Miami Marlins' one-run wins not indicative of future success

The Miami Marlins are surviving this playoff race more than they are owning it.

The Marlins have scratched and clawed their way to being 3.5 games behind in the jam-packed National League Wild Card race. But the way they have done is not impressive at all; Miami has the most wins in the National League since the second half began, but they have primarily done it in the one-run variety. Prior to yesterday's impressive 10-3 win over the Arizona Diamondbacks, the Fish were 17-12 in the second half, yet somehow were outscored by 14 runs!

Even now, Miami does not boast impressive numbers behind their perfectly average 62-62 record. The Marlins have been outscored by 26 runs so far this year. They own a Pythagorean record of a 59-win team, meaning Miami has outperformed their mark by three wins. And one of the ways that they have done that is succeed in one-run games. The Fish are a staggering 31-18 in one-run contests, which represents a .632 winning percentage.

For some fans, this is a positive sign. There are fans out there who may think that every one-run win is a sign of gritty determination from a scrappy group of kids who are coming together at just the right time. It is through the power of teamwork and never-say-die attitude that Miami keeps mounting these comebacks and pulling out victories from the jaws of defeat. And fans might think that every comeback win and walk-off celebration can only fuel the fire behind the Marlins, and that these wins can only make the team stronger mentally when they come up to a one-run situation again.

The actual descriptors could totally be true. You and I are not privy to the clubhouse, but the Marlins were still having fun last season, and the fact that they are winning is probably helping them still have fun this year. But all of that grit-and-determination stuff works great for movies, but not for real life. In real life, no team has magically found the stuff to win one-run games at a 60 percent pace. Most one-run games come down to a few lucky bounces one way or another, as both teams that night must have been evenly matched. The Marlins have gotten more of thoselucky bounces than not, and that does not mean they will get more of them in the future.

Dave Cameron of FanGraphs covered the same topic on the Baltimore Orioles last week. The Orioles have done it to a degree over the course of three seasons, and yet it still does not seem completely out of line.

But here’s the thing; the existence of an outlier does not prove that a model is broken. In fact, the existence of the right amount of outliers is actually evidence that the model works really well. The question isn’t whether we can find outliers in the data; the question is whether there are more outliers than we’d expect given anormal distribution.

It turns out the Orioles fit in the description of being that rare team that beat out expectations multiple years, and yet it is still possible (and perhaps likely) that, through sheer randomness, they ended up doing so. This is often described as the "Wyatt Earp effect," the idea that a statistically rare event can still occur at random, without a skill being involved.

The Orioles likely have not figured out the magic touch in one-run finishes, and they've done it for a while. Miami only has done it this season, and really they seemed to have depended on it mostly in the second half. Ten of the team's 18 wins this second half have been one-run victories, so prior to the second half, they were just 21-16 on those games, which in and of itself is above average. Did the Marlins suddenly learn how to win in the last three weeks?

Last season, Miami went 24-35 in one-run games. Did the additions of Jarrod Saltalamacchia, Garrett Jones, and Casey McGehee give the team a one-run spark?

Probably not. What is more likely is that the Marlins are performing a little above expectations. Earlier in the year, the Fish were better, boasting a better run differential and a .500 record. At the time, I did not expect the performance to continue, but I also said that the Fish were closer to .500 than they probably were expected to be before the season.

Overall, Miami is left where I expected they were when I discussed their April performance: still in limbo and able to go in any direction from here. If players like Yelich, Ozuna, and Eovaldi begin to make real strides towards success, then Miami has a chance to hold onto some of its good play from the early part of the year. If everything is subject to heavy regression to the standards of pre-2014, then we would expect to see Miami sink really far down the ladder. When you average those two possibilities out, you get what FanGraphs is expecting going forward: a .477 win percentage team (equivalent of a 77-win team) that has made some real strides in the first month but still has a ways to go.

That last line is relevant to Miami. Since that time, Miami's run differential has gotten worse, but the overall product has remained close to .500 for much of the year. But the team's Pythagorean record is that of a .477 win percentage team, right in line with that prediction some time ago. And if you look at their performance and strip out the context and sequencing that can lead to runs gained by bunching up hits or spreading out hits allowed, Miami has a .477 expected win percentage based on their hitting (15th in baseball among non-pitchers) and pitching (21st in ERA-, sixth in FIP-).

If it's held up for this long, why would it go any other direction? Going forward, we should expect Miami to hold to close to a .477 win percentage pace. That may mean the Fish may lose a couple more one-run games, games that you might have thought they would pull out because they "learned" how to win. It turns out winning just takes scoring more runs, and that there is no formula for winning the close ones. The Fish will have to do it the old fashioned way, just like every other team in the Wild Card race.