The Cardinals hit not one, but two extra-inning home runs Tuesday night.
The blasts, a two-run shot by Paul Goldschmidt that plated the Magic Runner in the top of the 11th inning and a three-run homer by Tyler O’Neill four batters later, helped the Cardinals beat the Brewers 6-1. It was a pretty rare display of power for an extra-inning contest since MLB introduced the new extra-innings “start with a runner on second base” rule prior to the start of the 2020 season.
Extra-inning home runs are down. Way down. Let’s start with the very basic numbers, and then we’ll dive a little deeper.
In 2019, there were 129 home runs hit after the ninth inning in 208 extra-inning games.
In 2020-21, there have been only 35 homers hit after the ninth in 123 extra-inning games.
The most obvious difference, of course, is the length of games. Most extra-inning games under the new rules end in the 10th or 11th (for the purposes of this story, we’re only looking at games originally scheduled for nine innings, not the eighth/ninth frames of a seven-inning double-header contest that goes extra). They average 11.3 extra-inning plate appearance per game, whereas in 2019 the average was 18.6 extra-inning PAs per game, which often made it to 12 innings and beyond. Fewer PAs, fewer home runs. Makes sense.
But the dip in home runs isn’t just about a lack of opportunities at the plate. Let’s look at the plate appearance per home run numbers for 2019 compared with 2020-21.
27.5 PA/HR for innings 1-9 (6,647 HR/182,656 PA)
29.9 PA/HR for innings 10 and beyond (129 HR/3,861 PA)
30.2 PA/HR for innings 1-9 (3,492 HR/105,318 PA)
39.6 PA/HR for innings 10 and beyond (35 HR/1,387 PA)
Pretty significant differences there, though it’s not a completely stunning revelation. It makes sense that, with a runner already in scoring position in 2020-21 extra-innings games, batters might not be swinging for the fences quite as much, if the primary goal is to push home that runner on second base. It makes sense, at least in theory, that hitters are adjusting their approaches based on the situations.
But on the other hand, the impact and effectiveness of defensive shifts have shown that a large majority of batters either are reluctant to — or cannot — change their approach at the plate. And even though trying to hit the ball the opposite way and trying not to hit home runs isn’t exactly the same thing, it still falls under the “can I change my approach?” category.
And it’s worth mentioning the baseball itself. It’s lighter for 2021, and offense has been down this season, across the board. Fewer home runs and fewer runs scored. It’s created a bit of a difference between 2020 and 2021.
28.5 PA/HR for innings 1-9 (2,284 homers/65,719 PA)
39.4 PA/HR for 10th and beyond (20 homers/787 PA)
32.8 PA/HR for innings 1-9 (1,208 homers/39,599 PAs)
40.0 PA/HR for 10th and beyond (15 homers/600 PAs)
There’s still a big gap in both years between regulation innings and extra innings, thanks to the rule.
I wanted to see what people in the game thought. I presented Angels manager Joe Maddon with the PA/HR numbers for extra innings, mentioned the difference from regulation innings and asked whether his hitters were intentionally taking a different approach. I figured that, if nothing else, Maddon would have a theory I could explore. I was right.
“That’s interesting,” Maddon said. “I would almost bet that, from the home side, if the score is tied, we’ve just got to score a run. That’s definitely going to put a different mindset in. The guy’s already in scoring position, so you’re not just trying to end the game with one swing. Automatically, your mindset changes. The difference is, on the visitor’s side, you do want to put up multiple runs. Of course you want to score one, but you’re trying to impact more than just one so you can cover (the home team) scoring the guy from second base.
“Maybe that has something to do with it.”
That theory definitely makes sense. If the home team knows it only needs one run in a tie game to win, why swing for the fences? Any sort of contact — a slow chopper, a fly ball to right field, a soft-contact bloop single or a gifted error — stands a good chance of getting the Magic Runner to third base, and more contact stands a good chance of getting him home with one out. Heck, the A’s completed a comeback win in the bottom of the 10th against the Twins on April 21 without even a single hit, scoring three times thanks to two walks and back-to-back errors on routine grounders.
“You know what? You put a ball in play,” A’s manager Bob Melvin said after the game. “… Put it in play, and something good can happen. And that’s what did.”
So let’s look. The numbers for 2021 support Maddon’s bet. Of the 15 extra-inning home runs hit so far this year, 13 were hit by the visiting team — 10 in the 10th, two in the 11th (those two Cardinals) and one in the 12th. Both of the home-team homers — Jordan Luplow and Kyle Schwarber — were hit with the game tied, not with the home team trailing.
“The mindset of the hitter,” Maddon said, “if the guy’s already at second base, is going to be different, I believe.”
In 2020, though? Different story
There were, as noted above, 20 homers hit in the 10th inning and beyond in 2020. Of those 20, 11 were hit by the the home team, just nine by the visitors. Yep, you read that correctly. Ten of those 11 hit by the home team were walk-off homers; four of them — by Freddie Freeman, Randal Grichuk, Manny Machado and Jose Ramirez — came with the home team trailing by one run heading into the at-bat. One, by Christian Walker, came with his team trailing by four.
Seven of the nine hit by the visiting team came with the game tied. Two came with a run already across in the 10th inning.
One rule, two seasons worth of sample sizes (both small, of course), two completely different results.
“MLB Tonight” analyst Billy Ripken, a former infielder who logged 12 seasons in the bigs, spends pretty much every day and night watching baseball, so I asked him, too. He’d clearly spent some time thinking about the subject ahead of our conversation and came prepared with questions of his own, a nice flipping of roles.
“See where punch-outs go in those extra innings, the strikeouts,” he said. “I’d love to think, in my old-school brain that still lives inside my head, that these guys are making a conscious effort to try and move a guy up 90 feet to get him closer, even though they might not be bunting. But the idea of putting the ball in play and watching something happen is positive. Nothing positive happens when the guy walks back to the dugout after he swings and misses three times.”
It’s a great thought, and would certainly speak to the theory that hitters really are making adjustments, taking a different approach dictated by a non-shift scenario.
So let’s look. Here are the strikeout percentages for regulation innings vs. extra innings, for 2019 vs. 2020-21 (again, same caveat about extra innings only including 10 and beyond).
22.9 percent in innings 1-9 (182,656 PA/41,849 K)
25.2 percent in 10th and beyond (3,861 PA/974 K)
23.7 percent in innings 1-9 (105,318 PA/24,973 K)
23.2 percent in 10th and beyond (1,387 PA/322 K)
Well, look at that. Ripken’s hunch was right. Even though strikeouts in regulation innings are up in 2020-21 compared with 2019 (and all time), strikeouts in extra innings are down, by two full percentage points. That’s a pretty significant number.
I texted him numbers and told him that his theory was correct.
“Check,” he replied. “That is still ridiculously high — in general, I’m not a fan.”
The “not a fan” comment was his phone-conversation observations about the general approach of modern hitters, and the managers, coaches and front office types who encourage the approach. For context, the league strikeout percentage during Ripken’s best year — his 4.1 bWAR in 1990 — was 14.9 percent.
“Guys today are truly trying to get off their ‘A’ swing three times because they think they can do damage,” he said.
But apparently that’s not happening as much in extra innings, with that runner magically appearing on second base. More contact is being made. I think it’s key here to look at how the extra innings compare with other inning segments, not just 1 through 9 as a whole. So let’s start with the strikeouts.
23.8 percent in innings 1-3 (36,600 PA/8,705 K)
22.5 percent in innings 4-6 (37,214 PA/8,382 K)
25.0 percent in innings 7-9 (31,504 PA/7,886 K)
23.2 percent in 10th and beyond (1,387 PA/322 K)
These numbers make sense. The strikeout rates are highest in the first three innings, when starting pitchers are freshest, and in the last three innings of regulation, when the best relievers generally throw. In the extra innings — especially for the home teams — relievers typically aren’t saved for Magic Runner innings.
Let’s use that Cardinals-Brewers game as an example. The Brewers had a 1-0 lead into the eighth inning. Here are the 2020-21 numbers (including Tuesday’s game) for the relievers Milwaukee used in the final four innings of the game.
8th inning: Devin Williams, 1.55 ERA, 15.9 K/9
9th: Josh Hader, 2.48 ERA, 15.2 K/9
10th: J.P. Feyereisen, 2.30 ERA, 8.6 K/9
11th: Brad Boxberger, 3.48 ERA, 9.6 K/9
Feyereisen and Boxberger are good pitchers, no doubt. But there’s a reason why strikeout pitchers Williams and Hader got the most important innings, the eighth and ninth (the Cardinals scratched across a run on Williams on a single, a walk and back-to-back fly balls that advanced runners). That happens all over the sport, which is why strikeout numbers are higher in the seventh/eighth/ninth than any other multiple-inning split.
But if that’s the case, and the best relievers generally aren’t being saved for the 10th and beyond, shouldn’t home run numbers be higher after the ninth? Let’s look at the three-inning splits for home runs.
27.9 PA/HR for innings 1-3 (1,313 homers/36,600 PA)
29.7 PA/HR for innings 1-3 (1,252 homers/37,214 PA)
34.0 PA/HR for innings 1-3 (926 homers/31,504 PA)
39.6 PA/HR for 10th and beyond (35 homers/1,387 PA)
Look how that chart climbs in a linear fashion, ending with 11.7 more plate appearances between home runs in innings 1-3 than Innings 10 and beyond. Huge, huge difference.
Now, let’s compare that to the last year with traditional extra-inning rules.
27.0 PA/HR for innings 1-3 (2,309 homers/62,375 PA)
26.5 PA/HR for innings 1-3 (2,362 homers/62,598 PA)
29.2 PA/HR for innings 1-3 (1,976 homers/57,683 PA)
29.9 PA/HR for 10th and beyond (129 homers/3,861 PA)
In both years, we see a drop in at-bats per home run when the elite relievers typically enter the game. In 2020-21, though, the “runner already on second” scenario makes an even bigger impact.
Maybe another of Ripken’s hunches was correct, too. Maybe the Magic Runner rule is encouraging players to take more of an old-school approach at the plate.
“I’m wrestling with the whole ‘man starting on second thing,’” Ripken said. “I didn’t hate it last year, because I understood why and I got it. It’s a 60-game campaign, for all the reasons we all know. But I’m not so sure it’s necessary in a full season.
“But there seems to be a conscious effort throughout baseball to maybe shorten up a bit and not swing away, and maybe shorten up enough to not strike out. That seems to be a good thing to me.”
One more set of numbers: If strikeouts are down, shouldn’t other numbers go up? Let’s look at extra-inning slash lines and OPS for each of the past three seasons.
2019: .233/.331/.394, .725 OPS
2020: .237/.351/.382, .733 OPS
2021: .261/.378/.414, .792 OPS
Small sample size in 2021 (600 PAs), of course, but that’s a healthy jump across the board.
At this point, “What does this all mean?” is a fair question. The numbers are pretty clear about what’s happened so far, concerning home runs: The new rule has dramatically cut down the number of homers after the ninth inning. Lots of walk-off hits, of course, but so far in 2021 there have only been two walk-off homers in 55 games that extended past the ninth inning.
So is this our new normal if the Magic Runner rule becomes a permanent part of baseball’s equation? It’s worth noting that it’s still a relatively small sample size, just 123 games. That’s only about 75 percent of a full season’s worth of data, which isn’t much for a sport that sees 4,860 games in an average regular season.
Guess we’ll have to wait and see.