Montgomery County, Maryland


There’s a bill to allow approval voting or ranked choice voting. It appears that the state needs to pass the bill to permit the county to do this, then the county would choose which system to use. Note this is where Rob Richie lives, in Takoma Park, Maryland which already uses IRV.

Here’s a piece which seems to be largely FUD against the bill, so it only mentions IRV and not approval voting.


The MoCo delegation to the Maryland legislature has introduced bills like this before (I know for sure they did one last year). Those previous bills only allowed RCV (it may have been vague enough to allow for a non IRV form of ranked voting, although IRV is the only ranked system any local media mentions). So this is an improvement over last year. Of course, if the bill fails again, it’s a moot point.
Actually, I wrote a letter to a local news website (which they published) advocating for the use of either Approval or Score voting in local elections. It cited the then-recent Democratic County Executive Primary as a case, the same election Luedtke did (Synopsis: Extremely anti-development councilmember wins with 29% of the vote because of vote splitting between 5 opponents. There were other problems, too, but that’s the gist of it.) I’d like to say that the letter had something to do with this development, but I don’t know for sure.

I’m surprised you didn’t comment on the other Luedtke bill in the article, which would make ballot access harder.


Last year’s bill:


Status Update:
The Education, Elections, and Housing Delegation Committee has voted unanimously to give the bill a favorable recommendation (with an amendment that does not affect the voting methods covered).
Next it will go before the entire county House Delegation and, if it is recommended, the county Senate Delegation. If it passes the Delegation, it will get its first reading.
Delegation approval of local bills is not required for a first reading, but bills that fail are customarily withdrawn. The delegation committee vote on this bill was delayed, so the bill has been introduced for first reading before the entire delegation will have a chance to vote on it, so that it can be scheduled for a state committee hearing.
The status of the bill can be viewed at:
and also at


How did this happen in the first place? It’s quite an accomplishment to get anything other than RCV before the legislature!


Status Update:
The House Delegation has voted unanimously to give the bill a favorable recommendation. I am posting a link to the video of the meeting, since there was some debate on the bill preceding the vote. Notably, quite a few of the problems raised were related to voter understanding (one delegate recalled a person at a past hearing commenting, “it’s hard enough to help my grandmother vote already.”), or to infrastructure, such as the difficulty of counting/recounting ballots, and compatibility of equipment. Of course, these are much more of an issue for IRV than an Approval.
The portion of the video that pertains to the bill is 29:29-44:47.


Notes on the public hearing:
The fiscal policy note was posted a couple of days before. It estimated that RCV would cost the county $1.7M and the state $162,000, and that Approval would cost the county $1M and the state nothing. This is a drop in the bucket, but it is consistent with the idea that Approval would be an easier change than RCV. That said, opponents of the bill often cited the cost as an issue.

Several local NGOs came out to support the bill, but it was all focused on “ranked choice voting” (which they of course conflated with IRV specifically). They generally used the generic arguments that sound good to people who haven’t thought about it much, such as ‘it eliminates the spoiler effect’ and ‘it ensures a winner with majority support’. They didn’t really mention why IRV and not Approval (or some other RCV counting algorithm). The delegates themselves were aware that there were other ways of counting ranked ballots, though.

The “RCV increases turnout” argument raised eyebrows.

The opponents were generally Republicans who thought that the council would just use the ability to choose the voting method to give themselves an advantage, citing the (admittedly severe) gerrymandering as reason not to trust them. (Although often gerrymandering was used in a whataboutist Some of them were pretty unhinged, though. For example, one compared the amendment to hold a campaign to educate voters about whatever system gets implemented to “reeducation camps in North Korea”. “One person, one vote” was fallaciously invoked. Some were concerned about the money (which was a drop in the bucket, of course). A lot of them were focused on how complicated tabulating IRV is. “I challenge [Delegate Luedke] to explain RCV in 3 sentences.” Approval didn’t really come up at all.


Seems Asset wouldn’t have money issues? Just let candidates mail in or congregate and vote their preferences for negotiations and vote-trading.


The money was an excuse. The projected cost to the county for RCV would amount to 0.04% of the county’s total expenditures in 2018. The projected cost for Approval would be 0.023%. (And the extra cost decreases to zero after a couple of elections.)

Anyway, some of the cost goes towards educating voters about what’s changing, which is inevitable regardless of what reform you use.


I’d expect more of a problem with this for Optional Asset - regular Asset seems simple enough to be understood almost at the ballot box.


My preferred version of Asset only lines up with yours in the single-winner case. In the multi-winner case I allow each voter to cast as many votes as there are winners. (That, or just use “pie voting” – mark n candidates and each of them gets 1/n votes. That has problems with getting confused with Approval though.)

That would require some education, but as long as pie voting and approval voting are not used in the same location, it should be fine. Besides, there will probably be pamphlets passed around about the new voting systems.


The issue with this is that in the multi-winner case, only one winner represents any voter, not all of them. If you’re looking for maximal voter control over where their vote goes and all, or a hybrid, Asset’s not the best. I’d point to Fractional Approval Asset Voting, which you call pie voting, but it’s not simple enough, and benefits few voters. We know from cumulative voting that nobody likes to split points up.


Unfortunately, it is essentially guaranteed that the bill will not pass this year, as the crossover deadline (the last date for each chamber to pass its own bills) is on Monday, and the committee has yet to act. The leader of a local IRV advocacy group whom I contacted said that the committee may vote to conduct a study on the bill in between sessions. This might not be a terrible outcome because it would necessitate direct comparison between Approval and IRV (and possibly some other ranked choice methods, since the bill doesn’t specifically define RCV.)