When I see the word "prediction," I think of something like this:

"It will be a very dramatic change in the sky, as anyone can see it. You won't need a telescope to tell me in 2023 whether I was wrong or I was right," Molnar said at the presentation

I don't see many clear, measurable, falsifiable predictions here. Some of the predictions are already true. How will we know if we had any understanding of current events if we can't check our hypotheses later? Saying "it's going to be a disaster" and claiming validation every time something bad happens is too easy.

Bryan Caplan has recorded a number of bets, some concerning Trump.

Scott Alexander has published a list of predictions for 2017 and keeps a scorecard. He also made ten specific predictions about Trump. Some paraphrased examples:

Total hate crimes incidents will be not more than 125% of their 2015 value at any year during a Trump presidency.

The U.S. Muslim population will increase throughout Trump’s presidency.

The Trump cabinet will be at least 10% minority.

No large demographic group will be forced to sign up for a “registry.”

Other sources make some predictions, but the ones I saw were often fuzzy and unclear, like talk of "impeachment" without specifying whether Trump would be removed from office.

Here are my guesses. In brief, I predict more of the same.

The Wall

There are now "more than 580 miles (930 km) of barriers in place" along the 1,954-mile (3,145 km) long Mexico-U.S. border.

I predict that the wall will be improved, but will remain less than half the length of the border. I predict that Trump will not force Mexico to pay construction costs (talk perhaps, but not actions like garnishing money transfers or withholding aid). Aid obligations in 2015 were $586 million, mostly for drug enforcement. The number varies a lot by year, but I predict it will not drop below the 2012 reported value of $215 million during the next four years.

Jobs

There were 12,265,000 employees in U.S. manufacturing jobs in October 2016. The trend has been downward since the 1980s, though there has been some recovery since 2010.

I predict that this number will be lower than 12,265,000 by the end of 2020.

Other

Twitter will still exist and annoy in 2020.

Trump will remain president for a full four-year term.

The number of abortions induced, as measured by the CDC, will continue to decline as it has for years, with no conspicuous change in the next four years.

The U.S. will remain a signatory to the Paris Climate Agreement.

The number of mosques in the U.S. not be less than the 2,106 counted in 2011 during the next four years.

U.S. GDP growth will be positive for at least three of the four years from 2017 to 2020, and not lower than -2% in any year.

Which of these do you think I got wrong? What are your concrete predictions?