<div dir="ltr"><div dir="ltr"><div>Thanks Kristofer! Your analysis is really helpful for me as I try to explain this to a broader audience than the folks on this mailing list. I strongly suspect that 99.999% of people who aren't on this mailing list have no idea what the term "pushover" means, and now I suspect that most people on this mailing list can't define it unambiguously. I've started a stub article here:</div><div><a href="https://electowiki.org/wiki/Pushover">https://electowiki.org/wiki/Pushover</a></div><div><br></div><div>I'm hoping all y'all on this mailing list help improve the new "Pushover" article.<br></div><div><br></div>I found the part in the "Tactical voting" page's changelog where pushover was defined relative to monotonicity on electowiki:</div><div dir="ltr"><div><a href="https://electowiki.org/w/index.php?title=Tactical_voting&diff=prev&oldid=3014">https://electowiki.org/w/index.php?title=Tactical_voting&diff=prev&oldid=3014</a></div><div><br></div><div>It would seem that <span class="gmail-mw-userlink gmail-mw-extuserlink gmail-mw-anonuserlink">James Green-Armytage (or someone assuming that name) defined "pushover" in terms of monotonicity back in 2005.<br></span></div><div><span class="gmail-mw-userlink gmail-mw-extuserlink gmail-mw-anonuserlink"><br></span></div><div><span class="gmail-mw-userlink gmail-mw-extuserlink gmail-mw-anonuserlink">A prior definition that was briefly on electowiki shortly after it was copied over from English Wikipedia states the following:<br></span></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div>'''Push-over''' or '''turkey-raising''' is a type of strategic voting in which a voter ranks a perceived weak alternative higher, but not in the hopes of getting it elected. For example, in a [[bloc voting|bloc vote]] where multiple votes are required, a voter may insincerely vote for a candidate they perceive as unlikely to win, in order to help their preferred candidate win.</div></blockquote><div><br></div><div>See <<a href="https://en.wikipedia.org/w/index.php?title=Strategic_voting&oldid=10141683">https://en.wikipedia.org/w/index.php?title=Strategic_voting&oldid=10141683</a>> to see the version on English Wikipedia at the time it was copied. I fear it was a mistake to define "pushover" in terms of monotonicity on electowiki, since it seems the "pushover" term is used for "turkey-raising" in some contexts (as I'm assuming based on Chris Benham's email and the contents of English Wikipedia in 2004 and 2005).</div><div><br></div><div>Rob<br></div><div><br></div></div><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Sun, Oct 1, 2023 at 6:08 AM Kristofer Munsterhjelm <<a href="mailto:km_elmet@t-online.de">km_elmet@t-online.de</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex">On 10/1/23 10:44, Rob Lanphier wrote:<br>
<br>
> This claim you make is interesting:<br>
> "[STAR] somehow doesn't 'violate monotonicity' and yet [...] is more <br>
> vulnerable to Pushover than RCV (aka IRV) which does.".<br>
> <br>
> Is that true? It seems to me that RCV's series of runoffs lead to many <br>
> opportunities for weak candidates to snowball via transfers from <br>
> eliminated candidates. The snowball effect in RCV usually snowballs to <br>
> the center of public opinion, but can sometimes roll toward the <br>
> outskirts as candidates get eliminated and their ballots get transferred <br>
> to a stronger and stronger candidate on the outskirts. With STAR (and <br>
> Score), I believe the candidate needs to have strong support from all <br>
> voters to get a high enough score to advance (since all ballots are <br>
> considered in the runoff round), but perhaps similar polarization can <br>
> occur under STAR over time. It's truly an interesting question which <br>
> method is more susceptible to pushover.<br>
<br>
I think this problem is about how to interpret pushover. Mono-raise IIRC <br>
comes in these two forms:<br>
<br>
1. If you raise A and A goes from winning to losing, that's a failure.<br>
2. If you lower A and A goes from losing to winning, that's a failure.<br>
<br>
Suppose A is the winner in STAR. Then raising A can't bump him off the <br>
top two who advance to the final, nor can it reverse A's pairwise <br>
victory over the other finalist B.<br>
<br>
Similarly, lowering A's score can't get A into the top two if he wasn't <br>
already, nor can it turn B>A into A>B. So STAR is monotone.<br>
<br>
I can see two ways to interpret pushover. The definition from Electowiki is:<br>
<br>
"Push-over is a type of tactical voting that is only useful in methods <br>
that violate monotonicity. It may involve a voter ranking or rating an <br>
alternative lower in the hope of getting it elected, or ranking or <br>
rating an alternative higher in the hope of defeating it."<br>
<br>
A strict interpretation considers "defeating it" to mean "turn the <br>
candidate from winning to no longer winning". That interpretation thus is:<br>
<br>
1. If you prefer B to A, A is winning, and you raise your ranking/rating <br>
of A with the intent of having the result change from A to B, then <br>
that's pushover strategy.<br>
<br>
2. If you prefer B to A, A is winning, and you lower your ranking/rating <br>
of B with the intent of having the result change from A to B, then <br>
that's also pushover strategy.<br>
<br>
I.e. the candidate you're altering the position of must be either the <br>
candidate who's winning or the candidate you want to win. By this <br>
interpretation, pushover implies monotonicity failure, because if <br>
raising A made A lose, that's a failure of the first kind, and if <br>
lowering B made B win, that's a failure of the second kind.<br>
<br>
STAR does not have this particular type of pushover.<br>
<br>
But here's a looser type of pushover:<br>
1. If you prefer B to A, A is winning, and you raise your ranking/rating <br>
of some other candidate X with the intent of having the result change <br>
from A to B...<br>
<br>
2. (same as #2 above)<br>
<br>
then STAR *does* fail. Suppose B beats X pairwise but A beats B <br>
pairwise, and the finalist set before strategizing is {A, B} so that A <br>
wins... then by increasing your rating of X, you might bump A off the <br>
set so that it's {B, X} instead, after which B beats X and wins.<br>
<br>
The "pied piper" strategy seems to be closer to this type than the <br>
strict interpretation. A is the mainstream Republican, B is the <br>
mainstream Democrat, and X is the outrageous Republican. By supporting <br>
X, the Democrats intend to induce some Republican voters to "rank or <br>
rate X higher", i.e vote for X rather than A in the primary. The <br>
intended effect is to knock A out, which leads to the general being <br>
between B and X, where B then (presumably) wins.<br>
<br>
(But not if X is Trump: then you get a backfire.)<br>
<br>
Strictly speaking, a monotone ranked method could also have this type of <br>
pushover strategy, e.g. a voter voting:<br>
<br>
B>A>C>D>E>F>X<br>
<br>
leads A to win, but<br>
<br>
B>A>X>C>D>E>F<br>
<br>
leads B to win. But because the strict version implies nonmonotonicity <br>
and ranking X higher is often accompanied by A being ranked lower, it's <br>
associated with nonmonotonicity for ranked methods. I'm not aware of any <br>
monotone methods with this kind of failure.<br>
<br>
-km<br>
</blockquote></div></div>