Counting the Congress

How I Count Senators to Evaluate President Trump

In Trump Watch, I outline three scenarios all of which depend on whether the Trump Presidency is perceived a success of a disaster. And, as I already pointed out, the only objective measure in the short term is how many legislators are willing to take action either for or against him.

The current count is 52 Success / 61 Disaster

So, now let's look into that.

The theory is that failures or successes of the Trump administration will eventually have electoral consequences to a congressman, which will, in turn, affect support. FiveThityEight is, in fact, tracking this through their votes. They call this the Trump Score, but to avoid confusion I will call it the 538 Trump Score. This is good info, and probably better than what I can give you. However, it does have a problem for our purposes. It does not come down to a single score.

Fortunately, (with some effort) I figured out how to export the numbers to a spreadsheet. For now, I'm looking at the Trump Plus-Minus, which is their predicted score minus the actual score. Currently, their average and median are almost 0 for all cases indicating, overall congressman are voting as predicted. However, that is only because both sides are highly partisan and voting as a block. And, so each side has deviated from their predicted score by about the same amount. I've detailed the numbers, here.

This in itself is a good measure, especially for telling if partisan solidarity is holding. However, it’s hard to see if Congress would be willing to take his lead or proactively take action against him.

There are his approval and favorability ratings. Certainly, these values are a precursor to his success or disaster. Why I'm not using them yet is the I'm not finding numbers I'm comfortable with, mostly because it's too soon. In particular, I've not been able to find aggregates broken by party affiliation or, better, states. Without this, I'm vulnerable to my being mislead like many where during the elections. However, as you will soon see, I will use them to validate or invalidate certain assumptions. More importantly, if I can get good numbers, I can revisit the mindsets I used before (re-evaluated by what we learned from the election).

So, we're back to counting congressman. The goal here is to get an estimate of the number of congressmen that are ‘independent' enough to go against their party. Mostly, because I can’t realistically count representatives I am sticking to senators. My defense for this is that the goals of the Senate and House are currently aligned so the Senate can be used as a proxy for the House.

Let’s consider the following scenario. The Senate Select Committee for Intelligence has ordered the White House not to destroy documents. What if in a couple of weeks, the Senate discovered documents were destroyed. We now have the same situation that harmed Nixon and Clinton. Would the Senate consider impeachment?

Unless McConnell and Ryan could get the votes, they would not even call for a vote. So, assuming President Trump's popularity has not changed, how many votes could the get?

That would be the Disaster Count.

Now, let’s take this another way. What if President Trump made a proposal that is consistent with his promises, but electorally dangerous to senators? For example, a large tax on imported goods. For now, assume the bill is well written and the President is able to articulate a good economic argument. (Can you believe I have to say that.) How many senators would fight for such a bill?

As before McConnell and Ryan would need to know they could get the votes to let there be a vote.

That would be the Success Count

So, the counts are derived by deciding how many senators are ‘countable' or not and then deciding which would support a ‘success' of ‘disaster'. The rest or 'base' is added to the counts for each scenario.

To be ‘countable’

• The senator represents swing state or a state different from his party. As most do, this is defined by the vote margin for or against President Trump 
• Or, the senator has demonstrated recent actions that show independence from the President and the Party. I will have to change this over time. So far, I'm just adding republicans with a 538 Trump Score bellow 100%. 

13 republican and 17 democratic senators represent swing states or states opposite their party.

Currently, only 4 republican senators have a 538 Trump Score bellow 100%. Only Rand Paul meets this criterion and has not already been listed as ‘countable'.

38 republicans and 31 democratic senators are NOT 'countable' and are the 'base'

You may have noticed that the scores are higher for both than my eyeball guess in the post, Trump Watch. I decided not to attempt the most conservative measure because I felt comfortable making the following assumptions.

As long as Presidents Trump’s popularity remains bellow 50%, and democrats solidarity remains strong, we can assume:

1. All swing states can be considered democratic states 
2. Republican senators not up for election in 2018 are free to be independent. This is reasonable because politicians have to have faith in their ability to campaign and can recover if there is enough time. 
3. Even democratic senators in republican states will remain loyal to the democratic party. This is reasonable because of primary threats that are realistic under the current fervor. 

I have all this in detail, here. In addition, there is now a link on the top toolbar ‘Congress Count'. Please note that I cannot update these values regularly so pay attention to “Last updated ...”

You may ask, could a President be successful and have a low Success Count. Yes, since there have been successful presidents without help from Congress. Theodore Roosevelt and Harry Truman are good examples.

And could not a disaster happen before the Disaster Count shows it? Unfortunately, yes, especially with international scenarios.

Both these are why I have to continually make adjustments as events unfold. There’s my cowardly qualification. You’ve been warned.