Is it possible to fraudulently elect one federal office without throwing all elections on the same ballot into question?

Question

Background

There is no shortage of 2020 Election Fraud conspiracy theories in the United States, and there was no shortage of attempts to overturn the final result of the Presidential Election. None of the processes that the United States uses to adjudicate matters like these led to any substantial changes (if any) of the final results.

One premise that came to mind may have acted as a double-edged sword had any of the lawsuits been successful. This premise is the reality that several elections for office are held on the same ballot card. Because multiple elections are held on one ballot, I would assume that if it was discovered that one election was fraudulent, then all elections on that ballot would be contaminated as well. However, this mechanical reality of voting never seemed to be addressed in the many allegations of fraud that were made...which would seem to be a vital premise to at least address, lest other elections be held under scrutiny that the accuser did not intend to hold under scrutiny.

However, because it was not explicitly addressed, it seems that the situation described in the second paragraph wasn't the case. It is this crucial scenario, and the lack of addressing it that led me to ask the following question:

Question

If multiple elections are held on the same ballot, how could alleged voter fraud in one of the offices not lead to consequences for the other offices on the ballot?

Answer 1

There are a couple of ways how one race could be subject to fraud; the implications on other races vary.

The simplest (and probably easiest to detect) option would be simply dumping additional votes that are only marked for that one race. This would be detected by an extraodinary number of votes for the race in question compared to other races which could be correlated to other places which would not have suffered from fraud. In such a case, maybe in most counties the ratio of presidential to House votes is 1 : 0.95 but in the counties where fraud is alleged it rises to 1 : 0.8 or 1 : 0.7. It would then be clear that there is only one race that is outlying and that this race would be the one affected by fraud; this race would then be subject to scrutiny while all other races would have a high likelihood of being confirmed.

On the other end of the spectrum, one option would be to tamper with the counting machines’ software (in a way that affects all counting machines in the state) leading them to count a couple of votes the other way for every set that passes through (potentially if certain other thresholds are met). I believe this type of fraud method would fly under any statistical radar that could be used in other cases (but also requires far more criminal energy by the fraudsters) and would probably only be detected by a full hand recount. This recount would then again obviously show one race out of line with all the others – unless the fraudsters took it one level further and fraudulently altered the outcomes of all races in the same manner. Here again we have a possibility of detecting fraud in only one single race and thus leaving other races intact.

Finally, in at least one case fraud was alleged based on the discrepancy between the results of the presidential race and those of down-ballot races; this allegation is based entirely on the assumption that all the other races were fine but only the presidential race was tampered with. The allegation is easily debunked on mathematical grounds but do note that the method used could be applied individually for each race and – if it were used in a mathematically sound manner – may be used to detect fraud in just one single race as opposed to fraud across all races on a ballot.

Obviously, however, there are also a number of possibilities how tampering with one race would lead to all races being called into question: if the fraud occurs by ballot stuffing but with fully filled-in ballots (as opposed to the example at the top where only one race is targeted) this obviously calls the entire election into question.

Answer 2

Yes, it is very possible to do that but proving that there is voter fraud and it only impacts one race on the ballot is another question. If the fraud in question is submitting/counting ballots that only vote for that one race or only changing votes for that one race it could be possible.

That being said if you find voter fraud in an election how do you prove that it only impacted one race (or the races you don't like the outcome in)? In the end for a lawsuit like that to be successful not only do they have to prove voter fraud in the races in question but they have to prove that the same voter fraud didn't impact the other ones.