For all the statistics baseball fans like to obsess over, there’s really only one that matters in any given game: scoring more runs than the other guys.

And after 25 games, the Cubs have been scoring a lot more runs than the other guys. They lead all of baseball with a run differential of +89, well more than double any other team.

According to ESPN, this is "the third best [start] in the modern era and the best since the 1905 New York Giants." The Cubs are tied with the legendary 1939 Yankees, a team that would rack up a +411 run differential on its way to a 106-45 record and a sweep of the Reds in the World Series.

It’s no surprise that after 25 games the Cubs have their best record since 1906.

Their 19-6 record reflects, depending on how you think about these things, some bad luck. Baseball Prospectus calculates three different versions of what a team's record should be, based on things like underlying statistics and opponent quality; according to those, the Cubs should either be 21-4 or 22-3.

Further statistical craziness: the Cubs have a run differential of +44 in games Jake Arrieta has started, accounting for half of their entire run differential in his six starts: he's allowed four runs (for an ERA of 0.84), and they've scored 40. [Clarification: the run differential includes runs scored by either team after Arrieta was pulled.] Which is part of their bad luck; they'd be better off if they were scoring 8.67 runs a game in John Lackey's starts to support his 4.32 ERA. Then again, they're averaging 7.60 runs in Lackey's starts, so that's OK, too.

The interesting thing to look at going forward is less the pitching than the hitting. It's unlikely that Arrieta, Jason Hammel, and John Lester will finish the season with ERAs under two. They all have strand rates over 90 percent; that will come down, and they will give up more runs.

But as of right now the Cubs have scored 153 runs, tied for the best in baseball with the Cardinals. And it looks like that could be sustainable. Dexter Fowler is playing way over his head, slashing .352/.473/.602 with a .438 batting average on balls in play. Dexter Fowler is not the fifth-best hitter in baseball, no matter what the stats say. Matt Szczur's .367/.441/.600 line is also unlikely to keep up.

On the other hand, there’s room for improvement. Anthony Rizzo is hitting .242 with a .206 BABIP. Jason Heyward—historically a slow starter—is hitting .211. Jorge Soler is hitting an abysmal .188, but he's raised his walk rate to 11 percent and dropped his strikeout rate from 30 percent to 21 percent; his BABIP is an extremely low .208.

So even if their pitching regresses to being merely excellent and the couple regulars who are riding what looks like luck fall back to earth, they have a couple slow-starting stars to pick up the slack. And one major change looks to stick: a couple of weeks ago, August Fagerstrom of FanGraphs said the Cubs could have the most disciplined lineup ever—in the post-1961 expansion era—when they had played seven games and walked 13.1 percent of the time.

"It’s not expected to stay that high, but it is expected to stay the highest," Fagerstrom wrote. It's now 13.3 percent. Only one regular on the Cubs, Javy Baez, has a walk rate below 10.5 percent, and he's not even an everyday player.

More statistical anomalies: the Cubs are the best baserunning team in the majors per FanGraphs data (10.4 BsR, compared to the second-place Padres at 6.4). I wrote about their walk rate and baserunning improvements last May; some of that can be attributed to the hiring of Joe Maddon. I also pointed out at the beginning of the season how Jason Heyward was a vast improvement on the basepaths over Jorge Soler, whom he was then expected to replace; Heyward has the best BsR score on the Cubs so far this season, and the second-best in the majors.

And they're the best defensive team—with the usual caveats for small sample size and tetchy metrics—by FanGraphs's Defense score (21.4, compared to the Rangers at 14.3) and Baseball Reference's Total Zone Fielding Runs (22 above average, compared to the Nationals at 17). Again, as I've written before, this looks like something that can be attributed in part to Maddon's hiring.

So … the Cubs are really good. You can be really good and really lucky, like the Cardinals were last year, and by cluster luck at least, the Cubs appear to have been pretty lucky too this year. But the Cubs have built a remarkably deep team. They have a good bench; they have a team full of positional flexibility; and their players have, generally speaking, broad skill sets. They'll have to survive changes in luck to keep up their pace as a historically great team, and Cubs fans know from bad luck. But it looks like they've built the team to do it.

And that other team in town? At 19-8, the White Sox happen to have the second-best record in baseball. Together the Cubs and White Sox were a ridiculous 22 games above .500 after 25 games.

Rarely are both teams this good at the same time. How rare? At this point in the season it's happened exactly once, in 1907, the year the Cubs won the first of their only two World Series championships.

Obviously a lot can happen between now and October, but if this keeps up, a rematch of 1906 could very well be in store.