Twenty-five years ago, millions of baseball fans around the country turned on their televisions expecting to watch a World Series game — and saw live footage of a deadly earthquake instead. The San Francisco Giants and the Oakland A’s, and the 62,000 fans watching them in Candlestick Park in San Francisco, felt the ground under them shake. The baseball commissioner thought it was a jet flying overhead. Oakland’s manager thought the crowd was stomping its feet. Then a section of the right-field stands separated in two by a few inches. Players ran to gather their family members and get out of the ballpark.

The earthquake killed 63 people — and might have killed more if there was no game on TV to keep people off area roads — and the series didn’t resume for 10 days. (The earthquake is the subject of the “30 for 30″ film “The Day The Series Stopped,” airing Tuesday night on ESPN.)

There have been deadlier earthquakes, and costlier ones, but few that surprised more people the moment they occurred. The millions of television viewers didn’t get what they expected because scientists couldn’t predict when an earthquake would strike.

The Loma Prieta earthquake helped fuel efforts to change that, but not much progress has been made. A decade after the quake, Robert J. Geller, professor of earth and planetary science at the University of Tokyo’s Graduate School of Science, wrote that earthquake prediction “seems to be the alchemy of our times.” Seismologists have mostly forsaken their quest for precise predictions, turning instead to more modest projects like telling the public when the probability of an earthquake has risen to 1 percent from 0.01 percent. They can’t predict whether or when they’ll be able to do any better.

Today, earthquake scientists in the United States and several other countries are working on producing “seismic weather reports” — a phrase Thomas H. Jordan, director of the Southern California Earthquake Center, uses to describe a continuously updated, local estimate of the probability of an earthquake. Just as you can look up the probability of rain in your area, the seismic forecast would let you look up the probability of an earthquake — but it wouldn’t be nearly as accurate.

These forecasts won’t tell people precisely where and when an earthquake will strike, or what magnitude it will be. Instead, the forecasts will show whether the baseline probability of an earthquake has risen.

But even that relatively modest endeavor faces many challenges, including computing power, communicating risk to the public, and swaying skeptics within seismology. “Crying earthquake (wolf) is a potent way of blunting earthquake awareness and preparedness,” Kelin Wang and Garry Rogers wrote in the journal Seismological Research Letters earlier this year. Wang, a research scientist at the Geological Survey of Canada’s Pacific Geoscience Centre, thinks that earthquake forecasting is a promising area of research, but the trick is translating those forecasts into something that won’t be counterproductive when it reaches the public.

Jordan is less concerned about spooking the public. In a response to Wang, Rogers, and other critics in Seismological Research Letters last month, Jordan said that Americans have experience processing low probabilities for catastrophic events. We’ve grown accustomed to hearing about heightened awareness of terrorist attacks, and wildfire warnings have become a feature of Californian life.

Ned Field, a seismologist for the U.S. Geological Survey, thinks there’s an audience for short-term forecasts. Prospective buyers of earthquake insurance, homeowners considering whether to leave town or homeowners wondering whether to build a basement could all benefit. (The USGS is already using an app to test how to spread the word in case of elevated earthquake risk in southern California.) He envisions marrying short-term forecasts with the agency’s model that estimates an earthquake’s economic costs and fatalities. That would give the public a sense not only of how likely an earthquake is, but also how much damage it could create if it occurs.

If they work, the sorts of forecasts Jordan and the USGS have in mind must avoid being overly definite, but they can’t be too vague, either. As my colleague Nate Silver wrote in his book “The Signal and the Noise,” just before a deadly Italian earthquake in 2009, scientific technician Giampaolo Giuliani spoke as if earthquakes could be predicted. He said an earthquake was coming, and based his prediction on an unproven technique of measuring radon-gas emissions. Meanwhile, more reputable earthquake scientists spoke from the other extreme, taking the view that earthquakes were no more or less likely at any given time. Both were wrong: Giuliani’s forecast missed the time, place and magnitude of the tremors that killed 309 people in L’Aquila. But the scientists were wrong, too. They discounted the significance of small earthquakes in the area, which in retrospect were foreshocks of the bigger earthquake.

The middle ground is the one the USGS is seeking: to tell the public when seismologists know there’s an elevated risk, without overstating their confidence in the prediction.¹

The agency has tried this before. In 2005, it introduced on its website a tool that allowed people to check the chances of an earthquake in their area. But the code powering the tool kept crashing, Field said, and the USGS removed it from its website in 2010.

The models used by Jordan and other scientists today analyze recent seismic activity to predict the probability of future earthquakes. Jordan and Field are focusing on a model derived from something called ETAS, or Epidemic-Type Aftershock Sequence Model, which projects the proliferation of tremors in the way a disease might spread.²

There are many other models besides ETAS. Just how many depends on how you count. Jordan estimates there are 400 models worldwide vying for pre-eminence and being tested by the Collaboratory for the Study of Earthquake Predictability, which he directs. Field says USGS alone is considering 5,000 models. One model can differ from another merely in how it represents the structure of the earth’s crust.

In theory, all these models should be competing in a test of which best forecasts seismic activity. In reality, there just aren’t that many high-magnitude earthquakes in the world that can serve as tests. That’s good news for anyone living on a fault line, but not for seismologists, who would like to know whether their models are correctly calibrated to pick up the greater risk just before a major earthquake. So instead the tests generally focus on whether the models correctly predict the frequency of lower-magnitude earthquakes. Globally, there is a reliable relationship between the rate at which these occur and the rate at which major earthquakes occur.³ But in any given spot, that relationship may not apply — and not all models can be tested globally because not all places have the level of measurement of crust structure and seismic activity that, say, California does.⁴ Also, a model tuned to pick up small quakes may not pick up bigger ones.

With so many competing models, there is another risk: The one that does best in tests might just be getting lucky.

Even the best model wouldn’t predict about 50 percent of the big earthquakes around the world. The quakes that go undetected would be the ones that seemingly come out of nowhere, without foreshocks. For instance, it’s unlikely the best model would have warned residents of Napa County of the greater risk of an earthquake before a big one struck this past August, killing one person and causing an estimated $1 billion in damages. And it’s hard to say if any model could have foreseen the Loma Prieta earthquake, according to Jordan, because not all the data the model uses to produce forecasts was available in 1989.

Jordan calls the hunt for a more precise earthquake prediction “a silver-bullet approach,” trying to find “some magic signal.” What he and his collaborators — there are more than 50 — are doing today “is very different … There’s nothing magic about it,” he said.

One prediction the forecasters are comfortable making is that we won’t get more definite predictions anytime soon — if ever. “I would not be at all surprised if earthquakes are just practically, inherently unpredictable,” Field said. “You never know; some silver bullet could come along and prove useful.”

In less than a week, you may have heard, there’s a midterm election in the United States of America. This is sort of a big deal for us at FiveThirtyEight. Such a big deal that our estimable tech team of Jeremy Weinrib and Paul Schreiber arranged a fancy live-blogging platform so you can snuggle up next to us for hours on election night. It’ll be cozy.

We’ve known for weeks that we’d need to give the platform a test drive, and we decided that we’d do that Wednesday, on the second night of the NBA season. We’d get together our crew of basketball writers (the ones who wrote our NBA team previews), buy some pizzas and use an algorithm to project whether Giannis Antetokounmpo has finally stopped growing.

But as the San Francisco Giants discovered last night, Jake Peavy has a habit of ruining the best-laid plans.

About the time Game 6 of the World Series passed a 95 percent win probability, we made the call to scuttle the NBA live blog. Instead, you’ll get to hang with us as we watch Game 7. We’ll argue that Jeremy Guthrie shouldn’t pitch more than three innings, locate where the Giants dynasty of the past five seasons would rank compared to others and, Yost-willing, debate the merits of the sacrifice bunt.

It’s going to be great. Or a total disaster. Come and find out which. 8 p.m. EDT Wednesday. Here on FiveThirtyEight.

Welcome to FiveThirtyEight’s live blog of Game 7 of the 2014 World Series. We’re doing some testing of this format, so please feel free to send us feedback.

The Kansas City Royals just concluded the most improbable, exhilarating, agonizing and — ultimately — meaningful season in the franchise’s past three decades of existence. That they lost the World Series to the inexplicably dynastic San Francisco Giants in a memorable Game 7 that literally came down to the final out doesn’t soften the blow for Royals fans, but it does put in perspective what Kansas City accomplished this season: they essentially came as close as any team possibly could to winning the World Series without actually hoisting a World Series trophy.

(As Grantland’s Royals-fan-in-residence Rany Jazayerli put it: “[Kansas City] went 11-4 in the postseason. That’s not only the best postseason record any team has managed without winning the World Series — it’s the best possible record a team can have under the current format without winning the World Series.”)

And yet, as the Giants’ celebration was unfolding under the Kansas City sky Wednesday night, there was the distinct sense that the clock had also struck midnight on the Royals’ 2014 cinderella story. Kansas City had not won their division, narrowly finishing first in the American League wild card standings over two teams (the Oakland Athletics and the Seattle Mariners) who’d posted vastly superior run differentials during the regular season. Going into the playoffs, the probability of a team like Kansas City, with an 84-78 pythagorean record, getting through the AL’s postseason minefield to the World Series was just 6.3 percent.

It took a lot of good fortune for the Royals to even make it as far as they did. But can they make it back?

To (roughly) answer that question, I built a simple model predicting the probability of a World Series contestant making another World Series appearance at any point in the five seasons after its initial showing. (The ingredients for it can be seen at the bottom of this post.) That model says the Royals have just a 30 percent probability of going back to the World Series at any point in the next five seasons (the average World Series participant returns about 46 percent of the time).

Coincidentally, that 30 percent probability means Kansas City currently has roughly the same chances of a World Series return as the model assigned to the 1985 Royals (which had a similar age and weighted pythagorean winning percentage as the 2014 Royals). That version of the team, as Kansas City knows well, never made it back to the World Series. Other similar teams include the 2007 Colorado Rockies and 1984 San Diego Padres, both of which drew easy comparisons during the Royals’ October run — and both of which were World Series one-hit wonders. Of the 10 historical World Series contestants most similar to this year’s Royals, only one — the 1992 Atlanta Braves — ever found their way to another World Series within the next five seasons.

Then again, the model isn’t too much more optimistic about the Giants team that beat the Royals, assigning San Francisco a 34 percent probability of ever going back to the World Series with their current group. Yet the Giants have already bucked similar odds twice — they returned after the 2010 edition of the team appeared to have a 33 percent chance of doing so, and made another repeat trip after the 2012 edition was assigned a 37 percent probability of going back to the World Series. For their part, the Royals have the eighth-best farm system in MLB according to Baseball America, and have had top-10 prospect classes in three of the last four such lists issued by Baseball America. The talent is ostensibly there to keep the Royals’ 2014 run from being a total fluke.

The truth is that most World Series entrants fail to return within the next few years. It’s hard enough to make it to one World Series, let alone two in the span of five seasons. MLB’s current playoff structure ensures a high degree of randomness, with mechanisms in place to prevent the best teams from running the table. The Royals are most likely to fade away like so many other World Series teams have throughout baseball history.

***

For the model readers out there: I used data since MLB expanded its playoffs to include four teams in 1969, fitting a logistic regression model that used an average of the team’s previous five seasons’ worth of pythagorean winning percentages (weighted 5-4-3-2-1, from most recent to least recent, to give more influence to more relevant seasons) and the average of the team’s mean ages for its batters and pitchers. (I also considered additional variables, such as the market size of the team’s home city and whether or not the team actually won the World Series in question, plus dummy variables for MLB’s free agency and wild card eras, but none of those proved statistically significant.)

While Kansas City’s players were younger than the typical World Series team — their mean age was 28.3, compared to 29 for the average World Series team — their weighted pythagorean winning percentage (.497) was league-average at best over the previous five seasons. The pythagorean-record factor has historically been 50 percent more important than a team’s average age, and the Royals had the eighth-worst such mark of any World Series participant since 1969.

Stocked with a cast of insanely talented young players, Tuesday’s MLB All-Star Game in Cincinnati was a perfectly pleasant affair, and a nice showcase for the current (and future) state of the sport. But it wouldn’t be an All-Star Game if it weren’t also noted that the outcome of a silly exhibition contest continues to be used to determine home-field advantage in the World Series.

That’s pretty dumb, and there’s no shortage of people in the game who’d like to see it changed. (Except, apparently, new MLB Commissioner Rob Manfred.) There are also plenty of alternative suggestions for what should determine World Series home-field instead. But as dumb as it is for the All-Star Game to “count,” what really matters is the effect the rule has had on the 12 Fall Classics since it was put into place.

To quantify the ramifications of the policy, I calculated pre-series win probabilities for all World Series teams from 2003 to 2014 using their regular-season pythagorean records.¹ I looked at how much those odds shifted depending on whether the frivolous, All-Star-based home-field rule was used, or two alternatives: the equally arbitrary (but at least consistent) pre-2003 policy of alternating home-field between leagues each season, and a simple rule that bestowed home-field upon the team with the superior regular-season record.

Going from the current format to an old-school approach that alternates home-field by league every year, the pre-series odds would have shifted by an average of +/- 1.4 percentage points each year over the past 12 seasons. Home-field itself would have been different seven times in this alternate universe, though it probably wouldn’t have made much difference to the outcomes of most series. Twice in 12 years (2005 and 2013), the margin between the teams was slim enough that swapping home-field would have changed which team was favored.

The team with home-field in reality won each of those series, so the argument could be made that linking home-field advantage to the All-Star Game swung a pair of championships. But each of those series was also nearly 50-50 regardless of who held home-field, so you’d expect a different set of results in those series (compared with reality) only about 75 percent of the time if they were replayed. In total, if we re-simulated the last 12 World Series in this universe, we’d see about 6.2 different winners over that span, but only 1.7 of them could be directly traced to ditching the All-Star Game-based home-field format. Per decade, that works out to about 1.4 titles exchanging hands because of the change in format.

That number gets smaller if we compare the actual results since 2003 to another hypothetical universe in which home-field is assigned to the team with the better regular-season record. (This seems to be the most popular suggested format change among the reformist crowd.) Under those rules, the favorite would change for only one World Series (2013), and if we replayed the past 12 years numerous times, we’d see only 0.7 different champions as a result of the format change than we saw in reality (which averages to 0.6 per decade). The more that things would change, the more they’d kinda sorta stay the same.

This isn’t to say that MLB shouldn’t look into switching to a more sensical method of determining home-field advantage in the World Series — preferably one that didn’t involve an exhibition game. But the stupidity of the current policy probably outpaces its actual effect, given that the difference between it and a record-based approach is one different (perhaps more deserving) champion every 17 years.

Why should politics get to have all the fun with chats? In preparation for the World Series, which starts Tuesday, we summoned the biggest baseball obsessives on FiveThirtyEight’s staff to Slack to talk about the New York Mets and Kansas City Royals. As usual, the transcript below has been lightly edited.

neil_paine (Neil Paine, senior sportswriter): After the long season, it all comes down to the Royals and the Mets. So, first things first, let’s just put it out there: How do these two teams compare? Kansas City had a better regular-season record (95 wins vs. 90 for New York), but what do the deeper indicators say?

hjenten (Harry Enten, senior political writer and huge baseball fan): Well, let me take the surface-layer answer here as a non-sportswriter (though one who follows some of the deeper statistics) — based off runs scored and allowed (i.e., the Pythagorean win formula), the teams are very close. The Royals should have won 90 games, while the Mets should have won 89 games.

rob (Rob Arthur, baseball columnist): I would echo Harry and say that the teams are quite even. BaseRuns gives the Mets a small but significant edge, but the Royals relief corps plays up in the postseason, when they can use their top three relievers more.

hjenten: I think that, of course, ignores the fact that the Mets were a far different team after they traded for Yoenis Cespedes in late July. The Mets went a combined .627 in August, September and October, while the Royals went .567.

neil_paine: Rob, to your point, it seems like the Royals used “sequencing” to their advantage, but there may be reason to think that at least some of that is real and can persist in the World Series?

rob: The Royals are a very contact-oriented team, with the lowest strikeout percentage in the majors by a large margin. They are also excellent against high-velocity fastballs, which pitchers often go to when they are in a jam. Those attributes may give them a tiny edge in terms of sequencing, but there’s no question that they’ve been overachieving as well.

neil_paine: And for what it’s worth, it should probably be mentioned that the Royals played a tougher schedule (certainly during the regular season); Baseball-Reference.com’s Simple Rating System has the Royals ranked 10th in baseball in strength of schedule and the Mets 29th, which probably feeds into the difference in their Elo ratings even now:

hjenten: Of course, the Mets just beat the team that had the NL’s best Elo rating going into the league championship series, the Cubs …

neil_paine: That’s true — Elo gave the Cubs about a 60 percent chance of winning that series, and the Mets won (swept!!) anyway; now Elo gives KC a 55 percent chance of beating New York, so grain of salt and all that.

Harry, you brought up a difference in the two teams as the season went on. Obviously the Mets added Cespedes at the trade deadline, but the Royals also did some dealing — some of which has worked out better than others.

rob: The Royals grabbed Johnny Cueto and Ben Zobrist at the deadline. And while Zobrist has been his typical self — with a .933 postseason OPS (on-base plus slugging) — Cueto has not looked comfortable. It’s hard to tell how much of that is normal performance fluctuation and how much is real. There are reasons that Cueto might not be at his best: a new catcher in Salvador Perez, the fatigue of a long season and many innings, and the ever-present threat of a hidden injury.

hjenten: Yes, and the Royals were actually slightly worse in the final few months of the year than they were overall. The Mets, on the other hand, were clearly better after those trades.

neil_paine: How much of that do we think is because the Royals all but locked up the division and the playoffs so early? And maybe more to that point, how much credence do we give to playing your best baseball right now? Certainly that’s what the Mets seem to be doing.

rob: I don’t put very much stock in the second-half stats, partially because the Mets put up those stats against the weakest second-half schedule in MLB. As for the more recent postseason performance, a lot of the Mets’ playoff run has been fueled by an insane stretch from Daniel Murphy, which likely will not continue.

It’s also worth noting that Cespedes is injured, so if he’s the cause of their second-half surge, the Mets may be in trouble.

hjenten: Well, here’s why I would put a little more trust in the late-season records. Yes, the Mets beat up on some crummy teams, but they also went 7-2 against the Nationals from July 31 on and more than whacked the ever-loving snuff out of some of those bad teams. They scored 320 runs and gave up 243 (a 103-win pace, according to Pythagoras); meanwhile, the Royals scored 290 and gave up 269 (an 87-win pace). In other words, NY was playing far better ball than KC over the season’s final two-plus months.

And it’s not like Kansas City has dominated the postseason, either. If not for a miracle against Houston, they’d be sitting at home like we are. As for Cespedes — yeah, he got hurt in that final game, but I haven’t seen any signs that he’s not going to play. We’ll see.

neil_paine: Since you guys mentioned Cueto, Murphy and Cespedes, let’s talk about the Mets’ hitting against the Royals’ pitching. I think we can all agree that Murphy will eventually cool down and stop hitting like a real-life Roy Hobbs, so where will the Mets’ offense come from when that happens? Do they have any players on cold streaks that might regress to the mean (in a positive way) and cover for Murphy?

rob: One reason for Mets optimism is the fact that so many of their players have been underachieving in the postseason: David Wright, Lucas Duda and Travis d’Arnaud all spring to mind. So even if (when!) Murphy cools off, one of these guys might step it up. They have an OK offense by BaseRuns, so they shouldn’t need Murphy to be superhuman forever.

hjenten: I think we’re obviously dealing with small sample sizes in the playoffs, but during the regular season, the Mets’ best hitter was not Daniel Murphy. D’Arnaud (who hit two homers in the NLCS) posted a 128 OPS+ in limited action, and Mike Conforto (who homered in the division series) had a 132 OPS+. Obviously Duda (132) and Cespedes (157) were also outstanding hitters. Duda had been in a slump, but he showed some signs of life in Game 4. To me, the offense is less of a question than whether Jacob deGrom, Thor and Matt Harvey can continue to pitch as well as they have been doing.

neil_paine: We’ll certainly get to those three later, but before we do: Is KC’s rotation as concerning as it might seem? Can its dominating bullpen make up for it?

rob: The Royals’ rotation is worrisome, especially with Cueto potentially being in trouble. But as you mentioned, Neil, the Royals have the tools to make up for their shaky starters: an incredible defense (best in MLB last year by a HUGE margin) and a top-notch group of relievers. Here, too, the Royals may be able to take advantage of sequencing, by putting their good relievers in at the moments when the lead is most threatened. (This relies on Ned Yost knowing when to put his best relievers in, but he’s been much improved in that regard of late — Game 6 of the ALCS aside.)

I think it will come down to whether the Mets can chase the Royals starters early and expose the weaker members of the bullpen. Because once it gets to their best relievers in the late innings, the Royals become very hard to beat.

hjenten: And the Mets bullpen, outside of Jeurys Familia, is quite troubling. Their eighth-inning man, Tyler Clippard, has been anything but steady, giving up three runs in 4 2/3 innings this postseason. (That’s after giving up 10 runs in 14 2/3 innings in September/October.) And their lefty specialist is Jon Niese, who was rubbish as a starter. I’d say their best pitcher out of the pen at this point (besides Familia) is Bartolo Colon. That’s never a good sign.

rob: Don’t underestimate Fireman Bart!

neil_paine: But we’d be remiss if we didn’t also point out that the Mets’ starting rotation of deGrom, Syndergaard, Harvey (you mentioned these guys earlier, Harry) and Steven Matz have been otherworldly in the postseason so far. I crunched the numbers, and New York is allowing the second-lowest playoff FIP (fielding independent pitching) — relative to the league average — of any pennant winner during the wild-card era (since 1995).

Rob, you talked about the overwhelming velocity of New York’s starters when the Mets played the Cubs in the NLCS and how the strikeout-prone Cubs might have been particularly ill-equipped to deal with their heat. But now we almost have the polar opposite kind of lineup with the Royals.

rob: Right, the Royals are contact-heavy. And there’s some limited evidence that they are better than average at hitting heat. So as far as the matchup between the Mets’ starters and KC’s lineup, this looks to be strength against strength, and it’s not at all clear how that battle will end up.

It’s worth noting as well that temperature plays a significant role in offense, and it should be pretty chilly.

hjenten: On the weather front: Going for a low of 49 in KC on Tuesday night, 40 on Wednesday night. 42 in NYC on Friday night. 45 on Saturday night. So it will not be warm.

rob: Cold temperatures mean less offense, so when you mix the cold, high velocity and the quality Mets starters, we could see the Royals doing a lot more whiffing than they are used to.

hjenten: The Mets’ great starting pitching is not a fluke, either. During the regular season, deGrom had a 2.70 FIP, Harvey 3.05, Syndergaard 3.25. Matz was the worst, with 3.61 in limited time — still 7 percent better than the NL average. That stands in contrast to the Royals’ starters (during their time with KC): Only Ventura (3.57 FIP) was better than any of the Mets starters were during the regular season.

neil_paine: Through that lens, it looks like a pretty big mismatch of starting rotations.

hjenten: Now, can the Royals, who hit for contact, avoid that? Maybe. As you mentioned, the home-run-hitting Cubs were the opposite of the Royals in that regard. The key for the Royals is to get into that Mets bullpen early. If they can, the Mets are in trouble.

rob: And bear in mind that FIP won’t be totally fair to the Royals’ pitchers. Because their pitchers aren’t fielding-independent: Every liner they give up goes toward Lorenzo Cain or Alcides Escobar or some other, similar-competent defender.

neil_paine: That’s a great point, Rob. Although relative to the league, the Royals in this postseason actually look worse by ERA than by FIP!

hjenten: Here’s another thing. Compared with the regular-season numbers for deGrom (2.54 ERA), Harvey (2.71), Syndergaard (3.24) and Matz (2.27), the only Royals starter with a better ERA than any of them was Chris Young, at 3.06. No one else was under 3.50. (And it isn’t as though the average ERAs for the AL and NL were drastically different — the NL’s ERA was about a tenth of a run lower.)

rob: Ha — well there goes that idea.

neil_paine: We mentioned the cold and how it might not be conducive for hitting. Does this give an edge to a team like the Royals, who are pretty adept at the small-ball things that win one-run games? Or have we seen enough of that out of the Mets this postseason (even if it’s uncharacteristic, given their regular season) that it might not be as much of a lopsided matchup on the basepaths?

hjenten: The Mets didn’t really run in the regular season, with only 51 stolen bases (worst in the NL). Meanwhile, the Royals stole 104 times (second in the AL). But in the postseason, you’re right, it’s different: The Mets have had nine steals, while the Royals have had seven. Cespedes can run, and Granderson can definitely run (three stolen bases in the NLCS).

rob: In theory, the Mets have the baserunning edge anyway. That’s hard to believe when you look at specific players like Lorenzo Cain (who scored the series-winning run from first on a single in Game 6 of the ALCS), but bear in mind that Cain is counterbalanced to some extent by Kendrys Morales, Mike Moustakas and others.

neil_paine: So it sounds like that won’t be as much of an edge for KC as it’s often made out to be.

rob: Yep. I think a lot will hinge on the Royals’ appetite for (and execution of) those small-ball tactics (like sacrifice bunts) that are despised by modern sabermetrics. It works for them — or at least it has worked, over and over — but the percentages suggest that it isn’t helping them in the long run. We’ll have to see whether Kansas City’s magic runs out.

hjenten: And all it takes is one play in the World Series.

neil_paine: OK, so let’s bring it all the way back for the big picture on the Series itself. What are your predictions?

rob: I will say Royals in seven, but let’s be realistic: A 55 percent favorite (as Elo estimates) is barely a favorite at all. The outcome is almost certainly within the margin of error of any publicly available forecasting tool.

So it will not surprise me whichever way this goes. And after this postseason, I am just excited to see some more weird baseball.

hjenten: I also want to say the Royals in seven. But to be different, I will say Mets in five.

neil_paine: OK, you heard it here first — Royals in seven … maybe.

Thanks for chatting, guys, and we’ll be back later in the World Series. See you then!

No matter what happens at the end of this year’s World Series, a curse is getting broken. On Tuesday, the Chicago Cubs will open up proceedings against the Cleveland Indians, and by Nov. 2, one of them will have claimed the title. For two teams with the some of the worst championship luck in all of baseball, this series will offer long-awaited catharsis for one — and even more misery for the other.

No team is ever a shoo-in to win a championship. But all else being equal, great teams should claim the title more often than merely good ones. To confirm that hypothesis, I looked at every team’s end-of-season Elo rating (a measure of team strength) and whether that team won the World Series, in every postseason era — from when two pennant-winning teams went straight from the regular season to the World Series, to the modern 10-team field that battles through multiple playoff rounds.

As expected, a team’s probability of winning the World Series increases as its roster gets stronger. But the effect of having a good team is completely overwhelmed by the number of opponents a squad must face to get to the World Series. Consider the worst and best teams to have ever made the playoffs by Elo’s reckoning: the 2005 San Diego Padres (with a rating of 1489) and the 1906 Chicago Cubs (1635). In a two-team playoff system — which is what the 1906 Cubs faced — that Padres team would be expected to win the World Series only 23 percent of the time, while the Cubs had a commanding 72 percent chance of taking home the title. In the current 10-team playoff system, the Padres’ odds would shift down to 4 percent — but the 1906 Cubs would drop much more dramatically, all the way down to 28 percent. That’s why the modern playoffs are a crapshoot: No matter how good a team is, by the time it gets to the postseason, its chance at the championship isn’t radically better than that of the worst team in the playoffs.

As a corollary, any ballclub that appears in the postseason often enough — no matter how mediocre its teams are — should eventually be guaranteed a World Series win. But for more than a century’s worth of Cubs squads, no level of greatness has been able to get them over the hump. I determined just how unlucky each franchise has been over its postseason history by taking its Elo rating and the size of the playoff field and then calculating how likely the team was to win the Series each year using the process I outlined above. I added up those probabilities from all the years in which a World Series was held and compared them with how many titles the teams actually won, and I found that the Cubs are the unluckiest team of the last 113 years.

Just based on the pretty-good teams the Cubs have featured in their 18 playoff appearances, my model expected them to win six or seven championships. Instead, they’ve only won two since 1903, and both were more than 100 years ago. (Notably, this year’s Cubs triumphed in the NLCS over the second-most-unlucky team, the Los Angeles Dodgers.) The Cubs have had more years to be unlucky than most teams, since they’ve existed for a long time. But even on a per-playoff-season basis, the Cubs have been the least fortunate franchise in baseball.

But you already knew the Cubs were unlucky. The misfortunes of the Cleveland Indians, on the other hand, had attracted far less ink before this season, despite their status as the eighth-most-unlucky franchise in the same time frame. The Indians don’t have quite the cursed reputation of the Cubs, and that’s fair. But when you consider that they’ve also been the seventh-most-unlucky team in terms of converting regular-season wins into playoff appearances since 1998, it’s easier to believe that Cleveland is hexed. (By contrast, the Cubs have been the second-luckiest team at getting into the playoffs in that same period.)

One of these franchises will see its championship drought end soon. If the baseball gods have any mercy, they’ll reward the Cubs for assembling what will probably go down as one of the best squads of all time. Yet, as a Cubs fan myself, I’ve been trained to believe that seasons can only end in heartbreak. But whether or not it ends this year, statistically speaking, the curse can’t persist forever. Probably.

VIDEO: Cleveland fooled us twice

In preparation for the World Series, which starts Tuesday night, we invited ESPN MLB writer/editor Christina Kahrl and our own baseball columnist, Rob Arthur, into Slack to chat about the Chicago Cubs and Cleveland Indians. As usual, the transcript below has been lightly edited.

neil (Neil Paine, baseball editor and sportswriter): Well, we’re finally down to two teams, the Cubs and the Indians, both of whom have long championship droughts on the line. So my first question for the room is just a big-picture one: How did these teams stack up in the overall sabermetric numbers during the regular season?

rob (Rob Arthur, baseball columnist): Both teams were good, but the Cubs were also great, fantastic, amazing and 10 other superlatives on top of that. In the first half of the season, they had as good a run differential as any team ever — right up there with the 1927 Yankees. They were merely dominant in the second half, but in either half Chicago was better than Cleveland: the Cubs had a +91 run differential in the second half alone, which is only 10 less than the Indians racked up all season. And, remarkably, some measures (such as cluster luck) suggest this Cubs team got unlucky.

Which is not to say that the Indians were a bad regular-season team — they had the fourth-best run differential in baseball. But they also probably got a little fortunate from a cluster luck perspective, and their pitching, while solid, was also weakened due to injuries by October. So this matchup is probably a bit lopsided in favor of the Cubs, at least if we go by regular-season numbers.

christina (Christina Kahrl, ESPN.com baseball writer and editor): I would think whatever metric you used, you’re going to get happy answers about the Cubs and Indians that don’t involve a stack of head-scratchy one-run outcomes or players having extraordinary seasons outside their expected range of performance. (Well, except maybe Tyler Naquin’s strikeout rate.) But across 162, these were two very good teams. Outside of the Cubs’ sporadic offensive disappearances, we’ve seen two of the best regular-season teams also play well in October. If not for injuries to the Indians’ rotation, you could have seen that these two teams belonged here months ago.

There are interesting distinctions, of course. The Cubs and Indians both walk plenty, but the Indians aren’t in quite the same class when it comes to power production. But they’re both very balanced offenses, with good amounts of contact (call it BABIP or just execution on balls in play), power, patience and speed. The fun gets into the differences between how Cleveland manager Terry Francona used his bullpen to compensate when the rotation melted down, and how the Cubs churned through relief combinations before trading for Aroldis Chapman at the deadline. To some extent, both teams are where they are because of how well their answers worked out.

neil: So we’re not seeing fluky teams! These two teams might legitimately be some of the very best in baseball! Seems like a departure from recent World Series history.

christina: And yet — maybe it’s because I’m in Chicago — because of those injuries in the Indians’ rotation, folks are already anticipating a walkover. The last 15 years or so should perhaps suggest a little less overconfidence on this score. I can’t help but think of the 2006 or 2011 Cardinals as notable examples of underdog winners.

rob: Right, and given that it’s only seven games, anything can happen.

neil: Yeah, I was gonna ask because Rob mentioned that it was “a bit lopsided” — in baseball, that still doesn’t really mean either team is very likely to win over the other. At most maybe it’s 60-40, or 65-35, for the favorite?

christina: Well, the Cubs should be favored, for all sorts of reasons about how awesome they are (not just because the Indians’ rotation is a shambles). And I think you’re right in terms of how far that lean should be. But I also remember “October unbeatable” Jon Lester losing a must-win game in 2014, so I tend not to believe in absolutes.

rob: Yeah, and interestingly, everything from betting markets to our Elo ratings to FanGraphs’ simulations puts the probability between 60 to 70 percent for the Cubs. So that speaks again to the randomness of baseball — I think it would be hard to argue that the Cubs aren’t better than the Indians, but despite that edge they only have about a 2-in-3 chance.

christina: To put it another way, this series doesn’t feel like the 1998 World Series, where there was almost no reason to watch unless you were a Yankees fan.

neil: Hey! Those Padres had a pretty good seas… — ah, I can’t finish that thought. It was a rout. But this one, less so, it sounds like.

Now, have we seen anything during the playoffs to make us think either team is better or worse than the yearlong numbers would indicate?

rob: Yes, I think it’s fair to say that the Cleveland bullpen — and Francona’s clever use of it — gives the Indians a strong advantage that isn’t reflected in their regular-season numbers. The Cubs don’t really have anything comparable to that; although their bullpen is strong, Chapman doesn’t seem comfortable outside of the eighth or ninth innings. (Even then, he’s looked shaky at times.) I don’t think we can say with much confidence how much exactly fireman Andrew Miller is worth, in terms of series win probability. But I think he probably keeps things to closer to 60-40 than 70-30, as some outlier predictions would put it.

christina: I do wonder how well the Cubs will do if the Indians get to their ’pen in the fifth, sixth or seventh innings. The Indians’ lineup has many strengths — it’s front-loaded with Carlos Santana leading off, it’s deep, and Francona isn’t afraid to use his bench. So in those middle-inning matchups, especially during games with the DH, I wouldn’t bet on Joe Maddon securing advantages as easily as he does against some NL opponents. A lot depends on whether the Indians get to the Cubs’ starters early — running up pitch counts, making them work from the stretch — and then forcing the game into the hands of relievers like Justin Grimm or Carl Edwards.

neil: Speaking of the managers, this seems like it’s going to be a battle of two extremely smart, saber-savvy tacticians — perhaps the likes of which we’ve never seen before.

christina: Well, let’s be fair, Howser vs. Herzog in 1985 was pretty awesome.

neil: If you wanted Whiteyball, you got it with last year’s Royals. This year — well, it’s not exactly Moneyball that these two teams play, but maybe something in the same tradition at least.

christina: But to your point, yes, it’s going to be a very interesting series in that regard, watching a couple of brilliant skippers with histories of putting players in a position to succeed. For those folks who say “managers don’t matter,” here are two great tacticians who are also extremely smart about how to manage people across six months, and who get the difference between managing the regular season and managing in October.

rob: Yes, although Maddon’s strength seems to lie in the parts of baseball that still aren’t visible to us: chemistry, the clubhouse and getting the best performances out of players. Francona is probably good at that, too, but bullpen management is a visible manifestation of his skill, whereas the best we can do to quantify Maddon’s ability is look at how his teams consistently have positive run differentials.

christina: Yeah, I wouldn’t put either over the other as far as people management. “Tito” and Maddon both deserve their reputations.

neil: So, aside from the battle of managerial wits and the two bullpens, what else will you be keeping an eye on as key matchups in the series?

rob: Christina mentioned above that Lester’s been incredible in the playoffs. That’s true — he’s Bumgarner-esque — but he has a critical weakness: the yips that prevent him from throwing over to first. In theory, that should make it easy to steal bases on him, but opponents have been curiously reluctant to exploit Lester’s flaw. The Dodgers tried — and failed — to do so, largely by dancing around between first and second, and Lester turned in another awesome start. But I do wonder if Francona’s tactical savvy can translate into more stolen bases and potentially weaken the Cubs’ best starter.

neil: Do the Indians have base runners who might especially be able to take advantage of something like that?

rob: The Indians had the third-best baserunning team in the majors, according to FanGraphs’ metrics. The Dodgers were 11th, although they had some good base stealers who just failed to convert. Jeff Sullivan posited that it’s a mental block for potential base stealers, as they are so unused to getting leads of 25 feet (or more!) that they don’t know what to do with them. That’s why I think it will mostly be a matter of Francona getting the base runners to actually take off, and not the skill of the base runners themselves. Almost any major leaguer should be able to get to second base before the throw when they have a 35-foot lead, as some of the Dodgers’ baserunners did:

christina: We should also remember that base stealers were 23 for 26 against Jake Arrieta, as well, so this isn’t just a Lester problem. I can see arguments that Willson Contreras might help control the damage in games that don’t feature the Lester-David Ross battery, but we’ll see.

neil: Sounds like we shouldn’t be surprised if Cleveland’s baserunning makes life difficult all series for what is otherwise a scary good Chicago rotation.

christina: They’ll need to try, because they only thing that’s going to take that Cubs’ defense down a notch is the friction multiple baserunners and men in motion can create. Play a static, big-inning offense where you wait around for hits, and the Cubs will find ways to kill your scoring opps. Russell-to-Baez-to-Rizzo is going to merit its own poetry.

rob: The defensive skill of the Cubs infield is a major factor that stops potential base runners. It’s all too easy to get caught in a TOOTBLAN* with Javy Baez’s creativity on one side of second base and Addison Russell’s sure hands on the other. In that way, it will be strength against strength.

(* Ed. note: That’s “Thrown Out On The Basepaths Like A Nincompoop,” for the uninitiated.)

christina: I’m also wondering which Arrieta or Kyle Hendricks we get. That could shape the series. Take Hendricks: The Indians are the best team in baseball at killing pitches 90 mph or slower. They’re third in baseball in OPS against off-speed pitches. If anyone is going to get to Hendricks in his magical year, it might be the Indians.

rob: I agree that Hendricks and Arrieta are less sure bets. Generally, a major strength of all of the Cubs pitchers is that they suppress batted ball velocity. I believe that’s a genuine skill that the Chicago rotation possesses, but it also seems like a skill that’s more variable than say, throwing 98 mph fastballs that your opponents can’t catch up to. So I wouldn’t be terribly surprised if the Cubs have a couple of disastrous starting pitching outings and get BABIP’d to death.

neil: All right, let’s close this out with some official predictions. Who ya got, and in how many games?

rob: I’ll take the Cubs in 6. They are the better team, and one thing we only briefly alluded to is how tired and tattered Cleveland’s rotation is. I think the Cubs will dampen Cleveland’s bullpen advantage by overworking them, and that will be enough to close the Indians out. But not easily.

christina: It’s really tough, because while Cubs in 5 is probably the safest choice, there are so many things that could go wrong with that (or even just extend the series) that I’m sticking with my prediction over on ESPN.com: that the Indians find a way to win in 7. Because, how safe are the safe bets? But I’ll admit, there’s also an element of my wanting this to be an epic series, to give us something to remember beyond one of these two teams’ “curses” ending.

neil: Indians in 7? Christina, I knew you were a Chicagoan, but now I see you either are not a Cubs fan, or the most quintessential Cubs fan possible.

christina: Hah. Funnily enough, people mistake me for a White Sox fan, but I’m agnostic. (I’ve stuck with the team of my childhood, the A’s — hence my bitterness about Mr. Lester in 2014.) When I polled Chicagoans last week on Twitter, the second-largest group beyond the 39 percent of Chicagoans who call themselves Cubs fans who think they’ll win it all was the 31 percent who said they’re Sox fans who hope they blow it.

Besides, if the Cubs win, I can claim I didn’t jinx it, right?

neil: Very true, you are zigging where those not-so-covert Cubs fans we saw everywhere on Saturday night are zagging.

christina: I did the double-reverse, anti-curse, non-jinx prediction. Shazam!

neil: Well, I’ll split the difference and say Cubs in 7. That feels like the way this season is, and always has been, destined to end — though as we know, sometimes real baseball gets in the way of destiny, narratives and whatnot.

Either way, though, it looks like one of the more entertaining on-paper World Series in recent memory. I can’t wait!

VIDEO: Cleveland fooled us twice

You’re reading Back of the Envelope, an experiment that aims to bring shorter, quicker content to FiveThirtyEight.

All postseason, FiveThirtyEight’s MLB projections have had a lot of doubts about the Cleveland Indians — our forecast didn’t even think they would make it out of the division series. In the video above, Neil Paine explores the chances that Cleveland’s expectation-defying streak will continue in the World Series against the mighty Cubs.

Subscribe: Apple Podcasts | ESPN App | Download | RSS | New to podcasts?

Welcome to the latest episode of Hot Takedown, FiveThirtyEight’s sports podcast. On this week’s show (Oct. 25, 2016), we preview the World Series. First, we take a look at the Cleveland Indians and ask if credit for their success belongs only to their bullpen. Then, we chat about the Cubs and wonder if they are uniquely built for postseason baseball. We also ask Cubs fan Rob Arthur if he thinks the Cubs’ rotten luck is about to come to an end. Then, it’s NBA time again. We break down FiveThirtyEight’s CARMELO projections for the season and debate whether the Golden State Warriors can live up to their historically high expectations. And to close out the show, we share a significant digit about the WNBA finals.

Also, remember to check the Hot Takedown feed on Thursday for the third installment of our documentary series Ahead Of Their Time. The series looks at coaches and players who did something radical for their era and were later proven right by analytics. The third episode, coming on Thursday, is about the stathead who helped ruin English soccer. You can find the previous two episodes here. Links to what we discuss:

• Rob Arthur says even the data thinks the Cubs may have been cursed.
• Neil Paine writes that all it took for the Cubs to win the pennant was great hitting and pitching.
• Ben Lindbergh at The Ringer thinks the only way to beat Andrew Miller is to avoid him.
• Neil Paine writes that superteams like the Warriors usually underperform their numbers.
• Nate Silver explains what’s new in FiveThirtyEight’s NBA projections this year.
• You can find our CARMELO NBA projections here.
• ESPN’s Kevin Pelton gives us the reasons that we won’t see a Warriors-Cavs rematch this year.
• Significant Digit: 528,000. That was the average viewership figure for Game 5 of the WNBA Finals. More people actually watched Game 1 than the deciding game.

You’re reading Back of the Envelope, an experiment that aims to bring shorter, quicker content to FiveThirtyEight.

Do I want — really, actually, genuinely want — the Cubs to win the World Series? It’s a question I grappled with at length after a couple whiskeys with my best friend from back home as they marched toward the pennant. I love the Cubs. I’ve always loved the Cubs. There is little else (some humans excluded) I love more than the Cubs. So what’s my problem, anyway?

The Awl addressed the paradox with a cautionary tale from a Red Sox fan: “Without all the losing, the Red Sox are now just another pretty good team. The aura of mythology that swirled constantly around them was gone.” And that’s a big part of it. Losing is what the Cubs are known for. But we bleed-blue fans don’t root for lovable losers for the sake of it. We don’t do it because we’re proud of the title drought or miserable seasons. We don’t happily bask in the black-magic glow of curses or take masochistic joy in the outstretched arms of Bartmans.

We root for them because eventually they will win, they just have to! It’s math! Curses aren’t real! And when they do win … my God, it will be ecstatic bliss. Angels will sing. Fathers, sons, mothers, daughters will sob tears of joy. Bill Murray will be there. Etc.

The question is when do we want to cue the messianic chorus. Maybe Cubbie fandom is something like holding an American option — a piece of paper we Cubs fans carry around in our pocket and turn in at the bliss counter one day when they Win the World Series. In exchange we receive some number of singing angels. The longer this takes — the more heartache we bank — the more angels. That’s just math, people. If the Cubs had won it when I was, say, 3 years old, the rest of my fandom wouldn’t have been imbued with a greater, Sisyphean meaning. So, on my deathbed, yes, I should want nothing else than to cash in. If I’m watching from my high chair, I’d probably like to wait a while. To bide my time until 8:08 p.m. Eastern on Wednesday night, I hastily sketched a chart of how my 31-year-old self should weigh these priorities, complete with an “exercise boundary” that tells me when I should want the Cubs to win so I can cash in my pain for those angels.

So what does my analysis reveal? Will I be rooting for the Cubbies? Of course. I just can’t not. In late October, math no longer applies.

You’re reading Back of the Envelope, an experiment that aims to bring shorter, quicker content to FiveThirtyEight.

When is a loss just a loss, and when is it a harbinger of things to come? On Tuesday night, the Chicago Cubs and the Golden State Warriors took hard losses in very different settings: Game 1 of the World Series (a 6-0 Cleveland Indians win) vs. Game 1 of the NBA season (a 129-100 drubbing by the Spurs). Weirdly, FiveThirtyEight’s Elo projections say each game had a similar effect on the Cubs’ and Warriors’ chances of winning a championship:

The stakes didn’t match at all, but the fallout did. That’s uncommon for two teams at opposite ends of their season but makes sense when you dig into why. Both of these swings reflect when these games were played: For the Cubs, the World Series is essentially a seven-game season, so every game will shift championship odds quite a bit. The Warriors, meanwhile, had a preseason championship probability that rested entirely on team rating (since the team didn’t have a record). That meant the odds could take a big swing early in the season.

The road back to being more than 50 percent likely to win the championship is quite different for the two teams. All the Cubs need to do is win Game 2, and their deeper roster will seduce the model again. The Warriors have further to go. While 38 percent is still an absurdly high probability to win the title one game into the season, they’ll need to break off an impressive run before they can convince the model, and the league, that they’re the overwhelming favorites they once appeared to be.

CHICAGO — The Cleveland Indians won a 1-0 nail-biter on Friday night in Game 3 of the World Series. Two number-three starters succeeded in shutting down two strong offenses, allowing the game to come down to the final at-bat. But while the relievers were overpowering as usual, the most significant influence on this game wasn’t the wind, a single Indians hitter or managerial cleverness, but a seemingly inconsistent strike zone.

Home plate umpire John Hirschbeck has a reputation for calling balls and strikes erratically, and that was on full display last night, creating shifting strike-zone boundaries that bedeviled both offenses. For the Indians, Josh Tomlin turned in an unexpectedly solid line, allowing only two hits. At times, Tomlin was burned by bad calls, leading, for example, to a fourth-inning walk by Kris Bryant. But when the strike zone is called inconsistently, hitters tend to strike out more often and make weaker contact. That’s because pitchers can choose to target inconsistently called areas of the zone when it benefits them, while hitters can only decide whether to swing or not at what’s offered. When they’re uncertain, batters often opt to swing at pitches outside the zone, resulting in glancing contact and easy outs.

Chicago Cubs starter Kyle Hendricks, who usually gets favorable strike calls due to his impeccable command, struggled mightily in allowing six hits and two walks in only 4.2 innings. The shifting zone did aid him in racking up six strikeouts, above what you’d expect based on his regular-season stats.

Even as the inconsistent strike zone helped the pitchers, neither was overpowering. And with bullpens fresh after the day off, both starters were pulled before the 6th inning with the score 0-0, an event that has never happened before in MLB postseason history. That handed the game to the relievers, including an early appearance from Andrew Miller. They were as commanding as expected, except for one lapse by the Cubs’ Carl Edwards Jr., who allowed Coco Crisp to single in the lone run of the night.

The Cubs came close to evening the score in the bottom of the ninth. With two runners in scoring position and two outs, Chicago dynamo Javy Báez was up to bat against Cleveland closer Cody Allen. He struck out whiffing to end the threat, leaving the Indians up 2-1 in the Series.

The outlook for the Cubs is worrisome going forward: Their series win probability by Elo is down to only 37 percent. In his last start, Corey Kluber looked invincible, and the Cubs will have to face him in Games 4 and 7 of this Series (if it goes that far). That means they will need to pull off at least one upset against the 2014 AL Cy Young winner to clinch the series. While such a feat appears difficult, the Cubs managed an even more surprising performance against Clayton Kershaw in the NLCS, so it’s certainly possible. Nobody said ending a 108-year title drought would be easy.

CORRECTION (Oct. 29, 12:05 p.m.): An earlier version of this article incorrectly described Corey Kluber. He was the 2014 AL Cy Young winner; he is not the reigning winner.

CHICAGO — With a 2-1 World Series deficit and home-field advantage slipping away, the Cubs needed this game. Instead, the Indians soundly defeated the Cubs 7-2 on Saturday night in Game 4, silencing the normally raucous Wrigley crowd and drastically decreasing Chicago’s chance of taking home the championship. In a World Series marked by low scores, Cleveland has shut down the Cubs’ bats more than ever this year.

Pitching has defined this World Series. The average number of runs per game so far has been only 5.5, which is tied with 2015 and 2011 for the lowest total since 1983. Offense is down even more when you take into account the higher regular-season scoring in 2016: This year has seen the largest gap between World Series runs scored and the regular-season average since 1966.

Paradoxically, neither team’s pitchers have been altogether overpowering. In Game 3, an inconsistent strike zone kept both teams from plating many runs. On Saturday, the Indians batters managed to capitalize on mistakes while Corey Kluber kept the Cubs quiet. Kluber’s final line (6 IP, 1 ER) is somewhat deceptive: Throwing on only three days’ rest, his stuff seemed to lack the crispness and velocity that usually characterizes one of the best pitchers in the American League.

But Kluber’s 81 pitches went through the sixth inning, enough to hand the game over to the invincible Cleveland relievers. Outside of a solo home run allowed to Dexter Fowler, the Indians bullpen stopped any further scoring. Between the shaky starters and overpowering relievers, the Indians have totally controlled the Cubs offense. The four World Series games so far have seen the Cubs score only 7 runs, which is a lower total than they’ve racked up in any four consecutive games in the 2016 regular season.

Part of the problem was bad luck and sloppiness on the part of the Cubs. That included two errors by Kris Bryant, the Cubs’ normally sure-handed third baseman. Another problem was a gusting wind that turned at least one probable homer into a double. But credit must be given where due: The Indians are executing their gameplan to perfection, getting small but reliable leads and then deploying their absurd bullpen to maintain them.

It will be hard for the Cubs to come back from this 3-1 deficit. As the Cavaliers taught us earlier this year, a 3-1 lead isn’t insurmountable, but Elo rates the Cubs’ the total chance of winning the Series at a measly 15 percent. (That’s a smaller chance than FiveThirtyEight’s election forecast model currently gives Donald Trump to win the White House.) But if Chicago is going to have any chance of a Series win, they’ll have to awaken their bats in Sunday’s game.

Subscribe: Apple Podcasts | ESPN App | Download | RSS | New to podcasts?

It only took 108 years, but our sports podcast Hot Takedown finally got to discuss a Chicago Cubs World Series win. We break down the strange managerial decisions in Game 7, discuss whether this year’s Cubs or the 2004 Boston Red Sox are Theo Epstein’s crowning achievement, and FiveThirtyEight’s Oliver Roeder joins us to discuss the legacy of this year’s Cubs team.

Twenty-five years ago, millions of baseball fans around the country turned on their televisions expecting to watch a World Series game — and saw live footage of a deadly earthquake instead. The San Francisco Giants and the Oakland A’s, and the 62,000 fans watching them in Candlestick Park in San Francisco, felt the ground under them shake. The baseball commissioner thought it was a jet flying overhead. Oakland’s manager thought the crowd was stomping its feet. Then a section of the right-field stands separated in two by a few inches. Players ran to gather their family members and get out of the ballpark.

The earthquake killed 63 people — and might have killed more if there was no game on TV to keep people off area roads — and the series didn’t resume for 10 days. (The earthquake is the subject of the “30 for 30” film “The Day The Series Stopped,” airing Tuesday night on ESPN.)

https://www.youtube.com/watch?v=ORD9ICL4PsE

There have been deadlier earthquakes, and costlier ones, but few that surprised more people the moment they occurred. The millions of television viewers didn’t get what they expected because scientists couldn’t predict when an earthquake would strike.

The Loma Prieta earthquake helped fuel efforts to change that, but not much progress has been made. A decade after the quake, Robert J. Geller, professor of earth and planetary science at the University of Tokyo’s Graduate School of Science, wrote that earthquake prediction “seems to be the alchemy of our times.” Seismologists have mostly forsaken their quest for precise predictions, turning instead to more modest projects like telling the public when the probability of an earthquake has risen to 1 percent from 0.01 percent. They can’t predict whether or when they’ll be able to do any better.

Today, earthquake scientists in the United States and several other countries are working on producing “seismic weather reports” — a phrase Thomas H. Jordan, director of the Southern California Earthquake Center, uses to describe a continuously updated, local estimate of the probability of an earthquake. Just as you can look up the probability of rain in your area, the seismic forecast would let you look up the probability of an earthquake — but it wouldn’t be nearly as accurate.

These forecasts won’t tell people precisely where and when an earthquake will strike, or what magnitude it will be. Instead, the forecasts will show whether the baseline probability of an earthquake has risen.

But even that relatively modest endeavor faces many challenges, including computing power, communicating risk to the public, and swaying skeptics within seismology. “Crying earthquake (wolf) is a potent way of blunting earthquake awareness and preparedness,” Kelin Wang and Garry Rogers wrote in the journal Seismological Research Letters earlier this year. Wang, a research scientist at the Geological Survey of Canada’s Pacific Geoscience Centre, thinks that earthquake forecasting is a promising area of research, but the trick is translating those forecasts into something that won’t be counterproductive when it reaches the public.

Jordan is less concerned about spooking the public. In a response to Wang, Rogers, and other critics in Seismological Research Letters last month, Jordan said that Americans have experience processing low probabilities for catastrophic events. We’ve grown accustomed to hearing about heightened awareness of terrorist attacks, and wildfire warnings have become a feature of Californian life.

Ned Field, a seismologist for the U.S. Geological Survey, thinks there’s an audience for short-term forecasts. Prospective buyers of earthquake insurance, homeowners considering whether to leave town or homeowners wondering whether to build a basement could all benefit. (The USGS is already using an app to test how to spread the word in case of elevated earthquake risk in southern California.) He envisions marrying short-term forecasts with the agency’s model that estimates an earthquake’s economic costs and fatalities. That would give the public a sense not only of how likely an earthquake is, but also how much damage it could create if it occurs.

If they work, the sorts of forecasts Jordan and the USGS have in mind must avoid being overly definite, but they can’t be too vague, either. As my colleague Nate Silver wrote in his book “The Signal and the Noise,” just before a deadly Italian earthquake in 2009, scientific technician Giampaolo Giuliani spoke as if earthquakes could be predicted. He said an earthquake was coming, and based his prediction on an unproven technique of measuring radon-gas emissions. Meanwhile, more reputable earthquake scientists spoke from the other extreme, taking the view that earthquakes were no more or less likely at any given time. Both were wrong: Giuliani’s forecast missed the time, place and magnitude of the tremors that killed 309 people in L’Aquila. But the scientists were wrong, too. They discounted the significance of small earthquakes in the area, which in retrospect were foreshocks of the bigger earthquake.

The middle ground is the one the USGS is seeking: to tell the public when seismologists know there’s an elevated risk, without overstating their confidence in the prediction.⁸

The agency has tried this before. In 2005, it introduced on its website a tool that allowed people to check the chances of an earthquake in their area. But the code powering the tool kept crashing, Field said, and the USGS removed it from its website in 2010.

The models used by Jordan and other scientists today analyze recent seismic activity to predict the probability of future earthquakes. Jordan and Field are focusing on a model derived from something called ETAS, or Epidemic-Type Aftershock Sequence Model, which projects the proliferation of tremors in the way a disease might spread.⁹

There are many other models besides ETAS. Just how many depends on how you count. Jordan estimates there are 400 models worldwide vying for pre-eminence and being tested by the Collaboratory for the Study of Earthquake Predictability, which he directs. Field says USGS alone is considering 5,000 models. One model can differ from another merely in how it represents the structure of the earth’s crust.

In theory, all these models should be competing in a test of which best forecasts seismic activity. In reality, there just aren’t that many high-magnitude earthquakes in the world that can serve as tests. That’s good news for anyone living on a fault line, but not for seismologists, who would like to know whether their models are correctly calibrated to pick up the greater risk just before a major earthquake. So instead the tests generally focus on whether the models correctly predict the frequency of lower-magnitude earthquakes. Globally, there is a reliable relationship between the rate at which these occur and the rate at which major earthquakes occur.¹⁰ But in any given spot, that relationship may not apply — and not all models can be tested globally because not all places have the level of measurement of crust structure and seismic activity that, say, California does.¹¹ Also, a model tuned to pick up small quakes may not pick up bigger ones.

With so many competing models, there is another risk: The one that does best in tests might just be getting lucky.

Even the best model wouldn’t predict about 50 percent of the big earthquakes around the world. The quakes that go undetected would be the ones that seemingly come out of nowhere, without foreshocks. For instance, it’s unlikely the best model would have warned residents of Napa County of the greater risk of an earthquake before a big one struck this past August, killing one person and causing an estimated $1 billion in damages. And it’s hard to say if any model could have foreseen the Loma Prieta earthquake, according to Jordan, because not all the data the model uses to produce forecasts was available in 1989.

Jordan calls the hunt for a more precise earthquake prediction “a silver-bullet approach,” trying to find “some magic signal.” What he and his collaborators — there are more than 50 — are doing today “is very different … There’s nothing magic about it,” he said.

One prediction the forecasters are comfortable making is that we won’t get more definite predictions anytime soon — if ever. “I would not be at all surprised if earthquakes are just practically, inherently unpredictable,” Field said. “You never know; some silver bullet could come along and prove useful.”

In less than a week, you may have heard, there’s a midterm election in the United States of America. This is sort of a big deal for us at FiveThirtyEight. Such a big deal that our estimable tech team of Jeremy Weinrib and Paul Schreiber arranged a fancy live-blogging platform so you can snuggle up next to us for hours on election night. It’ll be cozy.

We’ve known for weeks that we’d need to give the platform a test drive, and we decided that we’d do that Wednesday, on the second night of the NBA season. We’d get together our crew of basketball writers (the ones who wrote our NBA team previews), buy some pizzas and use an algorithm to project whether Giannis Antetokounmpo has finally stopped growing.

But as the San Francisco Giants discovered last night, Jake Peavy has a habit of ruining the best-laid plans.

About the time Game 6 of the World Series passed a 95 percent win probability, we made the call to scuttle the NBA live blog. Instead, you’ll get to hang with us as we watch Game 7. We’ll argue that Jeremy Guthrie shouldn’t pitch more than three innings, locate where the Giants dynasty of the past five seasons would rank compared to others and, Yost-willing, debate the merits of the sacrifice bunt.

It’s going to be great. Or a total disaster. Come and find out which. 8 p.m. EDT Wednesday. Here on FiveThirtyEight.

Welcome to FiveThirtyEight’s live blog of Game 7 of the 2014 World Series. We’re doing some testing of this format, so please feel free to send us feedback.

The Kansas City Royals just concluded the most improbable, exhilarating, agonizing and — ultimately — meaningful season in the franchise’s past three decades of existence. That they lost the World Series to the inexplicably dynastic San Francisco Giants in a memorable Game 7 that literally came down to the final out doesn’t soften the blow for Royals fans, but it does put in perspective what Kansas City accomplished this season: they essentially came as close as any team possibly could to winning the World Series without actually hoisting a World Series trophy.

(As Grantland’s Royals-fan-in-residence Rany Jazayerli put it: “[Kansas City] went 11-4 in the postseason. That’s not only the best postseason record any team has managed without winning the World Series — it’s the best possible record a team can have under the current format without winning the World Series.”)

And yet, as the Giants’ celebration was unfolding under the Kansas City sky Wednesday night, there was the distinct sense that the clock had also struck midnight on the Royals’ 2014 cinderella story. Kansas City had not won their division, narrowly finishing first in the American League wild card standings over two teams (the Oakland Athletics and the Seattle Mariners) who’d posted vastly superior run differentials during the regular season. Going into the playoffs, the probability of a team like Kansas City, with an 84-78 pythagorean record, getting through the AL’s postseason minefield to the World Series was just 6.3 percent.

It took a lot of good fortune for the Royals to even make it as far as they did. But can they make it back?

To (roughly) answer that question, I built a simple model predicting the probability of a World Series contestant making another World Series appearance at any point in the five seasons after its initial showing. (The ingredients for it can be seen at the bottom of this post.) That model says the Royals have just a 30 percent probability of going back to the World Series at any point in the next five seasons (the average World Series participant returns about 46 percent of the time).

Coincidentally, that 30 percent probability means Kansas City currently has roughly the same chances of a World Series return as the model assigned to the 1985 Royals (which had a similar age and weighted pythagorean winning percentage as the 2014 Royals). That version of the team, as Kansas City knows well, never made it back to the World Series. Other similar teams include the 2007 Colorado Rockies and 1984 San Diego Padres, both of which drew easy comparisons during the Royals’ October run — and both of which were World Series one-hit wonders. Of the 10 historical World Series contestants most similar to this year’s Royals, only one — the 1992 Atlanta Braves — ever found their way to another World Series within the next five seasons.

Then again, the model isn’t too much more optimistic about the Giants team that beat the Royals, assigning San Francisco a 34 percent probability of ever going back to the World Series with their current group. Yet the Giants have already bucked similar odds twice — they returned after the 2010 edition of the team appeared to have a 33 percent chance of doing so, and made another repeat trip after the 2012 edition was assigned a 37 percent probability of going back to the World Series. For their part, the Royals have the eighth-best farm system in MLB according to Baseball America, and have had top-10 prospect classes in three of the last four such lists issued by Baseball America. The talent is ostensibly there to keep the Royals’ 2014 run from being a total fluke.

The truth is that most World Series entrants fail to return within the next few years. It’s hard enough to make it to one World Series, let alone two in the span of five seasons. MLB’s current playoff structure ensures a high degree of randomness, with mechanisms in place to prevent the best teams from running the table. The Royals are most likely to fade away like so many other World Series teams have throughout baseball history.

***

For the model readers out there: I used data since MLB expanded its playoffs to include four teams in 1969, fitting a logistic regression model that used an average of the team’s previous five seasons’ worth of pythagorean winning percentages (weighted 5-4-3-2-1, from most recent to least recent, to give more influence to more relevant seasons) and the average of the team’s mean ages for its batters and pitchers. (I also considered additional variables, such as the market size of the team’s home city and whether or not the team actually won the World Series in question, plus dummy variables for MLB’s free agency and wild card eras, but none of those proved statistically significant.)

While Kansas City’s players were younger than the typical World Series team — their mean age was 28.3, compared to 29 for the average World Series team — their weighted pythagorean winning percentage (.497) was league-average at best over the previous five seasons. The pythagorean-record factor has historically been 50 percent more important than a team’s average age, and the Royals had the eighth-worst such mark of any World Series participant since 1969.

Stocked with a cast of insanely talented young players, Tuesday’s MLB All-Star Game in Cincinnati was a perfectly pleasant affair, and a nice showcase for the current (and future) state of the sport. But it wouldn’t be an All-Star Game if it weren’t also noted that the outcome of a silly exhibition contest continues to be used to determine home-field advantage in the World Series.

That’s pretty dumb, and there’s no shortage of people in the game who’d like to see it changed. (Except, apparently, new MLB Commissioner Rob Manfred.) There are also plenty of alternative suggestions for what should determine World Series home-field instead. But as dumb as it is for the All-Star Game to “count,” what really matters is the effect the rule has had on the 12 Fall Classics since it was put into place.

To quantify the ramifications of the policy, I calculated pre-series win probabilities for all World Series teams from 2003 to 2014 using their regular-season pythagorean records.⁷ I looked at how much those odds shifted depending on whether the frivolous, All-Star-based home-field rule was used, or two alternatives: the equally arbitrary (but at least consistent) pre-2003 policy of alternating home-field between leagues each season, and a simple rule that bestowed home-field upon the team with the superior regular-season record.

Going from the current format to an old-school approach that alternates home-field by league every year, the pre-series odds would have shifted by an average of +/- 1.4 percentage points each year over the past 12 seasons. Home-field itself would have been different seven times in this alternate universe, though it probably wouldn’t have made much difference to the outcomes of most series. Twice in 12 years (2005 and 2013), the margin between the teams was slim enough that swapping home-field would have changed which team was favored.

The team with home-field in reality won each of those series, so the argument could be made that linking home-field advantage to the All-Star Game swung a pair of championships. But each of those series was also nearly 50-50 regardless of who held home-field, so you’d expect a different set of results in those series (compared with reality) only about 75 percent of the time if they were replayed. In total, if we re-simulated the last 12 World Series in this universe, we’d see about 6.2 different winners over that span, but only 1.7 of them could be directly traced to ditching the All-Star Game-based home-field format. Per decade, that works out to about 1.4 titles exchanging hands because of the change in format.

That number gets smaller if we compare the actual results since 2003 to another hypothetical universe in which home-field is assigned to the team with the better regular-season record. (This seems to be the most popular suggested format change among the reformist crowd.) Under those rules, the favorite would change for only one World Series (2013), and if we replayed the past 12 years numerous times, we’d see only 0.7 different champions as a result of the format change than we saw in reality (which averages to 0.6 per decade). The more that things would change, the more they’d kinda sorta stay the same.

This isn’t to say that MLB shouldn’t look into switching to a more sensical method of determining home-field advantage in the World Series — preferably one that didn’t involve an exhibition game. But the stupidity of the current policy probably outpaces its actual effect, given that the difference between it and a record-based approach is one different (perhaps more deserving) champion every 17 years.