One of the truly gratifying aspects of this blog is the expertise offered by readers, both in the comment threads, and through private email. For example, my analysis of the legal issues surrounding the election contest lawsuit would not be nearly so coherent (or accurate) without the invaluable help of Lawyer X. Likewise, I have been greatly aided in understanding the intricacies of the GOP’s proposed “proportional deduction” by a reader whom I would call Statistician X… if not for the fact that you all know him as DJ, a frequent contributor to the comment threads.
In recent days, the Rossi camp’s spin has veered away from the felon vote, and back towards the “total mess” thesis… an unsurprising tactic considering the number of offsetting illegal votes the Democrats have alleged. But this shift is also likely due in part to the report recently issued by the Democrats’ expert witness, Christopher Adolf, who effectively exposes the Republicans’ data as “unrepresentative”, their methodology as “flawed”, and their entire case as “hopeless”.
DJ has graciously summarized Prof. Adolf’s report for me, and not surprisingly, found that his arguments largely echo the same arguments DJ has made in the comment threads here and elsewhere. As DJ modestly explains, any statistician doing an objective analysis should reach the same conclusions. But I’ll just let DJ explain it in his own words… think of it as a kind of guest blog.
Professor Adolph states the first flaw this way: “Use of a non-random, non-representative, and incomplete sample of invalid votes … is useless for answering questions about the net effect of all invalid votes on the state-wide election outcome.”
The take home message is that the Republican’s invalid voter data are not suitable for any type of statistically or scientifically valid analysis.
Essentially, there were systematic biases in collecting the invalid ballots. The invalid votes are not drawn randomly or representatively across the state. Adolph did a very simple analysis to demonstrate this fact using the Republican and Democrat invalid voter lists. The Republican list was certainly cherry-picked. The Democrat list was cherry-pocked, too, but that is partially an artifact of the Democrats going to the precincts that the Republicans ignored.
Adolph points out that “[o]nly a complete statewide census, or a random or representative sample, has any hope of offering an unbiased estimate of the effect of invalid votes.” What should the Republicans have done? “Only if the Rossi campaign put as much effort into searching for invalid votes in precincts they won . . . as in precincts they lost can we begin to trust the experts’ reports.” (This is identical to the “equal scrutiny” argument I was making.)
Professor Adolph goes into much detail on the ecological fallacy in both Professors Katz’s and Gill’s analysis. He points out that “Modern, accepted [statistical] methods would provide strong warnings to the researcher that ecological inference in this case is impossible.” In other words, Katz’s and Gill’s arguments constitute an ecological fallacy. Both of the Republican expert witnesses make a single critical assumption that vote choices within any sub-population in a precinct are the same as the average vote choice in the precinct as a whole. “This assumption is strong, implausible, and unwarranted. Moreover, it flies in the face of decades of warnings from social scientists and statisticians. . . .” (Adolph).
To demonstrate this, Adolph first provides some technical background on Gary King’s solution to the ecological inference problem (which is state-of-the-art social science). I will skip the technical details, except to say that ecological inference can work under some conditions. One condition is that the groups about which we wish to make inference are represented in high enough numbers to provide mathematical limits (or bounds) on their characteristics.
For the 2004 election, it would be necessary that a reasonable fraction of the votes within precincts were invalid votes in order to infer the limits of their voting behavior from the aggregate voter data. Such is not the case. In all 1,344 precincts in which invalid votes have been identified, none of them have a high enough fractions of invalid votes to set these “bounds” and infer felon voting patterns out of the aggregate patterns. This is mathematical fact, not opinion. “The bounds [test] show[s] it is possible that every invalid vote in every precinct was cast for Rossi. Alternatively, every vote may have been cast for Gregoire
Man..you have put some serious effort into dicrediting and confusing proportional analysis Goldy.
The bottom-line is there ARE illegal votes.
No Judge is going to ignore that fact.
Those illegal votes must be validated with hard evidence..no doubt.
But once determined…it appears you are still saying the Judge will just ignore them because of the complexities in apportioning them.
How can a Judge ignore illegal votes???
Hair splitting that might prevail in court, ok.
But man, too many facts out there for you guys to do anything more than damage control.
One party, one state, one ruler.
Cynical@1: And Judge Bridge has said that the fact that there were illegal votes is irrelevant, unless it can be proven that those illegal votes changed the outcome of the election.
The mere presence of illegal votes is not sufficient. Because every single election ever held has (and will have) illegal votes.*
* Caveat: Assuming that the election in question is a general election – one run with an official list of registered voters. For some set of elections, the set of eligable voters is sufficiently small that it is simple to validate the voters. For instance, one of the controls in the papal election is the fact that each of the cardinals knows each other.
Instead of reacting to the logic, reasoning and facts, Cynical spouts a knee-jerk reaction that doesn’t address any of the cogent arguments made in the posting.
Cynical — this response underscores the desperation in the republican’s case. Wouldn’t it be better to have saved the $10million both parties will end up having spent for Rossi to hear that he lost from the courts (just like he heard he lost from the voters)? Is that really good, conservative use of money? Even desperation has its limits….
How could a Florida judge find that it was OK to deny 1,400 voters their legal right to vote so that Republican anus-guards could prevent 119 illegal voters from voting. Mr. Cynical raises some good questions, he just doesn’t raise enough of them to expose his real motives— which are stictly partisan and have nothing to do with honesty or fair elections.
and democrats are FOR honest and fair elections? coughbullshitcough
Dem mantras
Vote early, vote often
Count every vote
If in doubt, call the vote a Gregoire one
Certify, smertify; we don’t need no stinkin rules
Smear and spin
Pretend Fla was stolen from them (and mention till the end of time)
Tight rules only disenfranchises the poor, the weak, the humble
we don’t need no stinkin rules
We have rules. You can find them in the RCW and WAC. Too bad that’s not good enough for you.
scottd: “We have rules. You can find them in the RCW and WAC. Too bad that’s not good enough for you.”
Those rules ARE good enough for everyone… except King County Elections officials, who failed to follow them (and have admitted as much in depositions).
Mark: Then Rossi should have no problem in court. All he needs to show is how laws were broken to unfairly deny him office.
I don’t think he’s doing so well in that venue, but what do I know? We’ll see soon enough.
I’ve already stated that I’ll accept the final result of the legal contest. Will you?
Scottd@8
You might try reading some of those rules and see how many of them were broken by K.C. that is point that righton@7 is making. We have rules and KC tossed them aside like they did not exist. The left can continue to spin but clear thinking people can see this election for what it is. I find it very interesting that the left is so quick to accept the admitted fraud committed in this case. Why are you not outraged? No need to answer that, we know the answer already. If it was Rossi that won the hand count and these problems existed and this Fraud was admitted to under oath, the left would be rushing the Governors mansion and try to remove Rossi themselves. You guys are a joke and have no principles. How you reconcile this in your minds is amazing to me. (actually if you have KC do it for you I am sure it reconciles just fine)
Clearly an idiot @ 11: Where have I accepted fraud in this case?
If it’s true that Nicole Way improperly or fraudulently filled out the absentee ballot report, she should be fired and maybe prosecuted. Election officials should put in place procedures to make certain such a thing never happens again. I feel the same about other problems found in King County and other election jurisdictions.
If Rossi can show that these mistakes, errors, and misconduct harmed him, he has a case. If not, tough luck.
And if you don’t like those rules, feel free to work to have them changed. What could be fairer than that?
BTW: I’m not clear where you stand on the role of law in our society. Do you think we should resolve disputes in court on the basis of law? Or do you think we should just settle things based on who whines most effectively?
Manipulating statistics is the strategem of last resort for desperate losers, and that’s what the GOPers have been reduced to. Yes, there were illegal votes, but who did they vote for? I don’t know. Neither do you. That’s the point. You aren’t allowed to guess, you need proof. That’s DJ’s and Professor Adolph’s point, and undoubtedly will be Judge Bridge’s ruling as well. The Republicans have been dishonest about this election contest from the get-go, and nothing exemplifies their dishonesty more than the attempt by their lawyers to admit evidence of illegal votes for Gregoire while blocking evidence of illegal votes for Rossi.
You are amazing…let me summarize your position.
“We know that laws were broken and that rules were ignored and that this election is totally f’d up but prove that you won.”
Analogy for you Scott;
You and I run a 100 lap race; we don’t have to run on the track at the same time. I run for a while and take a break as needed and you run for awhile and break when you need it but whoever runs 100 laps the fastest wins $ 1,000,000.00. The rules states that the race officials will be responsible for documenting our times and laps. We complete the race. On numerous occasions the race officials “accidentally” miscounted laps. They also forget to write down my times for laps 15, 34, 58 and 72 and your times for laps 10, 20, 25 and 30. Someone thought I only ran 99 and another 103. Someone said you ran 98 but another said you ran 104. But any way you look at it no one shows a total of 200 laps being run. So what they do is make all their lap tally sheets balance and add in our “average“ time into the laps they left the time out of and certify me the winner by a total of 1.5 seconds. I get the $ 1,000,000.00. Prove that the Errors and admitted falsifying of the Lap Tally sheets cost you the race and $ 1,000,000.00. That is quite the burden of proof. It is very likely that you won the race but I might of as well. We will never know because the race officials screwed up and cheated. But you can’t prove that you would have won so I keep the $ 1,000,000.00. Nice……wanna race?
Scottd
It is KC that has a problem with rules and laws not I. To think that the Rossi side has to prove that the admitted fraud in this election cost him the race is ridiculous. You are ok with; I cheated but prove you would have won if I didn’t. The report used to certify the election is fraudulent and that is not grounds to toss out the result to you? I admit that we don’t know if this changed the result but you cannot base an election result on fraud / falsified documents. They may not be able to prove a change in the result but they can prove the result was based upon fraud, which invalidates the result.
Tell it to the judge — or, tell Rossi to tell it to the judge. Really, that’s why we have courts.
I’m not a lawyer, nor do I have access to all of the facts and records in this case. Neither do you. From where I sit, it seems like Rossi is doing a poor job of presenting his case, so he’s going to lose. Maybe it’s because he has no case — but I’ll leave that up to the judge to decide.
Does KC have a problem with the law? I don’t know — see above. But it’s pretty clear that you do because you’re unwilling to accept the result from our legal process unless it yields the result you desire. I think you need to brace yourself for a long season of discontent.
(u)Clearly smarter @ 14
“We know that laws were broken and that rules were ignored and that this election is totally f’d up but prove that you won.”
It is not at all clear that laws were broken. If you read RCW 29A.60.200 carefully and slowly, you will see that the law demands timely certification even if the results are not completely accurate. The vote tallies must be known with “reasonable certainty”. The law’s purpose is to ensure timely certification, not ensure absolute accuracy.
RCW 29A.60.200 Canvassing board – Canvassing procedure – Penalty. … The county canvassing board shall proceed to verify the results from the precincts and the absentee ballots. The board shall execute a certificate of the results of the primary or election signed by all members of the board or their designees. Failure to certify the returns, if they can be ascertained with reasonable certainty, is a crime under RCW 29A.84.720.
So, Ms. Way and her boss Mr. Fell had a problem: “…a newly installed computer was unable to give a precise count of ballots returned” (Seattle PI).
What do they do? (1) They could hold up the certification, which would result in the Canvassing Board begin charged with a class C felony, or (2) they could estimate the numbers using whatever data were available. (A third option might have been to quickly hire and train a small army of counters.)
The law does not prevent them from taking option (2), so long as the final results are known with “reasonable certainty.” That makes (1) a reasonable way to proceed.
Notice that solution (2) did not change any aspect of the number of votes given to candidates—that is, there was no change whatsoever to the election outcome. All Ms. Way’s actions did was introduce a slight error on a tally form. It appears (from newspaper reports, anyway) that there was very good reason to estimate the number for the tally sheet: there was no better way to meet the requirements of RCW 29A.60.200.
All this episode really shows is that KC needs a better computer program for doing the absentee ballot accounting.
Oops, one sentence @ 17 should have read:
That makes (2) a reasonable way to proceed.
Swelled head at 14:
That’s a fun (and well-fleshed-out!) little thought experiment you came up with, but it’s not analogous to what happened in this election.
If you want a footrace analogy, try this: Gregoire and Rossi have just run a marathon, and are coming up to the finish line neck-and-neck (which is a surprise, because Gregoire was supposed to have left Rossi in the dust several miles back). They break the tape almost simultaneously; it’s going to be a photo finish! Rossi immediately claims victory; the crowd thinks he eked out the win. But after examining the photos, it turns out Gregoire actually finished first. Rossi is pissed off and his fans are livid. They start looking for technicalities that would require the race to be re-run. The camera’s exposure setting was slightly off, it turns out; the person with the light meter screwed up or fudged the reading. And Rossi’s trainer comes up with evidence that a gust of wind at the runners’ backs helped push them over the line, and asserts that it pushed Gregoire ahead of Rossi, although he can’t prove that. Rossi files a protest and the race organizers hold a hearing, but it looks like they’re going to let the race results stand.
Goldy,
So are you predicting that Rossi will lose?
BTW, I am still waiting for the sources and methods for the gas tax thread.
Spin spin spin. And then …..there is … Nicole.
Flip a coin. Seems like the conservatives have enough facts to build a case of gross neglect in vote counting and fraudulent reconciliation reports. But they can’t with integrity say that Rossi won because their cheerleader, Vance, repeats the mantra that nobody can tell who won.
Seems like the fair thing would be to flip a coin. Of course that will upset everyone here who thinks they know everything about the law.
Scottd@16,
“But it’s pretty clear that you do because you’re unwilling to accept the result from our legal process unless it yields the result you desire. I think you need to brace yourself for a long season of discontent.”
It is your side that is always mentioning Florida and whined about Ohio in that last election.
Oh yeah, there are over 3 years of bitternes left in your long season of discontent. Can you keep hate alive that long?
How does Nicole help Rossi?
pbj @ 23: I’m not sure who “my side” is. I’ve never said anything about Florida or Ohio. I’m all for investigating alleged impropriety, but absent a successful legal challenge, I’ll concede Bush’s victory.
Are you going to be able to do the same when Rossi loses his case?
Reply to 1, it seems to me that proportional analysis discredits itself. No thoughtful person can help but conclude it’s nothing but guessing.
David and his race comeback
The first 2 photographs showed Rossi the winner. So then they asked some guy sitting on the field who won. He called up all his colleagues, most of whom are alumni of the team Gregoire was running for. They couldn’t find all their notes, but some of the notes and stopwatches they held seemed to indicate Gregroir won.
And the guys who phoned into the guy trying to judge the race, couldn’t get through, so he chose not to use their input (militiary guys trying to submit ballots)
And while the guy (Logan) was muddling though the data, some people in stands came up and said they’d seen Gregoire win, so he added them to the judging panel.
Reply to 14, nice try but Gregoire doesn’t have to prove she won. The law puts the burden of proving she didn’t win on those challenging the certified result. Proving the election was a “mess” doesn’t prove Rossi won, it only proves the election was a mess. That’s not enough. You have to prove Rossi won.
Reply to 22, what law are you relying on for your coin flip proposal? As I understand the law, Gregoire has no obligation to submit to a coin flip; she is the governor until Rossi proves in court that he won the election, or until January 2009, whichever occurs first.
Reply to 27, I have a better analogy. Let’s say Rossi and Gregoire play a game of football. Let’s say Rossi at the half Rossi is ahead 14-10, at the end of the third quarter Rossi is ahead 14-13, and the end of the fourth quarter Gregoire is ahead 16-14. Two questions: (1) How many games were played? (2) Who won?
Scottd@25,
This isn’t really about Rossi or Gregoire, it is about the intergity of the system. Currently there is no integrity when people commit fraud and knowingly file false election reports. To dismiss it as a thoughtless mistake or forgetfullness shows a disregard for fair elections.
First of all Bridges said he cannot order a new election and he cannot install Rossi as governor. So even if Rossi “wins” he doesn’t automatically become governor.
Part of me is all to happy to let Gregoire continue down her radical liberal path because the ruin at the end of the road cannot be blamed on anyone other that the Democrats. Oh yeah, there were Republicans who voted for the burdensome Gregoire gas tax and trust me, I am working at the party level to get rid of them with even more zeal than Democrats because an enemy within is alway worse than an enemy without.
This trial is only the beginning. We will probably be following this through to the SCOWS where I am well aware of the political makeup and that the Democrat court will probably squash it.
If the outcome is that is gets rid of the crooks at King County elections, then it will have been worth it.
Yeah, you lefties run the risk of getting trapped like we righties got trapped by Ollie North, etc. So blind to preserving your cause you overlook the dishonesty, badness of what is going on.
If you pulled back, i doubt you’d want a know fraudulent system
pbj @ 31: I think Bridges said he could either find that Rossi had actually won the election, or that the election was invalid, depending on the evidence and circumstances.
In the first case, Rossi would become governor. In the second case, there would be no governor and Brad Owen would become acting governor until the next regularly scheduled election. There is some confusion over whether this would mean Nov 2005 or Nov 2006.
The thing Bridges said he could not do was order a new election — i.e., the re-vote Rossi so lamely suggested. Bridges said there is simply no provision in the law for such a thing.
So, Rossi is not without recourse. He can either show that he actually won and take office. Or, he can show that mistakes and misconduct appear to have caused him to lose, which would invalidate CG’s election and put Owen in office and give Rossi a chance to try again in the next general election.
His problem is that he can’t seem to demonstrate either of these — so we’re left with perpetual whining as the likely outcome.
Or take office and call for some Emergency bill. Legistlature knows how to do those.
Patrick seems to think that well over 1000 ILLEGAL VOTERS VOTING can be identified and validated and there is nothing that can be done. Ludicrous!
Once secret ballots…legal & illegal…are COMINGLED, you must have some reasonable method of removing those illegal ballots.
The alternative is akin to allowing ballotbox stuffing to occur and once thru Accuvote…what, no relief????
You’ve gotta be kidding Patrick.
Care to place a $100 wager on where some form of proportional analysis will be applied to the identified ILLEGAL VOTES???
I’m willing to place that bet with you.
We’ll settle up via Goldy.
Is it a bet Patrick???
C’mon…put yer money where yer mouth is!!
Reply to 35, are you offering to wager your money or BIAW’s? I understand you process a lot of $100 bills for BIAW …
Get serious, Mr. Cynical. Using statistics to GUESS is not PROOF of anything! It’s not my responsibility to suggest how Republicans should prove who the illegal votes were for. They’re the ones trying to overturn the election, not me. The law requires them to PROVE Rossi won! I don’t know how they can do that, and I don’t care, but I do know that I’m not willing to have the candidate I voted for removed for office based on someone’s wild guess.
I’m not saying there isn’t a statistical method that would work. However, I don’t think assuming felons vote the same way as other people in their precinct is valid. Felons are a distinct population with their own demographics and voting patterns. I think you have to look at how felons vote. Also, any analysis that just looks at illegal votes in King County is a load of bull. You have to look at the whole state. It’s absurd that GOP lawyers want the court to only consider illegal votes for Gregoire while ignoring illegal votes for Rossi. I guarantee you that Judge Bridges won’t go for that, or if he does, he will be reversed on appeal. Want to bet on that?
Uh, are you trying to convince us that his case is hopeless or yourself?
Patrick @ 29,
Flipping a coin is contemplated in Washington law. Although I was just half kidding, out of frustration due to the mess that has been made of this election. I suppose that such close scrutiny of election process is like watching sausage made.
The law allows the Secretary of State to “decide by lot” the results of a statewide election. RCW 29A.60.221(2). It is technically not a remedy for a legal contest to the election but rather a remedy only for a tie. Other states have similar provisions.
Patrick–
You went from criticizing proportional analysis to reframing the whole issue. Of course the Judge will allow Dems to bring in their ILLEGAL VOTES. I never said they couldn’t. However, it was not the R’s job to do the Dems work. The Judge will deal with whatever hard evidence is presented and validated to his satisfaction….then apply a reasonable proportional analysis formula.
It seems you have changed your tone Patrick to worm out of a clear bet. I knew ya would!!
Patrick,
Stranger things have happened.
If the judge finds that exactly 129 votes for Gregoire were illegal, then you could have a tie (and possibly draw lots per RCW 29A.60.221(2)).
If (and I know it is a lot of “ifs”) the judge rules the election a tie, then who knows what the judge will do? Conventional wisdom say that the Lt. Governor would take over. I happen to know and like Brad Owens (a Democrat) and think he would do a fine job.
Whatever.
Liking it @ 42
“If the judge finds that exactly 129 votes for Gregoire were illegal, then you could have a tie (and possibly draw lots per RCW 29A.60.221(2)).”
First, the Judge will not overturn the election because there are 129 illegal votes identified. This would assume that all the illegal votes went for Gregoire—something that is exceedingly unlikely. Some number on the order of thousands of illegal votes must be identified to overturn the election using a proportional analysis, since the illegal votes must be apportioned among Rossi, Gregoire, and other candidates.
Secondly, in a tie for Governors office in Washington, a joint session of the Legislature decides who gets the office. This means that Gregoire would have been elected Governor in the event of a tie.
pbj @ 31 you talking about integrity is like GWB talking about his ‘honorable’ military service to our country. What a JOKE
I’m getting really sick of Rossi’s “Revote WA” spam emails. Somehow I got on their list and they absolutely refuse to remove me. I’ve put “unsubscribe” in the subject, I’ve asked them in the email, I’ve tried to be removed now three freakin’ times but they still keep spamming me with their lies. This is a perfect example of the way Rossi has treated Washingtonians and I for one can’t wait until he loses the lawsuit. Maybe then they’ll finally stop sending me mail I never wanted in the first place.
Dave@45–
Simply delete it when it comes.
If you don’t have something to bitch about you aren’t happy.
So you take some little minimal thing and act like someone is about to stab you or something.
Typical LEFTIST PINHEAD.
Someday on your tombstone it will read:
HERE LIES DAVE…
DAVE DIED AN ANGRY LEFTIST PINHEAD.
Oh fuck off, Cynical. You don’t think I -have- been deleting them? I don’t want the email. I don’t want to be spammed by your golden boy, Rossi. But I have a plan if they email me again, and they’re not going to like it.
Dave–
Wow…you have a plan!
Is it a S-E-C-R-E-T Plan Dave??
It’s probably a plan so secret You don’t even know what it is dumbass!
Get a life Dave–
Get a life.
it’s amazing to see the word fraud continually used–fraud by whom? Fraud is an intentional misrepresentation designed to achieve a result other than that which would naturally occur. Who did that? Perhaps individual voters will be charged with fraud; that has nothing to do with the election contest itself. For it to have any bearing on the contest, it will have to be official fraud–and so far there’s not a bit of evidence to support it, either in public record or in allegations by Rossi. Legally speaking, if Dean Logan came out today and said he put 1,000 Gregoire votes into the machines himself, I’m not sure it would help Rossi–because the causes for contest needed to have long since been entered. What’s in there now, is what they have to go on. The illegal alien votes were stricken for this very reason–they came too late.
On proportional analysis–great to see statistical methodology put to good use. Two points: the first flaw Adolph points out may not be entirely relevant, legally speaking. Rossi has put forth X number of illegal votes, and the law says if those illegal votes can be shown to change the election, it will be set aside. It says nothing about an appraisal of all potential illegal votes, known and unknown. So to say that because we know that Rossi’s tally of illegals is neither complete nor representative, seems neither here nor there.
On the other hand, a point that I hadn’t yet considered is much more valid: their numbers are too small to apply inference to. I had approached the flaw from the other side of the equation–the larger population (all votes by precinct) was not known to be enough like the illegal votes, to apply the former’s result to the latter. But this makes more sense: in any given precinct, the number of illegal votes to consider is so small, that assessing a proportion of ANY kind is necessarily fraught with sample bias. I had mentioned this before in passing, but not anywhere near as coherently. If you have 1,000 illegal votes in a precinct, the sample is large enough for an applied proportion to be meaningful. If you have 100, there is more room for error, but at a bare minimum you might be able to roughly extrapolate some conclusions about the pattern. But 10 votes is just not enough to apply percentages to. The difference between 54.9% and 55.1% is one full vote at the 10-vote level, and a 20% increase (ie, from 5-5 votes to 6-4, 6 being 20% more votes than 5). How much sense does it make to apply a .2% difference at the precinct level, as a 20% difference at the actual vote level? Not much.
Thanks for the report. I may steal the link and expound on it myself at A2 after going through it.
tj–
You are grasping at straws….
It’s too late dude.
Read the R’s pre-trial filings.
It’s still a broad case with many avenues.
The Judge will methodically look at each allegation…judge them seperately…then onto the next.
In the end he will compile ALL issues and either the R’s have enough or they don’t.
Have you read the R’s pre-trial tj??
It’s posted on SP.
By the way, Stefan is awesome at posting links to enitre source documents so we can read…and decide.
yes, I have. They do not mention fraud. And issues are not compiled; votes are. Rossi doesn’t have enough.
torridjoe @ 50
“the first flaw Adolph points out may not be entirely relevant, legally speaking. Rossi has put forth X number of illegal votes, and the law says if those illegal votes can be shown to change the election, it will be set aside. It says nothing about an appraisal of all potential illegal votes, known and unknown. So to say that because we know that Rossi’s tally of illegals is neither complete nor representative, seems neither here nor there.”
Rossi put forth X invalid voters, but cannot say how they voted. Therefore statistical methods are called upon.
If the court accepts use of a statistical method, then the selectivity bias in the Republican’s list IS very important. The statistical methods inherently assume that the invalid votes are represented in the proper proportions–that is, they are a “random sample” of all invalid votes in the state. This isn’t at all optional statistically. Professor Adolph’s, and even more so, Professor Handcock’s analysis make this very clear.
Of course, judicially, it may be another story, but Courts are supposed to use valid science when they resort to scientific methods.
It’s not optional to the question of whether they’re representational, but it IS optional to the question of whether it matters viz those actual votes. For legal purposes, what I’m saying is that the Judge may consider those which are presented, as the total universe of felon votes. That he might believe others exist out there, wouldn’t change his consideration of the votes presented as the only ones he can rule on. To consider non-appearing felon votes, even if we assume they’re there, means to accept as evidence their existence, which he can’t do if they’re not presented. He said circularly, but you see where I’m getting at.