB3 success rate was so low over a full season was in 2002:

Fast forward again – past the release of bat tracking data and that epic Yankees-Dodgers series – and it’s getting harder and harder to ignore this trend. Between May 12 and today, the SB3 success rate is up slightly, but attempts have gone down again. Thus, the league is still on pace for a 24.1% drop-off compared to last season. Meanwhile, the SB3 success rate on the year is only 77.8%. That’s right around where the success rate hovered before the rule changes; from 2015 to 2022, it never rose above 78.8% or fell below 76.7%. But in 2023, it jumped to 84.3%.

To be clear, I’m not surprised the SB3 success rate is regressing. Evidently, the league determined the ideal SB3 success rate was somewhere around 77%, and there’s no reason the rules changes would significantly affect that. What is confusing is that the SB3 success rate and attempt rate are so much lower than they were last season. In theory, the problem with a high stolen base success rate is that it suggests runners aren’t stealing enough. Yet, in this case, the success rate is down even though runners are attempting to steal third far less frequently:

One possible explanation? Tragedies of varying degrees have befallen all three names that stood atop the SB3 podium last season. Esteury Ruiz has been optioned and injured. Ronald Acuña Jr. is out for the season. Corbin Carroll has spent most of the year in a dreadful slump; he’s currently on pace for 28 steals, partly due to an on-base percentage that’s down by 50-plus points, and even if he catches fire and his OBP surges, it’s unlikely that he’ll have enough base-stealing opportunities to swipe 50 again. Those three combined for 9.2% of all SB3 last season, and their collective 87% success rate boosted the league average. Their absence from the top of the leaderboard this year can’t explain everything, but it’s not ludicrous to wonder if it’s part of the problem. Except for the De La Cruz of it all.

Like I teased at the top of this article, the 22-year-old phenom leads the majors with 13 SB3 and 15 SB3 attempts, good for an 86.7% success rate. That puts him on pace to steal third 29 times this season in 33 attempts. Both of those figures would put last year’s leader, Ruiz, to shame. He only managed 21 SB3 in 24 bids. Furthermore, although De La Cruz is in a class of his own, he’s not doing all the heavy lifting. The top five players in SB3 attempts this season are on pace for more SB3, more SB3 attempts, and a higher success rate than the top five last season. Whatever the explanation for the decline in SB3 might be, it has nothing to do with the guys at the top of the leaderboard:

SB3 numbers at the team level point to a similar conclusion. The top-two teams this year – the Reds (duh) and Nationals – are on pace for more SB3 than last year’s top two. However, the 28 teams ranked no. 3-30 are all on pace for fewer SB3 than their counterparts from last season. Consider that the median team last year stole third base 15.5 times, while this year, that median has fallen to 10.9. Similarly, the median team success rate last season was 87%; right now, it’s 80%. On top of that, nine different teams are on track for at least 12 fewer SB3 this season than last. Conversely, only one team, Washington, is on pace to increase its SB3 total by more than 12. All this is to say that the root cause of the disappearing SB3 isn’t at the top of the leaderboards, it’s everywhere else.

Considering the downturn in SB3 is so widespread, I’m inclined to accept the simplest explanation: Runners are stealing third less often because they’ve had fewer opportunities to do so. For one thing, the percentage of plate appearances taken by left-handed hitters this year is the highest it’s been in the 21st century. That means there have been more plate appearances where the catcher has a clear path to throw down to third. Nearly 35% of plate appearances this season have been taken by lefties, up from 33.7% in 2023. That may not seem like a big difference, but over a full season, plate appearances quickly add up. Indeed, compared to 2023, the league is currently on pace for 549 fewer PA by right-handed hitters in situations with third base empty and a runner on second. That’s not inconsequential.

That said, I think the primary explanation is that defenses have improved when it comes to holding runners on second base. Accordingly, they have decreased the SB3 success rate and, just as critically, prevented plenty of would-be third-base stealers from taking off in the first place. Good defense means fewer stolen base opportunities. Notably, we saw a similar improvement early last season. Runners were successful on 48 of their first 50 SB3 attempts (96%) before pitchers and catchers started to figure out how to stop them. The success rate over the rest of the season was only 83.2%. Eventually, the rate seemed to plateau; this article from The Athletic on August 9, 2023 shows the SB3 success rate just under 85%, and that’s pretty much where it would be when the season wrapped up nearly eight weeks later. Still, it was plain to see that defenses made an impressive adjustment to a difficult situation on the fly.

With an offseason to prepare and a full season of data to work with, it stands to reason that teams could figure out how to better adapt to the new rules going forward. This is a difficult hypothesis to prove, but, for what it’s worth, we know that pitcher pickoffs are up by 23.5% this season, and pitchers have gotten significantly better at holding runners on first base. Unfortunately, Baseball Savant doesn’t have pitcher running game stats for second base yet. However, the Statcast baserunning stats, which take into account “runner position on the base paths,” suggest that advancement opportunities per game are down by about 5% in situations that start with a runner on second base. This metric only considers advancement opportunities on balls in play, but all the same, it’s further evidence that defenses have done a better job holding runners on second this season.

Thus, what we’re seeing right now might be the new normal. Or, at least a lot closer to normal than what we saw last year. Steals per game at third are still up by 31.6% compared to 2022, and the success rate is right around the accepted break-even point. Perhaps the third base gold rush of 2023 has settled down, but the new rules – and Elly De La Cruz – are continuing to do their job to increase activity on the basepaths between second and third.

All stats through June 19.