Here's How We Do Paper Ballots

Start a company in your county with the sole intent of counting paper ballots.

In my small county, in a recent election using DRE's, the amount per vote counted was done at a cost of $7 per vote. No shit. I could have made a thousand bucks that day if my company had been in business!

How do we beat Diebold, et al? We beat them through competition. While I could have made a thousand bucks in that last election, I could have settled for 100 bucks and beat the price of the DRE's by a thousand dollars, easily!

The DRE's are a ripoff in more ways than one, so they can easily be crushed in the free market economy of counting votes. Not just price wise, but with far greater accuracy and trust.

They are done for, at least in my county! The free market rules!

How many voters in your county? What would the steps be?

Will you write a NYAQ and post it here?
Not a FAQ, a NYAQ.
Not Yet Asked Questions.
But, you know, the ones you will be asked.

NYAQ's, eh? The devil IS in the details. I was hoping I'd get back on line these 8 hours later and have a few question answered, to tell the truth. Alas, I have come up with a few myself, in these 8 hours, and I will be posting them forthwith - later on this thread. I have been pondering this idea, y'know.

$7 per vote?

damn that's high

we might have to focus on recounts that are mandated to be counted by hand, at least for now. What do you think?

ballot next year. This is a big mountain to overcome, and it must (with this congress) be done at the state level. I was literally laughed at about this initiative when I talked about it to prominent Reform Ohio Now framers including an election lawyer. We have to sell it to the people, because it doesn't get support from the parties or BOEs.

The other issue, at least in Ohio, is that we have to have fair elections before we can expect it to win-a catchh 22.

Also, how would you address "nonpartisan" question -- i.e. that your "counters" would be trustworthy?

grrreat idea!

There is a small city in my small county. It recently elected the Mayor and Council members. The cost to rent the machines was $1500, and there were two hundred and some odd voters. $7 and change a vote.

I just now calculated the costs for last year's countywide election, and have come up with a figure of $500 a precinct. With about 500 voters per precinct the cost would have been about a dollar a vote. Still, if three employees can count those five hundred votes, and get payed $100 each, a profit of $200 could be made, or the county could save that $200 per precinct.

Imagine a store clerk who counts money all day for $40 takehome, getting a chance for some side work one evening that pays $100. I think those willing to enlist would be rather large. These would all be trusted full-time employees, and there could even be some bank clerks interested.

set some lovely gifts and perks aside for "promotional" (read: bribery) purposes.

It would take some very interested people being involved to get the ball rolling. The competition from the machine pushers will be intense. But with a guarantee that the vote was counted properly, timely and accurately, it seems many folks would be interested.

The company could be a for profit or a non-profit.

There will need to be a close coordination with the Board of Elections at all levels, the local election officials, and each precinct. The votes need to be counted at each precinct, eh?

It should take no more than three employees at each precinct, working for no more than an hour or two or three after the polls close. The employees could be paid or be volunteers. The counters will be professional - meaning bank clerks, cashiers, etc. Off hand I say about $200 a precinct would be the cost. Compare that with an estimated $500 a precinct in our last election.

Since the election machine vendors have such crappy equipment, it should be a breeze to sell the idea to the county.

How can they say no?

Admittedly, in my small county, the task would be far simpler than larger urban counties, but I imagine the fees charged by the machine pushers can be easily undercut even in large cities.

It is really just a matter of being prepared and getting the work done. A huge task in some cases, but if you find out what the costs are for machines, factor in the recount scenarios that will come with Papered DRE's, and the future of democracy as we used to know it, I think the task can be managed.

More to come...thanks yall!

Want to watch your vote get counted !

Want to be able to verify that your vote got counted !

Watch you county or precinct votes get counted and posted live and in real time online !

We use BIO-DEGRADEABLE Paper ballots bought locally !

We hire local printers to print the ballots !

We hire local people to count the ballots !

You are strongly encouraged to come and observe the ballots being counted !

How could those election theft machines compete with this.

Great idea BeFree.

asking for paper ballots is a required option in most states of the union. Starting a strong campaign keeping the faith in our election system but taking it away from the machines seems like a promising idea. Paper ballots are required to be counted, if nobody votes on electronic voting machines, there will be no longer a demand for it. However, people need to vote, so denying the machines during elections would be great, but demanding the right to vote is important.

You can't rig tabulators close to any way you can rig Diebold machines on a large scale very easily. That's for sure...

any takers?

In England, they threw out machines and went back to paper. Bank tellers count the ballots in public, observed by anyone who wants to watch. It works. England had a little problem with mail in ballots (but Oregon doesn't!). But I think the English model works. It's there.

NOW, here's the cool part. There will be a time in the next few months when everybody realizes it's a rigged game. The public is more than ready and the story is leaking out more and more.

We need to be ready. Our response: USE THE ENGLISH MODEL!

do you mean actually do this? Or is just talking about it enough?

Funny seeing you here? Ya want to talk about this? Or do you think it's a stupid idea, not worthy of discussion?

It is an idea, eh? One with kinks and twists and turns. Who knows but that the idea alone might scare the election people to be more circumspect?

Whatever, its out there. I just may be able to swing it, stay in touch. Seeya later.

Counting paper ballots when there are twenty or thirty selections on a ballot, can be quite consuming. It's why the poll workers, after a twelve hour day, are not too keen on sitting around counting ballots till midnight. E-voting takes care of that problem, for them.

My idea would be to bring in a fresh group of folks, trained to work fast and accurately, and overseen by election judges to count the ballots, precinct by precinct.

In big cities, the variable ballot styles create quite a challenge. I don't know that my idea would work in such areas, some type of scanner might be needed. In that case, my firm might contract to do audits of the machine counts?

I know for a fact that DRE's will always be cheaper than paper in the end. The problem comes from ballot printing. You can not use cheap laser printers. Printing has to be done on that SAT type paper that you used for the SAT test. It is accurate for the right angle the paper has to be. plus you have to make sure that there are more ballots than there are registered voters as there are provisional ballots as well. This means that you will trash half of the ballots every time as half the lazy Americans will not vote.

The problem with a second group of counters is that the cost of the labor is too expensive. I know that LA County uses high school kids that the county bribes high school teachers to bring in. That could be one way to cut costs as the high schoolers work for free. The other issue is California and Illinois and a few other states that like to vote on every cotton pickin' issue and official at once.

We in NJ like to have little elections for every little set of offices (fire officials, school budget, school board, town council). This way we can budget our corrupt bribes better instead of having to fork up the cash all at once.

One polling site:

- 5 DREs at a polling place at $4000 each initial investment = $20,000

vs

- 1 scanner $5000, 20 polling booths at $100 a piece $2000, paper ballots 1000 at .30 each $300 = $7300

plus, you tell me which polling place is more likely to have voters waiting ...

and a lighting strike frying the circuts of the DREs at a polling site, what do you do then? no backup as there is with paper ballots.

Any costs of paper ballots for scanners is still going to be less than the cost of added initial ivestment for DREs and ongoing maintenence agreements for said machines plus added labor and hassle from having 80 year old ladies having to set up several of these beasts at each polling place.

First off if the county pays $4000 then they are being ripped off. The first major county was Riverside at $4200 ea. The price should be around $3000 if the county is smart enough to know that companies like Dell are willing to work with smaller vendors like Avante to bring the price down to where it should be. Maintenance agreements should be the same for either system as both require more maintenance on the ballot generation end than the machines. The space to store machines is the main cost.

I remember that small towns in New Hampshire were paying $.80 a ballot three years ago. The paper has to have a perfect 90 degree angle. This is not paper that you can get at Staples. I will even say that the ballot is printed perfectly every time.

So the new math is $15000 for DRES that will last at least 10 years. We have three elections (on average) a year here in my little part of NJ. That includes separate elections (and in some cases primaries) for fire chief, school board, school budget, mayor, governor, and pres/congress. For our 1000 ballots (which coincidentally is the number of voters give or take 100 here), the cost at the NJ printing office in Mt. Holly (I got a quote) is $750 per election. So in twelve elections, the DRE starts paying for itself. That is four years where I live.

The lightning strike will screw up the scanner as well. Any natural disaster will screw up a polling place. The issue of delays occur with either type of machine. I have voted both ways several times. The line can be the same at either type of voting. That is usually that lack of management of the election officials to anticipate the number of polling places needed and of people to man the check in process. I have been behind the scenes at several elections over many states and there are some people who manage the flow of people like a professional at an amusement park and some that pile everyone outside in the rain.

I really am not trying to sing a DRE's praises. I just know what the arguments that are made in RFPs across the country. I have seen the cost analysis in detail. I have read winning RFPs so I know that the numbers were real not some company trying to get in on the action.

The arguments about the old ladies and the voters not being able to use the complicated machines will fall on deaf ears. I took a survey of election workers at a large California county (>500,000 voters). Of the 1200 poll workers, the average age was 59. 85% of them had an email address. The registrar in that county had monthly training throughout the year. She had maps of every polling place with the layout of the traffic flow and check in areas. I wish that there were enough of her to go around.

This topic is Election Reform. We have to talk more about the entire process here. I do not see enough talk about the polling places, the training, voter registration, and campaign practices. If it is scattered elsewhere then good.

I just was so frustrated with the focus so much on the ballot that we loose sight of the people and their roles in elections.

Thanks for letting me vent. My brother in law was right to let it out. Feels better.

Thanks and peace.

We have a ways to go to live up to the (entire) name of "Election Reform," it's true. (The name is fairly new to us, actually.) But keep stirring the pot, and we may catch on.

Meanwhile, forgive me while I stay on topic (more or less) and pick your brain on costs.

If I understand rightly, BeFree's OP referred to paper ballots _hand counted_, so I figure that knocks the capital costs down and the labor costs up (something I favor in principle, if the balance isn't too far out of whack). Any thoughts about costing that out?

I see that you posted on the thread that argued that "Touchscreen Voting Increases Annual Election Costs by 40%" -- was there something wrong with the analysis there? was the time horizon too short?

time and accuracy. Currently I run a business with my brother in law. As a small business, owners do all the odd jobs required. We received a product (about what would fit in a half of a sandwich baggie). We received 500 of them. As is customary with Chinese manufacturers, the cardboard box was flimsy enough that it teared in several places. This meant that we had to count the product as soon as we got it in case we had to make a claim for missing items. With bro, myself, and another, we arrived at three different counts on our first tries. As all three of us count products all the time, this is not a new process. It took us making little piles of twenty to come with a solution. It took us over an hour to agree on a number.

Imagine a ballot in say Rock County Illinois with over 400 ballot positions. A ballot position can be a "Yes" or a "No" or a candidate name or a judge "affirm" or "reject" and that does not include write ins which are done during the canvas. Say half of their 100,000 voters vote. That is 20 million reviews that have to be made. Each empty circle has to be "eyeballed" to make sure that duplicate marks are not made. You just can not count just the pencilled in marks. An alert person may average one second per review. That is 20 million seconds or 5555 hours. Mulitply this by three since there has to be at least three different people looking (a counter and two partisans). That is 16,666 hours. Now lets say that a good team really has only 6 hours of good counting in them for one day. That would be four shifts. The math is now 694 days for four teams (12 people).

We now call this a team day - 12 people, 24 hours. Now the force multiplier is easy. For one day turn around it will be 694 teams or 8333 people. That is 16% of the voters. OK, what is now the patience level of the voters? Say it is a week. That brings us down to 1189 people. That is pretty darn close to the population of the high school I graduated from. So we need access to a medium sized high school for a week. How much would a whole school cost to support? That I do not know. It is November in Illinois so it could be cold. You need food service, janitors, police, etc. Move the kids somewhere for a week or just give them days off. I would think at least $20,000 a day.

Oh yeah how much do we pay the people? No matter how patriotic we think we are we still have to pay them at least $10/hour? Also there are supervisors, county elections officials, etc. i just have to stop here. I just broke $2 million dollars for one election in 2005 (meaning all local elections). Countrywide that could mean billions spent on just counting the vote. I did not include printing of ballots, poll worker costs during the election, or other pre/post election costs.

The DRE vendors will do this math for you and say that if they supplied the same county with one DRE for every 200 people, you could buy a new machine for every election at $3000 a piece and still save $500K per election.

People mention Canada, but they do not have as many ballot positions as some major states like Il and CA.

(I was just saying in another post, about the Ohio recount, that anyone who doesn't understand why BoE officials would reasonably want to avoid full manual recounts may never have undertaken a truly massive and tedious clerical task.)

I'm not sure I buy your time-per-ballot estimates. (Don't get me wrong -- I'm not deeply invested in how this analysis turns out. I just want to get it right.) I'm looking at a 'random' precinct from Cuyahoga County (OH) in the 2004 general election
http://boe.cuyahogacounty.us/BOE/ballots/PDF/110204/BRHT03A.pdf
and I think I see 78 ballot positions in all, 46 races (many of the races in this case are uncontested). My count may be off, of course. It would be a bunch more in New York where there are multiple lines per race even for the uncontested races, although I think NY is unusual in that. Can you point me to a ballot for Rock County IL so I can see where all the ballot positions come from?

Apart from that, I really can't evaluate whether an estimate of 1 second per ballot position is reasonable. I would be more comfortable with observed times from an actual process designed to minimize count error. (I am thinking that an alert observer can, in principle, verify that a ballot position is blank in much less than 1 second -- but even if I am right, that doesn't necessarily mean that an actual count would proceed faster.)

I based it on seeing a real recount and how everyone takes their time. I was averaging over a long day. It may start out at .2 seconds then drag on when there is a messy ballot or stray marks. My numbers did not include time to review any questionable dots or erasures.

Maybe over the weekend I will have time.

Thanks,

I have worked elections where a site will call and say all polling booths are being used and people are backing up. It's easy to bring out a couple of reserve $100 voting booths. Then the only delay machine-wise is how fast a tabulator can take a ballot. With county budgets I cannot see many able to afford to have a lot of extra $3000 DRE in reserve for such a situation.

As to disasters, a DRE is corrupted, the record of votes is too and there is no backup. With a scanner or tabulator, the memory card goes bye bye (as I have seen happen in the middle of the day) at least ballots are there to be re ran or counted.

My experience was different with pollworkers, most were in their 70s and more than a few had trouble with tabulators in setup or closing.

I do appreciate your comments as an ongoing discussion of DREs is needed as well as the planning problems with saw with adequate staff, ballots, etc. by many counties.

Well, I have introduced the fact that one local election cost $7 a vote, using DRE's. Also, the fact that the last county wide election was $1 a vote.

I can get paper copies made for less than 10 cents. I could get volunteer counters to work for free, or pay professional money counters as little as $25 dollars for 3 hours of work.

So, I do not believe your statement: "I know for a fact that DRE's will always be cheaper than paper in the end"

Besides, cost is NOT the most important consideration. Accuracy, accountability, and a restoration of confidence in elections IS.

Breaking up a complcated voting process into simpler, more defined decisions is a wise idea.

Following that idea comes a proposal that all federal elections be separate. That would mean the President, Senator and House of Representatives are all elected the same exact way across the entire nation. Congress has the power to mandate that, and what better way is there of accomplishing that than through the use of Paper Ballots, Hand Counted?

are printed on the flimsiest paper you can get AFAICT. Looks like the kind of stuff we used to use as toilet paper (you'll be pleased to know we don't anymore, it's safe to visit).

But then we hand count ours...(votes that is)

There can be several dozen offices, propositions, judge retentions etc. on some state ballots in the US.

I think that is probably the biggest bar to HCPB in the US. We do sometimes have more than one race (referendums are rare) in which case we have more than one ballot paper, printed on different coloured paper.

Hand-counted paper ballots are a great idea, and there is no problem with them in most countries that use them, but it is a bit difficult to see how it could work in the US.

Switzerland has a lot of referendums - does anyone know what Switzerland does?

How many people would it take?

A few snippets from one of kster's links, http://www.ecotalk.org/VotingMachines-LegalVoting.htm (I'm not taking any position, just trying to gather information):

In Canada in 2000, "...even though there were 13 million ballots nationwide --- on average 250 or so easily read PAPER BALLOTS were counted at EACH of 50,000 precincts. The counting was done by about 6 counters at each precinct."

I'm not sure where the "about 6" figure comes from, because the AP story (following this assertion) states that "roughly 150,000 election workers" conducted the count. I think the estimate may be three election officials per precinct plus three partisan "scrutineers" (observers).

The AP story says that the count was virtually complete within four hours. Another part of the link (see "Example 1") says, "An estimate for the time required to count 300 ballots with ten races and an average of five candidates per race is two and a half hours." One could adapt to more ballots by splitting them and counting them in batches. In Ohio in 2004, the average precinct had about 500 votes, so one might shoot for 12 counters per precinct. Not awful so far.

Backing up... well, Cleveland -- proper, not metro -- has about 415 precincts and a 2000 population of about 478K, about 1150 people -- not adults -- per precinct. So at this point in the exercise we are looking at about 1% of the population being involved in the count, or about 10,000 in a metro area of 1 million. That is a lot of people.

But the other problem is the number of races. In one (fairly representative) Cuyahoga County (OH) precinct in 2004, there were almost 50 races and issues, although many of them were uncontested. I imagine that Cuya has an unusually high proportion of uncontested races, because it is so heavily Democratic. Of course since the U.S. has basically a two-party system, there usually aren't many races with more than two candidates. Still, it starts to seem like a nightmare to count the same ballots almost 50 times no matter how one divides them. We can increase the number of people and ballot piles (and then add totals from lots of piles), and/or we can expand the time by doing the count on the following day (which will complicate ballot security).

We can play with cost for a moment. Suppose that in a 500-person precinct we are paying 6 people for 3 hours at $10/hour, but stiffing the partisan observers -- that would be $180/precinct or 36 cents per vote. If we need to multiply that figure by five (whether people or hours) in order to accommodate more races, then it is $1.80 per vote. I don't take that particular figure very seriously; I am just illustrating possible calculations.

transparent, and public, same day hand counting of ballots in countries like Canada, France, Israel, India, Switzerland, Britain, and many other countries. See

http://www.ecotalk.org/VotingMachines-LegalVoting.htm

http://www.blackboxvoting.org/