How Do We Know Without Recounts?

We have all seen many articles and posts on the recounts in Wisconsin, Michigan, and Pennsylvania.  We are likely to see many more.  For now, here are a few points about the recounts:

I am entirely in favor of  thorough post-election audits and recounts.  

Jill Stein and those who have contributed to her fund are doing a service to democracy and for all voters. We should have routine audits and recounts after every election.  Currently only about half of states have post-election audits and routine close-vote recounts. Neither is sufficient alone.  An audit finding discrepancies, that if widespread, would change the winning candidate(s)  should result in a full recount.  Close vote recounts alone are insufficient.  Without audits we cannot be sure that the results are not off more than the trigger for recounts.  At a minimum audits should be risk-limiting, subject all ballots to audit,  check the entire totaling process, and assess ballot security.  Audits should also cover the registration and checkin process.

With new techniques such as single ballot auditing and ballot polling audits, post-election audits can be quite economical.  With detailed election reporting, auditing the total result can also be accomplished efficiently.

I am entirely in favor of the recounts initiated by Jill Stein. 

She and those who have contributed donations and time to the project are doing a service to our Democracy and every voter.  Since apparently, Wisconsin, Michigan, and Pennsylvania lack post-election audits the recounts are necessary.  At least, in Pennsylvania, the margin is now so close that a close-vote recount and checking of all totals should be automatic.

Even if there is no change in the state winners, Election Integrity has won already

The initiation of the recounts have already highlighted the lack of integrity in the current system.  We now know that it is hard to get recounts in these three states; that the recount laws, at least in Wisconsin, are inadequate to cause a “real” adversarial recount as we saw in Minnesota in the Coleman-Frankin recount; in Pennsylvania the law for recounts is ungainly requiring affidavits by three voters in each polling place in a short time; and highlighting the impossibility of really auditing unverifiable DRE, touch-screen, voting machines.

Yet, maybe we will not win that much in the end

Still I am skeptical that the audits will result in real change toward national minimum standards for voter-verified paper ballots along with sufficient audits and recounts.  Perhaps if the recounts reverse the result in one state or at least show a significant level of change in the numbers that will be enough to result in enforceable Federal minimum standards, or failing that reform in a number of states.

All the objections to the recounts are partisan

Lets start by conceding that Jill Stein and the Green Party hope to gain from doing this service.  What candidate or politician does something without a hoped partisan gain?

The initial complaints against the recounts came apparently from Hillary supporters.  To me, it seemed that they blame Jill Stein for Hillary’s loss as well as Bernie supporters.  Whenever there is a close election the apparent looser has many individuals and groups to blame, while the apparent winner has many to thank.  The closer the election the more small factors can directly contribute to the result.

Now the Trump team is objecting to the recounts, (after campaigning on maybe not accepting the initial result).

Thus has it always been. See our review of Ballot Battles.

Some Coverage:

Robert Koehler via Common Dreams: Vote Recount vs the Media Consensus <read>

In other words, the American president is essentially determined every four years by a sort of quick-draw consensus of corporate media conglomerates, not by a cautiously precise hand count of the votes that have been cast

There is evidence already for suspicion in Pennsylvania: Walter Mebane Jr. via the  Washington Post:  New evidence finds anomalies in Wisconsin vote, but no conclusive evidence of fraud <read>

Walter Mebane has a unique way of analyzing elections for suspicious results.  He analyses the digits in the numbers reported at a low level. The lowest digits should fall into a certain, non-random pattern in most elections.  If there is wide manipulation of data it is very difficult to mimic those expected patterns.  His analysis points to suspicion in Wisconsin.  Yet, for now its a bit less than “where there is smoke, there is fire.”

An article in Time supports better election night reporting data: How the Wisconsin Recount Could Help Fix American Elections  <read>

And this on the importance of election security by David Dill via Scientific American: Election Security Is a Matter of National Security <read>

It is not good enough to say, “We can’t prove fraud.” In every election we need evidence that vote counts are accurate

Facebooktwittergoogle_plusredditpinterestlinkedintumblrmailFacebooktwittergoogle_plusredditpinterestlinkedintumblrmail

Leave a Reply

You must be logged in to post a comment.