Does the A’s negative run differential really matter this early?

The Oakland A’s have a 25-17 record through mid-May, putting them first place in the AL West division and tying them for the MLB lead in wins. They also have a negative run differential, having been outscored by their opponents to the tune of eight runs.

It would seem like only one of those statements can be sustainable. Either they’ll keep winning games and their run differential will improve accordingly over time, or they’ll continue getting outscored and it won’t always keep working out this well and their record will drop. And since run differential tends to be considered a more predictive measure, it’s easy to see this as a red flag for the green-and-gold.

Here are a few reasons not to worry about it.

Small-sample gremlins

Time moves fast these days, but the nature of small samples hasn’t changed just because we follow news by the minute instead of the day or week. The fact is, we’re still only a quarter of the way through the season.

Remember the A’s awful start from the first week of April, when they went 0-6 and got rocked every day? It really happened, but right now those six games represent 14% of all the contests they’ve played. By the end of September, it will be less than 4% of their season. It will not continue to have this big of an effect on their numbers, so the question turns to whether you think it will happen again.

• First 6 losses: 13 runs, 50 allowed (minus-37 differential)
• Next 36 games: 162 runs, 133 allowed (plus-29 differential)

During those six games, the A’s got blasted by two good opponents in the Houston Astros and Los Angeles Dodgers, but they came back and beat the Dodgers in their series finale and then immediately beat the Astros in a rematch series. During the slump they had several key hitters dipping in and out of the lineup with various early injuries, including Sean Murphy, Ramon Laureano, Matt Olson, and Chad Pinder, and now they’re all healthy and producing like normal and newcomer Seth Brown has joined them.

During the slump the starting rotation got knocked around, but since then it’s been mostly reliable. In the bullpen, reliever Reymin Guduan allowed eight of the runs himself in meaningless mopup duty, and platinum defender Matt Chapman made an uncharacteristic error that led to three runs against the Dodgers. These things happen, but over the course of a full season they even out. When it all happens in the first week, it skews a metric like run differential to an extent that confounds it completely, and it takes months to correct.

Look at that split above. If a 42-game sample of run differential means something, then so does the most recent 36-game sample, which says the A’s are a perfectly reasonable first-place team who has been a bit extra fortunate in one-run games. It’s more likely that they’re a contender who had a disastrous opening week and then settled into their long-term groove, than that they’re going to play like the atrocious 1917 Philadelphia A’s one week out of every month the rest of the summer.

They really did lose those April games, and they really do count in the standings, but they snapped out of it. I’m just not sure how meaningful the skid is in the realm of predictive analysis, other than, yes everybody has a bad week now and then.

Expectation

Furthermore, it’s not like this A’s team is coming out of nowhere.

• 2018: plus-139 differential, and 97 wins
• 2019: plus-165 differential, and 97 wins
• 2020: plus-42 differential, and 97-win pace (in 60 games)

Granted, the roster has changed over time, and there are plenty of question marks left to resolve in this current iteration. But this is a proven October-caliber core with obvious upside, and that counts for something when you’re deciding whether to believe a small sample. Especially with an A’s team that has a history of starting slow, making the necessary adjustments, and improving over the course of the summer.

Historical precedent

Even if the A’s run differential were to stay low, with the club losing blowouts but winning close affairs, it’s not even impossible to reach October that way.

Off the top of my head, the 1997 Giants won their division with 90 wins and a minus-nine run differential. In 2012, the Orioles won 93 games and a Wild Card at just plus-seven. And perhaps most relevant, the 2007 D’Backs won 90 games and their division with a minus-20 differential, under the managerial guidance of none other than now-A’s skipper Bob Melvin.

I’m not expecting it to come to this, but, just saying. It’s happened before. The A’s own manager has even done it.

And it goes the other way too. The 2014 A’s were at plus-145 at the All-Star break, to go with their 59-36 record, and we all remember how poorly that ended.

***

Run differential is a useful metric, but it’s not a be-all end-all and it can be fooled by a fluke over the span of 42 games. The 25-17 A’s aren’t any more guaranteed to win their division than the 0-6 A’s were to be an unexpected disappointment, but I know which one I’m betting on.