Which of baseball's most unbreakable records might actually get broken in 2019?

74 home runs? 384 strikeouts? A 57-game hitting streak? History could be made this year, but some standards are more attainable than others. ESPN Illustration

In the years since Barry Bonds retired, records have been mostly replaced in baseball by fun facts. A record is something you chase. People know what it is in advance. A fun fact can be just as impressive, but it is bespoke and discovered only after it happens. Mike Trout has the most WAR in history through age 26. Jamie Moyer had more wins after his 31st birthday than Pedro Martinez had in his career. Mark Grace once hit .825 when he went the opposite way, the highest ever. Fun facts, but not records.

There will be some fantastic fun facts this baseball season, a few dozen from Trout alone, but try to find a real record that might be broken. Records, since Bonds retired, hardly ever get broken. Other than Aaron Judge breaking the rookie home run record in 2017, and the Yankees breaking the all-time team home run record last year -- both fairly fringy as records go -- can you think of any record in the 2010s that (A) you were aware of before 2010 and (B) has been broken since? Not really!

The main reason is records get set in extreme environments, and baseball mostly isn't as extreme as it was in the 1960s (when pitching was out of hand) or the 1930s (when offense was) or the 1880s (when there were only eight pitchers and they each threw 600 innings). Instead of records, we have to settle for seeing the greatest baseball players in the history of the world, putting up non-record statistics. Alas.

But what if a fan from the future came back in time and told us a record was set this year? What if she said it was one of the big ones, even? Let's say, one of the 20 most significant records that can be set entirely within the course of a single season? Which record do you figure it would be?

What follow are The Big 20, the 20 records I estimate are the biggest deal, ordered by how likely they are to be broken this year. We're ignoring the purely anachronistic records, the ones that are of such an ancient style of play there is no opportunity for a player to break them. We're treating only batting records set since 1901, and only pitching records set since 1920, as official records. And we're not debating the composition of The Big 20 itself, because it exists only for the framing of this article. What is up for debate is the question at hand. We're starting with the least likely.

20. 36 triples (Owen Wilson, 1912)

Buddy, get out of here already. There's no way anybody is hitting 37 triples.

This one is right on the cusp of being disqualified from The Big 20 as an anachronism. Wilson, the record-holder, hit 36 triples in a ballpark that was 460 feet to dead center (his teammate Honus Wagner hit 20 that year -- as a 38-year-old), in an era when triples were thrice as common as they are today. (Shoeless Joe Jackson set the AL record the same year.) Put it this way: Nobody has hit 24 triples in a season since 1925; Wilson hit 24 just at home, in Forbes Field.

But the thing is, even in the year the record happened, 36 triples was a stupid outlier of a record: Nobody before or after Wilson ever hit more than 26, and even in the minors (with weirder ballparks, clunkier defenders and, in some cases, much longer seasons) nobody has ever hit more than 31. Even Wilson never hit more than 14 in a season otherwise. He never led the league, other than that once. If a freak-show season like Wilson's could happen once, then we allow it could happen again, perhaps in a triples-rich ballpark like Coors Field, Chase Field or Comerica Park.

It just won't. If you add up the most triples anybody in the past decade has hit in the first half of a season, and the most triples anybody has hit in the second half of a season, you only get 27, still nine short of the record. Nine short of the record. Nine! Nine triples led the American League in 2016.

Owen Wilson really ruined this for us, to be honest. The intrinsic interestingness of a 27-triple season is hardly different than the intrinsic interestingness of a 37-triple season, except for the fact that the record is 36. This is both the bounty of a record -- it gives us a mile marker that tells us to pay attention, something important has happened -- but also its curse, because it diminishes everything short of it. We will never see a player hit 36 triples. We'll never see it even challenged. The rest of all of our lives are hopeless on the triples-record front. No point going on. If Wilson had just stopped at 27 we'd have something to watch for, but he hit 36 triples in one season, and that settles that question.

(Here's a fun fact: Wilson's final triple came on his last at-bat of the season, when he was out at home trying to score on an inside-the-park homer.)

Who'll do it and my totally made-up chance he does it: Trea Turner, 1 in 2.9 million.

19. A .426 batting average (Nap Lajoie, 1901)

Pal, be real. Last spring we went over the near-impossibility of hitting .400, and in a nutshell: If you combined the league's lowest strikeout rate (Andrelton Simmons) last year with the league's highest home run rate (Khris Davis) and the league's best BABIP (J.D. Martinez) into one perfectly optimized batting average superhero, even that guy would have hit a mere .399. Nobody's going to hit .400.

But OK, maybe. Could happen. People hit over .400 for a month all the time, so it's just a matter of bunching six of those months in the right sequence. (Tony Gwynn hit over .400 across a calendar year.) But imagine .426. If you'd given Ted Williams an extra 100 at-bats to get his .406 average to .426, he'd have had to hit .520 in those 100. Mookie Betts, who led the majors last year, would have to get hits in 73 straight at-bats to get his .346 average up to .426. The smartest thing baseball fans ever did was totally forget what the single-season batting average record is, and just focus on the number .400. We should do more of that, and this article isn't helping.

Who'll do it and my totally made-up chance he does it: Mookie Betts, 1 in 1.15 million.

18. A 56-game hitting streak (Joe DiMaggio, 1941)

Come on, guy. There is no active player who has ever, at any two points in his career, put together two hitting streaks that sum to 57 games. Freddie Freeman has two that add up to 50, Ryan Zimmerman has two that add to 49, and Brian Dozier has reached 48. That's how hard it is to imagine this. It is almost impossible to imagine any current player reaching 57 in a row. Only one player since Joe DiMaggio has come within even two weeks of the record.

It is significantly easier to just imagine a whole new ballplayer. It would have to be somebody who rarely walks (so that he doesn't see a streak thwarted by a three-BB game) and who rarely strikes out, and who is nuttily gifted in a way unlike any ballplayer we've seen in the past seven decades. As FanGraphs' Jeff Sullivan wrote this winter, it would have to be somebody like ...

Willians Astudillo!

Or, maybe more plausibly, another name mentioned in that article: Vlad Guerrero Jr., who last year hit .402 in Double-A as a 19-year-old and who, apparently, projects this year as a 20-year-old to have the game's fourth-highest rate of hits per plate appearance. Crucially: The odds are way against Guerrero turning into such an unprecedented player. But the odds are probably better than they are for any current major leaguer to simply small-sample-size his way through 57 landmines.

Who'll do it and my totally made-up chance he does it: Vlad Guerrero Jr., 1 in 62,000.

17. A .609 OBP (Barry Bonds, 2004)

The good news for an OBP challenger is that more than almost any of these records, it's a question of human agency as much as it is a question of human ability. Nobody is going to hit 74 home runs because it's nearly impossible to hit 74 home runs, no matter how much you want it. Everybody wants it. They just can't!

It's also just about impossible to have a .610 OBP, but to the degree it can happen it will happen because opposing pitchers simply choose to make it happen. Barry Bonds had a .609 OBP because 120 times he got an intentional walk. No effort required. No possibility of failure. Free OBP.

He also hit .362. He also drew 112 unintentional walks. And the chances anybody would be so good as to cause mass hysteria, as Bonds did, are extremely low. But that's the thing about mass hysterias: They rarely make sense, even after the fact, and predicting one in advance is impossible.

Who'll do it and my totally made-up chance he does it: Mike Trout, 1 in 61,800.

16. 13 shutouts (Bob Gibson, 1968)

What are we even talking about? There's no way this can happen. Felix Hernandez has 11 shutouts in his career. Neither league had 13 shutouts last year. The leader last year in each league had one. This is the stupidest suggestion yet. Thirteen shutouts clearly belongs in the anachronism pile.

But it's here. It's here because, while the modern game certainly discourages complete games, it doesn't make them impossible. Pitchers do complete games. Cliff Lee had six shutouts in a season this very decade, and while it's nearly impossible to pile seven more on top of that, he also had:

  • A complete game with one run allowed

  • An 8 2/3-inning start with no runs allowed

  • Two eight-inning starts with no runs allowed

  • Two eight-inning starts with one run allowed

  • Two seven-inning starts with no runs or one run allowed

He did all that without topping 126 pitches, so he's squarely in the modern era of pitcher usage. Six isn't that close to 14, but if you could flip, say, 25 outcomes to his favor I bet you could turn those eight "close" games into complete game shutouts.

Of course, this will never happen. They don't even let pitchers complete their no-hitters anymore. Jacob deGrom just had the second-lowest ERA this century and (probably not coincidentally!) they only let him complete one game.

Who'll do it and my totally made-up chance he does it: Trevor Bauer (who has no career shutouts), 1 in 51,799.

15. 262 hits (Ichiro Suzuki, 2004)

Of all the hitting records on this list, this one and the OBP record were set most recently, in 2004, but nope, not happening, and we should quickly get past it. No other active player has ever had more than 225 hits in a season, and few elite hitters are as walk-averse as Ichiro was permitted to be. Last year, the major league leader in at-bats (Trea Turner) only got 664; a batter would have to hit .396 in those at-bats to reach 263. The outer limit for at-bats -- say, a leadoff hitter for a high-scoring team who plays every game and has a 10th-percentile walk rate -- is perhaps 725, nine higher than the all-time record. And even the 725-AB guy would need to hit .363, a mark not met by any hitter in this decade.

Who'll do it and my totally made-up chance he does it: Vlad Guerrero Jr., 1 in 36,500.

14. 73 home runs (Barry Bonds, 2001)

Friends, I just do not know. There is no way a player could hit 74 home runs, right? But consider: Barry Bonds hit 73 in a season in which fewer home runs were hit than we had last year -- and a lot fewer than were hit in 2017 -- and in a pitcher's ballpark, and while getting walked 177 times, and while missing nine games entirely and batting third and fourth in the lineup.

Bonds had 664 plate appearances, more than 100 fewer than the Red Sox leadoff batters got last year. I know, I know, steroids, but is it that hard to imagine a superstar who is almost as good at hitting as Bonds was, playing in a more favorable home run environment, with more lineup protection, getting 100 more plate appearances and maybe 175 more at-bats than he did, hitting home runs with 80 percent the frequency Bonds did?

It is, right? It is that hard to imagine. Nobody in the past decade has hit as many home runs in the first half as Bonds did, or as many in the second half as Bonds did.

Who'll do it and my totally made-up chance he does it: Aaron Judge, 1 in 24,000.

13. 130 stolen bases (Rickey Henderson, 1982)

No, seriously, what are we talking about here? Nobody in the 2000s is even in the all-time top 25. Nobody in the past decade has more than 70 in a season. The game has changed: We value outs more than extra bases now. Whereas Rickey Henderson could attempt 172 stolen bases in 1982, nobody in 2018 attempted even one-third that many. Henderson was caught 42 times in 1982. No AL baserunner attempted more than 42 steals in 2017.

It's stupid that this is even as high as 13th. Except the one we had 14th was stupid, too, and also because of this: Billy Hamilton stole 155 bases in the minors in 2012, which set an all-time record. A big reason he set the all-time record is he went after the all-time record, and steals is one of those records you can go after. A big consideration for any record pursuit is: Does the pursuit itself increase the likelihood of a record being set (as in Cal Ripken's iron man streak), does it decrease the likelihood of a record being set (as in, perhaps, a home run chase, in which the closer you get the less likely you'll be to get pitched to), or is it irrelevant (e.g., ERA record). Sometimes, you could argue either way: The stress and strain of carrying a hitting streak 56 days would surely wear down a hitter, but it would also get him pitched to (probably) and keep his manager from pulling him in the sixth inning of a game. But there's no question about stolen bases: A runner gets to decide how often he goes and what his risk threshold is. Somewhere along the line, Hamilton decided he was going to break a record, and to some degree he could make that happen: He attempted 192 stolen bases, 70 more than he did the year before and 90 more than he did the year after.

A runner who tried to break Henderson's record in the majors would be facing a lot of resistance. There'd be a lot of complaining he was chasing stats. There'd be steals of third with two outs up by five runs in the eighth inning and nobody would appreciate that. He'd probably have to be more successful than Henderson was, get thrown out less, to convince his team to go along with this. And it's obviously not easy to steal bases against modern batteries: Hamilton, who was supposed to break baseball, has never stolen even 60 in the majors, and his success rates aren't even all that impressive. A runner would have to average 22 steals per month, a mark only one baserunner has achieved in one month over the past decade -- Trea Turner, 22 on the nose. There's also this: Henderson led the league in walks and had a .400 OBP. He reached first base -- not counting fielder's choices, but including singles, walks, HBPs and errors -- 233 times, more than any player in baseball last year, more than 70 more times than Hamilton's best season. Unless a team does the stupidest thing you've ever heard in your life -- like committing to using a speedster as a pinch runner early in every game, and telling him to steal every time -- this isn't going to happen!

It's more likely than 74 home runs, though. Human ambition is a stronger force than human ability. Furthermore, eventually 90 feet really isn't going to be enough, though it appears to be for the current baseball players we're all aware of.

Who'll do it and my totally made-up chance he does it: Trea Turner, 1 in 18,500.

12. A 26-game winning streak (1916 Giants)

No, no, NO, but ... maybe. We're into maybeland.

This one really is simple enough math: The best team in baseball this year will probably win about 100 games, and if we're generous and say it's 105, then they'll win 65 percent of their games. If we're really generous, because we know every record begins with a generous premise (career years, everything breaking right) let's say 110. That team has a 1 in 35,000 chance of snapping off 27 wins in a row, if we treat each game as equally winnable, with 136 27-game sequences to try it in: Roughly 1 in 250. If there was a 110-win team, we'd expect that team to win 27 in a row every two-and-a-half centuries.

Is every game equally winnable? No. A team might have a particularly soft schedule, improving the likelihood of good events clustering. A winning streak like this might start with a nine-game home stand, then divert to a road trip through Baltimore and Toronto, then back at home -- all of the games more likely than not to be won. (Cleveland, in winning 22 straight in 2017, had only four games against winning teams.) Success on Day X could bring Day X+1 advantages, if strong starting pitcher performances keep the bullpen fresh. And that's all without the mysterious phenomenon of momentum: Would each win make a team confident, or raise the pressure and give their opponents more grit in return? Who knows!

Who'll do it and my totally made-up chance they do it: Yankees, 1 in 2,300.

11. 1.12 ERA (Bob Gibson, 1968)

I have no idea how this got as high as No. 11. Gibson set this one in 1968, when they might as well have been playing in the dark for hitters' chances. Since that year -- after which baseball lowered the mound -- nobody has had an ERA lower than 1.53 (Dwight Gooden, 1985), which sounds pretty close to 1.12 until you think about it like this: Gooden would have had to throw 11 complete-game shutouts in a row to lower that 1.53 to 1.13. Jacob deGrom had a 1.70 ERA last year, the third-lowest since Gibson; if you turned his four worst outings into scoreless starts, it would only get him to 1.16.

Or think of it like this: Eight innings, one run = 1.125 ERA. Not good enough.

This is one of the few records, though, modern baseball is moving in favor of, instead of away from. Starting pitchers face fewer batters per game, which helps their ERAs. And starting pitchers throw fewer innings per year, increasing the likelihood of an outlier performance. (Smaller sample sizes help in any outlier pursuit.) We know that, over the past 50 years, the floor of a starting pitcher's ERA is around 1.50, 1.60. But we also know that the floor for a relief pitcher is much lower -- a handful of relievers have carried ERAs below one for a full year, up to 70 or 80 innings. So if deGrom can have a 1.70 ERA in 210 innings, and Blake Treinen can have a 0.78 ERA in 80 innings, do these eventually meet in the middle as pitching roles merge? Can't you kind of imagine an ace pitcher who throws 165 innings, five innings at a time? Blake Snell had a 1.33 ERA in his final 124 innings last year. That is higher than 1.12, but definitely lower than 1.53.

Who'll do it and my totally made-up chance he does it: Chris Sale, 1 in 1,750.

10. 191 RBIs (Hack Wilson, 1930)

This one started at No. 18 on the list, but it kept moving higher. Now, here's the case for it being No. 18: The top nine RBI seasons of all time came in a 12-year period during the 1920s and 1930s, and nobody has come within 25 RBIs of Hack Wilson's record since 1938. No active player has ever driven in 140. That's ... not that much closer to the record than Curtis Granderson's 23 triples in 2007 were to Owen Wilson's 36.

But here's the case for it at No. 10: Baseball hasn't changed that much since the 1920s and 1930s, when all these wacky RBI marks were set. Yes, 1930 -- Wilson's year -- was the most scoring-friendly season in history. But Lou Gehrig drove in 185 runs in 1931 and there were only 4.81 runs scored per game that year -- same as 2004, far fewer than 2000, and only slightly ahead of 2017. Further, the season is eight games longer now than it was then.

Intuitively, the pieces are in place for a real challenge to this record. Beyond intuition, though, are the plain facts: Wilson, when he drove in 191 runs, came to the plate with a total of 524 runners on base. He drove in 22.7 percent of them. And he drove himself in 56 times. Those are the three components of an RBI total, entirely: Runners on multiplied by percent driven in, plus home runs.

In none of those three components is Wilson all that special, nor is that era all that special. Wilson had 524 baserunners on; Albert Pujols had 526 baserunners in 2016, just three years ago. (And a higher percentage of Pujols' baserunners were in scoring position.) Justin Morneau had 558 baserunners in 2008. Derek Bell had 573 in 1996. Wilson's 526 rank just 69th all time.

OK, but Wilson drove in 22.7 percent of his baserunners, which is very high -- but, again, not that high: just 66th all time. Just the second highest in Wilson's own career. Allen Craig drove in a higher percentage in 2013. Will Clark and Pedro Guerrero both drove in a higher percentage in 1989, which was a low-offense year.

And then there's the home runs. Fifty-six home runs is a lot of home runs, but we've all seen players hit 56 home runs, even this decade. And the difference between 56 (a whole lot) and 47 (a more realistic lot) is a relatively small share of 191 RBIs.

So: We know players can get as many RBI opportunities as Wilson had, in an outlier season. And we know they can drive in as many of those runners as Wilson did, in an outlier season. It's just a matter of putting the right two outlier seasons together. And, to be honest and optimistic about it, there are probably a couple dozen guys who could do it. Allen Craig, folks!

Who'll do it and my totally made-up chance he does it: Manny Machado, 1 in 1,675.

9. Four homers in a game (Lou Gehrig, Scooter Gennett and others)

A quick story about Willie Mays, one of the 16 players who have hit four homers in game: One day in 1956, he stole four bases against lefty pitcher Vinegar Bend Mizell. After the game, they told Mays the record for a game was five. Shoot, Mays said. "If I knew I was near a record, I could have taken third base easy."

There's a lesson there: Knowing about the record is a big part of getting the record. But that's less and less true. Just as attaining a record has diminished in likelihood, so too has aiming for a record. Last year, Matt Carpenter hit three homers in the first six innings of a game. He also had two doubles, tying the all-time record for extra-base hits in a game. More importantly, he had a plausible path to five homers, a record that would have gotten his name remembered for many decades. He was on the visiting team, and his opponent was going to use position players to pitch the rest of the way, so he'd almost certainly get two more at-bats, and against non-pitchers.

Instead, Carpenter took the rest of the game off. "Not a lot of glory in that, anyway," he said of facing position players pitching. "I didn't even realize at the time that it was anything historic or anything crazy, honestly."

Carpenter gave two explanations there, and they contradict each other. One was Carpenter and the Cardinals didn't care about the record. The other was Carpenter and the Cardinals didn't know about the record, which ... well, that's incredible. Who doesn't know the record for homers in a game is four? But OK, maybe!

It makes all the difference in the world. Five homers in a game shouldn't be that hard. It is. I'll agree with that: It is very, very hard. In all of history, nobody has ever homered in five consecutive plate appearances, period. And in all of history, there have been only four plate appearances in which a batter was swinging for his fifth home run in a game.

But it shouldn't be. There were 4,456 batters last year who got five plate appearances in a nine-inning game, and they were disproportionately good batters. (Giancarlo Stanton batted five or more times 68 times, including 54 nine-inning games.) The chances of hitting homers in five straight at-bats is roughly one in a million for a top-tier home run hitter, so it's certainly not likely that somebody would do it (see: History), but that's a lot of swings. There were, further, 177 batters who got six plate appearances, and the chances of hitting five in six is closer to 1 in 100,000. (Five got seven plate appearances. Many more batters got six, seven or eight in extra innings, but a four-homer game is unlikely to be tied in the 13th inning.) And those fifth and sixth plate appearances come, almost by definition, against terrible pitchers pitching terribly.

It doesn't take six months of outrageous performance to hit five homers in a game. It doesn't even take 56 games of it. It just takes five at-bats in a row of a not-that-unlikely event happening. It's got to happen. Unless ... unless what Carpenter really meant was he didn't care to do it against a position player on the mound. Any game with a fifth home run on the line is likely to be a blowout, and any blowout in the late 2010s is likely to involve a position player pitching, and the unwritten rules of how seriously to try hitting against position players -- and whether there's any glory in that -- are still evolving.

Who'll do it and my totally made-up chance he does it: Giancarlo Stanton, 1 in 1,400.

8. 383 strikeouts (Nolan Ryan, 1973)

There have been two trend lines moving in opposite directions since Ryan set this mark in 1973. One goes up: pitcher strikeout rates. The other goes down: Pitcher innings totals. They are not independent of each other, but they are also not perfectly correlated, and sometimes one trend line angles more sharply than the other. Usually the innings totals have gone down faster than the strikeout rates have gone up, and neither Ryan's 383 or Sandy Koufax's 382 have been approached this decade. But sometimes, for some pitchers, the strikeout line has moved faster. In 2001, Randy Johnson struck out 372 batters in 250 innings. Last year's MLB leader had only 220 innings -- that line keeps going down -- but, meanwhile, Chris Sale set a record -- or is that a fun fact? -- for strikeouts per nine innings, at 13.5.

Innings will likely keep going down. Strikeouts will likely keep going up. A starting pitcher who strikes out 15 batters per nine innings is imaginable. A starting pitcher who throws 230 innings is imaginable. The same pitcher doing both in the same season is imaginable. Not likely, but predicting leaguewide trend lines is as hard as predicting players.

Who'll do it and my totally made-up chance he does it: Noah Syndergaard, 1 in 1,340.

7. An 84-game on-base streak (Ted Williams, 1949)

The case against it: Nobody has come within 10 games of Williams' mark, and nobody since Williams has come within 20 games of it.

The case for it: The post-Williams record is 63, set by ... Orlando Cabrera. That means that none of the great OBP hitters of the past 70 years has managed to get anywhere close to 84 -- no Bonds, no Rickey, no Mays, no Votto -- but it also means a great hitter would only have to be 30 percent better than Orlando Cabrera. I believe that man exists.

Who'll do it and my totally made-up chance he does it: Bryce Harper, 1 in 1,195.

6. 14.1 WAR (Babe Ruth, 1923)

As a reminder, Ruth set the WAR record in one of those Babe Ruth seasons most of us don't even really think about: 1923, when in lieu of setting a new home run record he did everything right. We in the modern era get to watch two of the all-time do-everything-right players, in Mike Trout and Mookie Betts, and last year each of them played at a 14.2-WAR pace for extended periods of time -- Trout in the months of April and May:

April: 13.7 WAR (prorated over 162 games)

May: 14.6
June: 12.2
July: 6.5
August: 3.2
September: 13.8

And Betts over the final two months of the season:

April: 11.5
May: 15.9
June: 7.5
July: 11.5
August: 14.4
September: 15.0

Each player's pursuit of the record was thwarted by absence -- injuries for Betts, injuries and bereavement for Trout -- but each finished above 10.0 WAR, just the second time since 1948 that two players did so. Trout's WAR prorates to almost 12 WAR over 162 games, which would be the sixth highest ever; Betts' prorates to 13 WAR, the second-best season ever.

This should, then, maybe even be higher than it is. But that final win that even a full-season Betts would have needed is awfully tough. Babe's 14.1 is a lot higher than Babe's 12.9 (the second-best season ever), when he hit 59 homers. Think of the best seasons in history -- they're easy to think of off the top of your head -- and realize each is more than a win, in almost all cases two or three or four wins behind 14.1. Considering what that 14.1 WAR is measuring -- the defense and baserunning and park effects of baseball that was played 95 years ago -- it's easy to imagine there is a lot of margin for error in that number. Maybe Ruth wasn't really a 14.1-WAR player, but a 12-WAR player with a bunch of measurement errors goofing up the total. Which suggests maybe nobody has ever been a 14.1-WAR player, and that realistically nobody can be one.

Except we just saw Trout and Betts each do it for months at a time.

Who'll do it and my totally made-up chance he does it: Both Trout and Betts, 1 in 875 apiece.

5. 67 doubles (Earl Webb, 1931)

This one gets challenged into late spring pretty regularly: David Ortiz was on pace for 72 in June 2016. Manny Machado was on pace through June 2013. Chuck Knoblauch was on pace in August 1994 when a strike ended that season. And plenty of other hitters have hit a dozen doubles a month through April or April plus May. Nothing about baseball has squelched doubles hitting, and doubles rates last year were just about the same as they were in 1931, when Webb set this one. The second-highest leaguewide doubles year in history was in 2007, followed by 2006 and 2008, so clearly "a lot of doubles" fits in modern baseball.

It's not going to get broken, to be clear. In the past 10 years, the best first half was Machado's 39 in 2013, and the best second half was Jose Ramirez's 29 in 2017. Put those together and you'd have 68, a record. But those are the two best halves of the past 10 years, out of thousands of halves by hundreds of very good hitters. It's not likely.

Who'll do it and the chances he will do it: Doubles are pretty random, and 50 names could fit here. (Webb never hit more than 30 in any other season.) Say, Xander Bogaerts, at 1 in 3,700, and 49 other guys at 1 in 3,701.

4. 62 saves (Francisco Rodriguez, 2008)

OK, so the record is 62. To beat this record, you need at least 62 opportunities. You know how many pitchers have ever had 62 save opportunities in a season?

Two. Two! Edwin Diaz had 61 last year, the third pitcher ever to top 60. Obviously, you can't possibly be good enough to get 62 saves in 61 opportunities, and if you're planning to blow a few you're going to need a bunch more.

But here's the other thing: Francisco Rodriguez, who had 62, blew seven. He had 69 save opportunities! And Bobby Thigpen, whose record of 57 Rodriguez broke, blew eight. So only two pitchers have ever had 62 save opportunities, but one had 65 and one had 69, so not only is 62 possible, but a healthy bit more is, too.

Not to diminish what Thigpen and Rodriguez did, but sort of to diminish it: The history of the save record is really the history of the save opportunity record. Rodriguez had a record number of opportunities. Before him, Thigpen had the record number of opportunities. Before him, Dave Righetti did -- the year he set the previous saves record, 46. Before him, Dan Quisenberry did -- the year he set the previous saves record, 45. So it might seem hard to break this record, but really it's just one of those things, and never put it past baseball to do one of those things.

Who'll do it and my totally made-up chance he does it: Seranthony Dominguez, 1 in 770.

3. A 59-inning scoreless streak (Orel Hershiser, 1988)

Some reliever is going to have a scoreless season one of these days. That's my whole argument: Relievers. Zach Britton had a 63-inning stretch when he allowed one earned run. That is, definitely not a scoreless-innings streak. But there are too many relievers capable of knocking out a 0.60 ERA for us to be too surprised when one has a scoreless season. Going to happen. I mean, obviously it's probably not, but it will.

Who'll do it and my totally made-up chance he does it: You've literally never even heard of the guy yet. He's the sixth man on Tampa Bay's bullpen depth chart right now, and before the season starts he'll be put on waivers, picked up by the Reds, start the year in Triple-A, and he'll set the record after he gets called up on April 14. His chances of doing it are about one in a million, but there are 2,000 active professional pitchers with equally good chances.

2. An undefeated starting pitcher

Wait, this isn't even a record! This was originally going to be the entry for 31 wins, and it was originally going to be around 19th, but it was determined to be too anachronistic -- more than 13 shutouts, even. Instead, we're declaring the record to be an undefeated season for a starting pitcher.

Nobody has ever done that. The most wins in an undefeated season is a mere 12, and the record for a pitcher who didn't relieve games is only nine. The best won-loss record by a pitcher who qualified for the ERA title (one inning per team game played) is 15-1, by Johnny Allen in 1937. You didn't know that, because it's not a real record, because it's not a good record. But an undefeated season would be, especially if it also broke the record for most consecutive wins in a season (19).

As deGrom showed last year -- when he went 10-9 with a historically low ERA -- this is a hard one to control. I once hypothesized a game-by-game replay of Pedro Martinez's 1999 season, which is probably the best pitching season of all time, in a world where Cleveland, rather than Boston, traded for him. (In 1999, Cleveland was the highest-scoring team since 1950.) I figured he'd have gone 23-1.

It's easier now, with shorter outings and fewer starts. It's harder to get wins, but it's also harder to get losses. An 18-0 season could happen this season, and I think it'd be a pretty big deal.

Who'll do it and my totally made-up chance he does it: Chris Sale, 1 in 460.

1. 20 Ks in a nine-inning game (Roger Clemens, Kerry Wood, Max Scherzer and Randy Johnson)

Here's an entire other article for you to read about why 21 K's in a game is the best record that can actually happen. Here's the key section:

A 21-strikeout game fits perfectly within the established limits of human performance and baseball's natural variation. Scherzer and Clemens each struck out 20 batters with at least one out to go (both got ground outs), and Johnson, Ramon Martinez and Wood all entered ninth innings with 18 strikeouts. Corey Kluber and Johnson both struck out 18 batters in eight innings before being pulled from starts. We've been close, so close. Indeed, there have been 26 starts in which a pitcher has struck out at least 10 batters and maintained at least a 21-K pace before being pulled from the game, and scores more pitchers have been pulled with shots at 21 still mathematically intact.

And, most importantly, we've seen pitchers strike out 21 (or more) batters across 27 outs -- "hidden" 21-K games, spread out over multiple outings. They've mostly been relievers, such as Ken Giles (23 K's in 27 outs in 2016) and Edwin Diaz (24 K's in 27 outs the same year). (A minor leaguer named Jacob Webb even had 25 K's over 27 outs in Class A.) Kluber did it, striking out the side in the first inning after his 18-K start, for 21 K's across nine innings. The sequence of events necessary for 21 strikeouts in a game has been proved possible.

Almost every record is the product of its environment. Nobody is going to hit 73 homers in a low-homer environment, just like nobody saved 62 games before modern bullpen usage solidified. This is a strikeout era. In the past week, we've seen three credible 21-K alerts: James Paxton had 16 in seven innings, with six outs to go; Gerrit Cole had 12 through six innings, but ended up with only 16; and Scherzer had 15 in 6⅓ innings before he was pulled. The means are there.

You might have noticed, though, something important in that blockquote: "James Paxton had 16 in seven innings, with six outs to go." Then he was pulled, at just 105 pitches. The record is dying, my friends. They don't even care about this incredibly cool thing that could have been done.

Paxton did get 31 swinging strikes, the most by any pitcher in any start last year. Decently fun fact.

Who'll do it and my totally made-up chance he does it: Max Scherzer, 1 in 80.