The Seattle Mariners are a pretty decent baseball team. At this point, 114 games into the season, I feel comfortable saying that. There have been stretches that have spurred immense frustration, and others that have produced nightly elation—but on the aggregate, the Mariners have been pretty good.
And that’s what we’re talking about here, the aggregate. If you follow me on Twitter, you likely know that I like to cite the aggregate, and I’m speaking specifically of the team’s run differential. This year it’s been fun to look at, as it now currently sits at +70, fourth-best in baseball. And if you also follow former Seattle Times beat writer Geoff Baker on Twitter, you know that this has been the source of some lengthy discourse.
First off, let me say this: Geoff isn’t wrong. And this isn’t some kind of diss post. This isn’t the first time we’ve engaged in some entertaining debate, and it likely won’t be the last. The first, actually, if I remember correctly, was when I took issue with him starting off a game recap with some downtrodden language on the Mariners trading away Jack Wilson when I thought it was more worthwhile to focus on what’s ahead, as a young M’s slugger I thought could be with the team a while hit a key go-ahead two-run double—that individual being Mike Carp.
But anyway, yes, Geoff isn’t wrong about run differential. If I understand the basic premise, it’s that the Mariners run differential doesn’t match their record because they’ve had an uneven distribution of runs. And that’s exactly right. A Pythagorean record assumes an exactly even distribution of runs, and the Mariners aren’t at theirs because they don’t have an absolutely normal distribution of runs—because almost no team does.
That isn’t where I take issue though. And maybe this me inferring the wrong thing, perchance being overly defensive where I shouldn’t, but it seems the implication is that the Mariners’ run differential holds little value without some selective shearing of the end total. For a stat that’s entire value lies in accurately portraying the aggregate, this to me is misguided.
Now, I’ll say now, I don’t want to mislead anyone on run differential and Pythagorean record. No one’s saying the Mariners should be 16 games over .500, exactly their Pythag. No one's saying that’s what they deserve, and no one's saying it is exactly accurate when it comes to evaluating the overall quality of this team.
That just isn’t how it works. All we’re saying is that, when looking at a stat that on its face normally does a good job of capturing a team’s overall talent level, the Mariners are likely underperforming by a bit—and they have, and had, the talent level to win more games.
The disagreement lies in the notion that the stat, as it exists, can capture a team’s overall talent. I, and a few other individuals I’ll point to here, think it does.
Before the season, the New York Yankees were projected to be about as good as the Seattle Mariners, perhaps just a little worse. And this was only the case because the Yankees, led by general manager Brian Cashman, committed half a billion dollars in salary. The reason they spent that ungodly sum, despite winning 85 games the previous campaign, was that Cashman knew they lacked the requisite talent to compete. And how did he know? Because they had a poor run differential. No, really, read it. He specifically cites run differential—the entire thing.
Just a few weeks ago, New York Mets general manager Sandy Alderson, generally respected in the game, pointed directly to run differential when publicly evaluating the quality of his ball club. Sitting about even then, he knew his team wasn't great, but run differential told him it certainly wasn't worth blowing the whole thing up.
Jerry DiPoto—the architect of the best team in baseball not using pixie dust, black magic and possum horcruxes—had this to say about run differential at the beginning of May:
"Run-differential is a real number. It’s the most basic way to look at in-season performance once you get past wins and losses. I think it’s fairly reflective of the talent of a team. … The overview of your team and of your season is simply how many runs do you score, and how well do you limit runs. Now what we’re working on is better sequence."
What he mentions here at the end is the big issue. Run differential can't be ignored—because that's how you think this team (-19) is one frontline starter from legitimate contention—but sequencing plays a big role in the conversion of scoring more than you give up into wins. So then, if you're going to mess around with run differential, I'd start with the sequencing, and not plucking blowouts from the aggregate.
As I've mentioned on Twitter, I like what Fangraphs has compiled with BaseRuns. Dave Cameron describes the reasoning for using something like BaseRuns here:
After all, the entire point of looking at run differential instead of actual wins and losses is because we’re acknowledging that wins and losses are affected by the timing of when runs are scored or allowed, and history has shown that run sequencing is mostly just randomness. So, developing an expected win-loss metric that removes the affects of sequencing is a good idea, but pythagorean record only goes halfway to that goal. It removes the timing aspects of converting runs into wins, but ignores the timing aspects of converting baserunners into runs. Evaluating a team by its run differential removes some of the sequencing effects of wins and losses, but leaves plenty of other parts, with no real reason why we should arbitrarily include some sequencing while taking other parts out.
Basically, it plucks out some flukiness. As was discussed earlier this season, the Mariners were benefactors of what's been described as "cluster luck"—things like high strand rate and batting above their weight class with runners in scoring position. BaseRuns aims to neutralize that, which is why the article quoted above, at the end of the Mariners' torrid June, had them at exactly their expected record.
Now, after a July in which they scuffled and a bit of a resurgence lately, things have changed slightly. BaseRuns doesn't say the Mariners should be 18 games above .500 like Pythag does, but it does say they should have won a little bit more.
Here's a quick glance at BaseRuns, with teams of interest plucked out. This is not a complete ranking, which is here. Here's how things shake out for the top of the AL, sorted by how many wins BaseRuns thinks each team should have. You can see the actual numbers on the left, the Pythag numbers in the middle and the BaseRuns results over on the far right.
So now, the Mariners' probably shouldn't have a run differential of +70, because that's pretty ridiculous for where they are—but when looking at the second tier of the American League, you can see they're better than the teams with which they're competing.
And that's why, when you look today, they have a better chance of winning the second wild card than anyone.