Why did those World Series games last so long?

Baseball's regular season has the pace-of-play problem with which we're all familiar: Many of the 2,430 games take on dead time until they are boring, lulling. It's a good dish, watered down until you don't love it as much.

The postseason's pace-of-play problem is different. It's not that games get boring, since a World Series game is almost always tense and urgent, and if you're on a cross-country flight with cable TV access you will enjoy every moment of it. But most people aren't on cross-country flights. Most people have full lives, and they have to squeeze in their baseball indulgences among other obligations, like family, and sleep, and moving at least once every four hours to avoid nerve damage. These postseason games are thrilling, but they are so lengthy that they become impractical for many otherwise enthusiastic customers -- a good dish that goes cold before it can be finished.

Nearly every game in this World Series was long, even by World Series standards. Of the 13 longest nine-inning World Series games this decade, six came this year. Game 3, a 4-1 Houston victory, took 4 hours, 3 minutes:

In some ways, the most discouraging part of the pace of these games is how well disguised the slowness is. It's not that the games are slow for reasons that are anomalous (like 15-14 slugfests) or that could be easily legislated away (like limiting constant mound visits by catchers, which have been sharply curtailed since the 2017 postseason) or that would be delightful (very good dogs running onto the field to frolic). Rather, they're slow because ... well, why are they slow?

We rewatched Game 3 of the Series, that unprecedented four-hour affair, with a stopwatch in hand and a few spreadsheets open, looking for the stuffing. Here's what we found.

Let's start with the baseline: While the average nine-inning baseball game this year was 3 hours, 5 minutes (an all-time high), the 65 4-1 games -- with less offense than typical games -- averaged just 2 hours, 55 minutes. That's our baseline. (The longest 4-1 game in the regular season, incidentally, was 3 hours, 29 minutes. There were two previous 4-1 games in this postseason that were at least 15 minutes longer than that. Postseason games are very long.) So Game 3 of the World Series was, at 4 hours, 3 minutes, 68 minutes longer than we would expect from the score and the year. What are those 68 minutes showing us?

1. For around 17 of them, they are showing us this:

Postseason commercial breaks are longer than regular-season breaks, by about 50 seconds. There are 17 between-inning breaks (one after every half inning except the final one), plus pitching change breaks (which are also longer in the postseason), so that adds around 17 minutes. This is baked into every postseason game, so if you want to know what an average postseason game is, start at 3 hours, 20 minutes.

2. For a little more than three minutes, they are showing us this:

There are, on average, 2.05 mid-inning pitching changes per game in the regular season, each of which adds about 3 minutes,15 seconds of dead time. (Pitching changes at the start of an inning add only a few seconds of extra time.) There were three such pitching changes in this game, which is in keeping with the World Series norm: There have been an average of 3.7 mid-inning pitching changes per World Series game since 2016. Managers have more urgent hooks for their pitchers (both starters and relievers) in the postseason, and a lot more innings generally go to relievers in the postseason, hence more mid-inning pitching changes. The three longest half-innings in Game 3 -- all of them lasting at least 22 minutes, compared to about 10 minutes for a typical half-inning -- all had mid-inning pitching changes.

3. For about five minutes, they are showing us this:

That's a single that the Nationals hit in the bottom of the ninth, which, as you know, wouldn't have been possible if the Nationals -- the home team -- had been winning the game. But the road team won all seven games in this World Series, which added about 10 minutes to each game -- or, more fairly, about five minutes, since the home team does bat in around half of baseball games.

We're up to about 25 minutes of our 68.

4. For about 16 minutes, they are showing us this:

That's Yuli Gurriel singling with a runner on first base and two outs. We noted that scoring was below average this World Series, but it was uncharacteristically heavy on baserunners. The Astros and Nationals combined for a .331 on-base percentage, the highest in a World Series since 2011, and significantly better than the .299 OBP that World Series hitters had this decade. Batters in this year's regular season had a .323 on-base percentage. It doesn't feel as if we saw an explosive, offense-packed series -- there were only nine runs scored per game, down from 9.6 per game in the regular season -- but that's just because nobody is getting hits with runners in scoring position. There have been a ton of baserunners, but they're not being driven in.

Put this together with the third factor -- more home teams batting in the bottom of the ninth -- and this World Series had about six more batters per nine innings than the average World Series this decade. And despite the low score, the Astros' 4-1 victory in Game 3 had 81 plate appearances -- almost 12 more than the average 4-1 game this year.

We already counted the time spent by batters hitting in the bottom of the ninth, so we won't count that again. But at a little under two minutes per batter, on average, the other eight extra batters account for about 16 minutes of extra baseball.

5. For about 20 seconds, they are showing us this:

That's just a pitch, one single extra 3-2 pitch. There tend to be slightly more pitches per plate appearance in the World Series than in the regular season, as strikeout pitchers face disciplined hitters and produce deeper counts. The difference was minimal for this game, though: In Game 3, there were 3.94 pitches per plate appearance, compared to 3.93 in the regular season, which is one extra pitch. So that's those 20 seconds.

6. For about two and a half minutes, they are showing us this:

We mentioned that more baserunners have meant more plate appearances, each one adding about two extra minutes. But more baserunners also means more men on base, and pitchers tend to slow way down when men are on base -- not just to try to stall the running game, but to go through more complex signals with the catcher, to pace themselves through longer innings and to keep themselves composed. The average time between pitches when a man is on base is about 5.5 seconds longer than when the bases are empty, according to Baseball Prospectus' pace metric.

Because there were a lot of baserunners in this Series, and this game -- and, especially, a lot of baserunners early in innings -- there were a lot more pitches thrown with runners on base than normal: 53% of all pitches in Game 3, compared to about 42% in the regular season. That adds up to 17 extra pitches thrown from the stretch, which -- at 5.5 extra seconds per -- adds about two and a half extra minutes. That gets us to about 44 extra minutes accounted for, and about 24 minutes unaccounted for. So where are those?

7. For at least 13 minutes, and probably a lot more, they are showing us this:

Exciting! Dramatic! It's pitchers taking longer than usual to pitch!

Nearly every pitcher in this World Series has worked more slowly than he did during the regular season. On average, World Series pitchers have taken an extra 1.5 seconds between pitches with the bases empty and an extra three seconds per pitch with men on base. Sometimes this is obvious to the viewer: WhenZack Greinkefell behind 3-0 to leadoff hitter Trea Turner, he took slow walks around the mound after the second and third balls, settling himself. When Anibal Sanchez was facing George Springer in a long at-bat during the second inning, he continually shook off signs and stepped off the mound. But mostly this shows up in an extra beat or two that you wouldn't notice unless you were timing it, at least until you wake up groggy from sleep deprivation the next morning.

Greinke took about six more seconds between pitches with men on base in this game, compared to the regular season, and about three extra seconds per pitch with the bases empty. Sanchez took an extra four seconds with men on, and an extra second with the bases empty. Relievers Brad Peacock, Will Harris, Joe Smith, Josh James and Joe Rosswere all slower in this game than they typically are.

All those extra seconds add up to 13 minutes, compared to the regular season, but even that total probably undersells it. The "pace" figures we're using, from Baseball Prospectus, exclude the time between any pitches that come more than one minute after the previous pitch, because it assumes that something unusual happened -- an injury, a visit from the pitching coach, a pickoff attempt. We weren't able to determine whether there are usually more pickoff attempts in the World Series, but we can say definitively that there were more pickoff attempts than usual in this specific game, Game 3. On average, teams threw over about 3.5 times per game in the regular season. Sanchez threw over five times just in the first inning, and continued to do so as the game went on.

So that undefined time -- pickoff time, mound-visit time -- could get us pretty close to 68 minutes. The rest can be blamed on Turner, who spent two and a half minutes in agony after fouling a pitch into his crotch.

From MLB's perspective, we can break this time into three segments:

1. Commercials (17 extra minutes)

2. The type of game that happened to be played (around 27 extra minutes)

3. Postseason-specific dawdling (around 24 extra minutes)

The first is easy to fix -- but they presumably never will. The second is impossible to fix -- but nobody would ever want to, since that was all the natural and exciting stuff. The third is what executives dream of erasing.

But it's not simple. Pitchers aren't taking more time between pitches because they've never considered moving faster. They're doing it because these games are incredibly important, and those extra seconds help them, and research has found that working slower helps pitchers throw faster. As Joe Garagiola Jr. told Jerry Crasnickin 2013, "If people are taking a little more time between pitches ... it's because somebody is going to be the world champion at the end of the next eight or nine days. You can't lose sight of that."

The fact that pitchers slow down so much in the postseason is evidence of how important to their performance pitchers consider those extra seconds. Taking those extra seconds away in the name of business would be like outlawing sliders in the name of business. It would rightly infuriate pitchers, prioritizing a secondary aspect of the sport (the surface-level entertainment value) over the primary aspect of it (competition between elite athletes attempting to be their best). It's no wonder the union has pushed back against pitch-clock proposals that would disproportionately penalize one class of employee for the league's relatively unimportant business interest.

If the league really wanted to cut the time pitchers spend between pitches in the most important innings of the year, they would do better to consider pitch-clock rules as explicit efforts to hamper pitcher performance, not to save time but to strengthen the actual competition on the field. Those sorts of changes are common in setting (or changing) the rules of game play: imposing play clocks and shot clocks in other sports, banning spitballs and regulating mound height in baseball.

The main thing keeping offense going in baseball right now is the juiced ball; otherwise, pitchers are and have been ascendant for the past decade, with strikeout rates setting records every season. If the dead ball goes away -- and since Major League Baseball claims not to know what caused the dead ball, it also doesn't know how to preserve it -- baseball could quickly enter a new dead ball era.

One small thing the league could do to return an advantage to batters is take away pitchers' ability to slow the game down to their own preferred rhythm. Not for pace-of-play reasons, mind you, but to maintain a nice balance between pitching and offense. If, as a convenient side benefit, World Series watchers get a good night's sleep ... well, that'd be OK too.

Thanks to Lucas Apostoleris and Baseball Prospectus for research assistance.