To set the range in Excel, right click on the axis and you can format the axis and select range tab to manually input the range you want.

This is an interesting post. Not only has the winner take all screwed Romney, but of course factoring in the Huckabee factor shows just how soft McCain's "support" has been. I have been thinking about how winner take all might have been benefiting the Republicans in that they can get past the challenge earlier than the Dems, but obviously that's not really true in this case.

Interesting, but with different rules people would adopt different strategies and, therefore, we cannot assume that the same results would have occurred.

I haven't heard much about the effect proportional allocation has on turnout, but it should help. Even if you think your candidate is going to lose, your vote can still help them.

There are lots of reasons Democratic primary turnout is so high this year, but I wonder if this has any effect. Maybe most people aren't aware of the delegate rules.

I agree with Josh, above, but do think this shows that McCain isn't quite as strong a candidate as we might think.

So Romney should have run as a Democrat, and Clinton as a Republican? You know, I think that would have been better for everyone.

These charts would have been a lot easier to grasp if red and blue meant the same thing in both charts. And if blue stood for the Democratic rules and red for the Republican, in both.

Of course, if the Republicans really used the Democratic rules they would also have super delegates. Wouldn't that be interesting!

On the one hand this is interesting. On the other hand, it's chasing at phantoms. If the Democrats had gone with the winner-take-all rules, then the candidates would have run different kinds of campaigns.

Maybe the real point here is that if all the Dem primary contests were held to the same standard, the delegate counts would be more even. Instead, caucuses mean that when Obama wins, it's winner-take-most, but when Hillary loses, its split proportionally.

I haven't heard anyone yet argue that, IN FACT, the Idaho, Washington, and Colorado caucuses reflected the choice of the entire Democratic electorate in the same way that the primary votes in California or Missouri reflected the choices of Democrats in those states.

"....but when Hillary wins, its split proportionately"

This makes me think about the way the media explains victories -- you get this sense that the basic analysis is that candidate x won because he's a winner and candidate Y lost because he's a chump. And while there's truth to the idea that better strategic decision making (and of course the avoidance of catastrophic decisions) contributes to victory, a lot of the time, it's luck. Candidates don't know for certain what's going to work. Nobody does. So to the extent that Mitt Romney succeeded in winning over as many voters as McCain, the idea that John McCain's primary victories are more important than the polling data about his party's lack of enthusiasm for his candidacy is wrong. I don't know much about how the enthusiasm of the base affects elections, but if it's very important, John McCain is not a strong candidate in the general election.

"Meanwhile, it's a bit odd that, for both parties, switching to the other side's rules would make the race more, rather than less, competitive."

I'm not sure it's that hard to explain. The competition is happening within the existing framework, so the campaigns are working to optimize their strategies for that framework and ignoring the other one. If you hypothesize that the frontrunners are frontrunners because they are more skilled at campaigning, then it makes sense that their lead would collapse if you compare metrics against which they weren't competing.

Alternatively, one might argue that the order of the primaries creates different advantages for different types of candidates depending on the delegate apportionment rules. WTA gives a big advantage to the strong starters, who tend to be media-favored or establishment candidates and social conservatives. Proportional counting gives more importance to later states, and thus gives underdog candidates more opportunity to campaign their way out of their underdog status. The current frontrunners got there by capitalizing on their structural advantage, but with the rules reversed the structural advantage flips and the gains the frontrunners made by working the system don't count as much and the underdogs get a boost.

SBG almost had it right.

To set the axis scale, either right-click on the axis you want to set, or double-click on it (I believe you are a macophile, yes?) If you right-clicked, select 'Format Axis' to bring up the proper window. If you double-clicked, the right window will pop up.

Then, go to the 'Scale' tab. You can set the minimum and maximum values of your axis there.

Happy Graphing!

On the one hand this is interesting. On the other hand, it's chasing at phantoms. If the Democrats had gone with the winner-take-all rules, then the candidates would have run different kinds of campaigns.

Doubtful.

Mark Penn's statement about states that matter (big ones like CA) versus ones that don't (every state Obama won minus IL), pretty much tells you everything you need to know.

That strategy makes the most sense in a winner-takes-all environment where second place doesn't matter. Winner-takes-all provides further justification for trying to wrap up the January primaries and the delegate lead with a big Super Tuesday.

It makes less sense in a proportional system where blowouts in small states and competitive second place finishes in the big ones keeps you statistically in the race.

Far more relevant is the amount of superdelegate votes. Proportional distribution of delegates makes it somewhat more easy for a second place candidate to keep campaigning. Superdelegate votes provides a rationale for staying in even though the candidate does not loses the primaries.

My takeaway from this thought experiment is that the Obama campaign has a clear understanding of the rules on the ground and how best to exploit them; that the Clinton and Romney campaigns do not and did not understand those conditions (and have wasted huge amounts of money in futile misapplied strategies); and that McCain is really, really lucky that Romney's campaign sucked.

(And, really, folks, all this is is a thought experiment; to get all riled up about how the respective campaigns would have done things differently under different conditions is to miss the point entirely.)

Meanwhile, it's a bit odd that, for both parties, switching to the other side's rules would make the race more, rather than less, competitive.

What exactly is odd about it? Changing the rules for the sake of competitiveness seems wrong to me. Its pretty clear that awarding delegates by proportion is much more representative of the electorates preferences, caucus's aside. The only way it would seem odd is if you think that inherently everything that Democrats do is right and Republicans wrong and hence its ironic that if switching the rules, there would competivness for its own sake. And we know you don't think that, sooooo....

Or CFMan, that campaigning had less to do with it and the results are just people's preferences. You are right, complete thought experiment.

campaigning had less to do with it and the results are just people's preferences.

I'm inclined to disagree. How they campaigned altered the race.

Clinton enjoyed double digit leads in national polls. In state polls, she enjoyed big leads in states Obama eventually won by wide margins. Obama's caucus victories are flat out a result of their campaign decisions. His big wins in the Potomac are entirely due to, yes, their campaign decisions.

Voter preference is only static in the absence of stimulus. When faced with an active campaign, it becomes dynamic.

Because I don't know how to control the range Excel places on the Y axis

Dude, seriously, that should be an embarrassing thing for a college graduate to admit. Take a short course or tutoring or something.

Perhaps the greater Romney/McCain parity under proportional representation has more to do with WTA not being a terribly effective method when there are more than two passably viable candidates. Due to this, the great disparity between the systems seen on the Repub side wasn't mirrored on the Dem side, as Edwards dropped out earlier and never won a single primary or caucus, compared to the continuing presence of Huckabee.

this is exactly why Rudy made sure NJ, CT,NY, and DE became winner take all states: he knew it would skew his advantage (back when he thought he was a contender) becuase he knew he was never going to be dramatically popular within his party. McCain reaped the rewards on Super Tuesday of Rudy's efforts.