This is a critical week in the Presidential election. No, that’s not because of the Vice Presidential debate, which will start right after this is posted. You can decide for your self just how important it is and then revise your estimate downward a week later. Don’t worry, we live in a time of negative interest rates so there is indeed a lower standard of importance than whatever you are thinking.

No, this is the week for the final jobs report. Not the last one before the election – that comes out the Friday before. That will be too close to the election to really sink in, so this is the last one that we can be sure will count. Will it be good and favor Obama III: The coming of Hillary? Or will it be a disaster and herald the arrival of, well, a much worse disaster named Trump?

Bet on a solid 180k gain that will seem decent enough to be called a win. But you never know with these reports. Here’s why.

Could you please fill out this survey? Thanks!

Longtime readers know that there are two jobs reports each month. The report assembled by ADP, which processes 10% of the nation’s non-governmental payroll, is based on a tremendous amount of data compiled in real time. It comes out the first Wednesday of the month, which is 5 October. The Bureau of Labor Statistics (BLS) report is the official one, developed from a survey of 146,000 local agencies charged with keeping track of jobs. It comes out the first Friday.

If you guess which one is considered the most accurate, you’re probably wrong. This is one of those things where the reporters and Wall Street and everyone else favors the least accurate data simply because it’s the most traditional. They’ve been doing it this way since 1947, so why change?

Both reports focus on the net gain in jobs, a number which is found by subtracting two very large numbers, the total employment, and getting a very small number. We’re talking 143M total jobs and a net gain, or difference between last month and this one, of 0.180M. That is a difference of 0.12%, so the error is enormous. But they have ways of evening it out and adjusting it for seasonal variations and it’s supposed to come out in the laundry.

The ADP net job gain per month in blue, the BLS in red. Data from the St Louis Federal Reserve.

The chart above shows the ADP employment report since 2015 in blue and the BLS report in red. If you look at the blue line you may see a drop from an average of around 200k to around 180k gained per month. If you look at the red line you see garbage.

Panic doesn’t help anyone.

The right answer for both is indeed 180k. Above that and we celebrate, below that we scream in panic about how terrible everything is. As you can see, it is more or less like throwing the dice, except the amount bet on dice is much smaller. Also, it’s a better bet.

There will be a lot of talk about this number either way simply because we are in an election year and there isn’t anything better to talk about. We could, for instance, discuss the issues raised by candidates. But I think most people are tired of hearing about the weight of Miss Universe Alicia Machado, so reporters are likely to turn to something more substantial, such as the Vice Presidential Debate.

When that fails, long about Friday, we’ll talk about the BLS report. Trust me on this.

Along with the number of jobs created comes the net unemployment rate. It is probably going to fall to 4.8% for one simple reason – it should have last month and didn’t. The net job creation, even at a slower 180k, is enough to drop the headline unemployment rate (U3) a little bit. This is based on who is looking for work at the time, and there may be some long-term unemployed people who are thinking about getting some work.

More than likely, it’s part of the noise. See chart above for an example.

So what will the final Employment Situation Summary, aka jobs report, of the election cycle look like? Call it 180k and anything very different will be talked about endlessly. There isn’t anything else close to a real issue which matters to people, so why not?