Later-No-Harm Method Calculator
by Kevin Venzke [Home]

This is a calculator for the (quite few) known methods which satisfy Douglas Woodall's Later-no-harm criterion:

In the context of a rank ballot voting method, Later-no-harm is the principle that a voter shouldn't be penalized if they add additional candidates (i.e. less desired compromise choices) to the end of their ranking, as opposed to leaving them out of the ranking entirely. We imagine on this page that any runoff (for TTR in particular) is conducted as an instant runoff, relying on voters' submitted rankings.

Later-no-harm is the property, proposed by Douglas Woodall, that a voter should not be at risk of harming their most preferred choices by adding additional choices. Thus the voter should hopefully feel safe in listing as many compromise choices as they may have.

The property is perhaps best known or most associated with IRV, as a purported advantage of that method. As a result, sometimes Later-no-harm itself is criticized (by opponents of IRV) using arguments that happen to be dependent on the specific way that the property is achieved under IRV.

If this is an error, it may be hard to discover this without awareness of other methods with the Later-no-harm property. Such awareness is likely to be rare. Woodall's DSC for example has only been proposed in an unpublished draft paper (2003) (although it is related to his earlier method DAC, which was proposed in Voting Matters in 1996). Some other methods have only been "published" on the Election Methods mailing list.

Another motivation is an interest in seeing if more Later-no-harm methods can be devised. The site can't help with this directly, but maybe it can inspire curiosity in others. It seems very difficult to invent new methods of this type. But the known methods do give a variety of results, as I think this calculator shows. That leads me to suspect that there may be some other interesting methods yet to be discovered in this space.

When an election is tied, I show (e.g. with color) all the possible winners in the main table, but there may still be some potential for confusion in the presentation of other details. Below the table, some more in-depth information for IRV, DSC, and Chain Runoff can be displayed.
Woodall proposes a counterpart to Later-no-harm called Later-no-help. Note that only FPP, IFPP, TTR, IRV, and ACP satisfy both properties. All the other methods violate Later-no-help, which means voters may sometimes assist a higher-ranked preference by adding additional preferences.
The concepts of Condorcet, Schwartz, and Woodall's Plurality criterion are included. I won't explain them in detail here, except to say that Condorcet/Schwartz methods are primary competitors for Later-no-harm methods in the "rank ballot space." And the Plurality criterion is included because it shows a special defect of a few of these methods, where in some cases a winner may seem to win the election based on other candidates' weaknesses, or some other odd technicality, rather than on their own merits.

Enter your ballots like this, one per line:
456: Alice>Bob>Carl=Debra
or equivalently to the above:
456: Alice>Bob
The number represents the size of the voting bloc. Decimals are OK. The size can also be left off and it will then be randomized.
Note that this calculator doesn't allow equal ranking above the bottom. This is because not all of these methods handle that kind of ballot, and the methods that can support it often don't really incentivize voting that way.
Candidate names can contain spaces. Each candidate in the list should be separated by > or =. Pipes (i.e. |) and any series of > will be interpreted as single >s. Not every candidate needs to be listed; candidates present on the ballots but missing from one faction's ranking will be interpreted as ranked tied for last, below any explicitly ranked candidates.

Click submit to generate an analysis.

Enter the ballots for an election:

Flag Counter