How you interpret Friday’s jobs report depends on whom you are voting for next month.

If you’re a Donald Trump supporter, you probably focus on job growth slowing relative to last year, the rising unemployment rate, or broader measures of unemployment that show an unusually high number of workers who want full time work but can get only part time work. If you are a Hillary Clinton supporter, you choose to emphasize the historically low level of official unemployment, rising wages, and a rising labor-force participation rate.

But while these metrics are useful in evaluating the economy, they are not necessarily great predictors of who will win the election. Back in May, Fortune looked at which statistics are actually best correlated with victory for the candidates of the incumbent or opposition parties and found that Hillary Clinton had the advantage. Now that we are just a month away, it’s time to check back in with the statistics that best indicate what voters will decide on November 8th:



ISM Manufacturing

According to a 2011 analysis by statistician and blogger Nate Silver, the best economic predictor of presidential electoral success for the incumbent party is the average ISM index reading from January through September of election year.

As you can see, the American manufacturing industry has spent much of the year in expansion, but the average reading so far this year is only barely positive, at 50.9. According to Silver’s analysis that correlates with a slight loss for the incumbent party’s candidate


Advantage: Donald Trump

Job Growth

The second most important statistic is the net growth in nonfarm payrolls between January and September of the election year. Since January, employment has increased roughly 1%, which correlates roughly with a tie in the popular vote.

Advantage: wash

Change in Unemployment

The third best predictor for presidential elections is the change in the unemployment rate between January and September of the election year. The unemployment rate has actually gone up slightly this year, from 4.9% in January to 5.0% today. That’s bad news for Hillary Clinton, as the incumbent party has never won reelection in a year when the unemployment rate was higher in September than it was in January since World War Two.

There are some mitigating factors that Clinton supporters can take solace in, like the fact that the unemployment rate is only barely higher in January, and that the rise is mostly the result of the workforce growing rather than businesses firing workers.

Advantage: Donald Trump

Though these statistics are clearly good news for Donald Trump, there are some caveats. We should hesitate to put too much faith even in the metrics that have a good track record of predicting election results. As Silver wrote at the time, none of the individual indicators correlates strongly enough with election results to explain more than 46% of presidential election results, which means they will give you incorrect predictions more than half of the time.

That Hillary Clinton is up in the polls right now is evidence that economic statistics aren’t the only thing that matters. But the economy is clearly in Donald Trump’s corner at this point.