Why are you doing this?
Well, mainly because I have fun doing it. I don’t have any illusions about being the next Nate Silver or anything. But I think my little version of this adds yet another perspective to look at the presidential race, and more is better, right? But mainly, I’m a news junkie, and presidential elections are an every four year frenzy for that sort of thing. And I enjoy crunching numbers, doing analysis, and making graphs. So this is a natural fit. Mainly though, I enjoy it.
Seriously, we already have 538, PEC, The NYT, The Washington Post, and all kinds of other folks doing this. What do you add?
Well, all those places are great. I eagerly consume everything they do and respect all of those efforts quite a bit. I’ve added to and enhanced what I do here based on things I’ve learned from them. (Specifically, the “tipping point” stuff is inspired by PEC’s meta-margin.) All those models can get quite complex though. They factor in many things, and they do much more robust mathematics than I do. One of my “points” in doing this the way I have since 2008 is that even with a very simple model, just a plain old average, nothing fancy, you can do very well.
Also, while a few of the other do this too, I concentrate a bit more on monitoring the ebb and flow of the race and looking at the trends over time rather than just what the prediction was today for November.
And of course, my views and analysis are always purely electoral college based. When I started in 2008, it was because I had been frustrated in 2004 that it was hard to get views of the race based exclusively on the electoral college, instead everybody kept talking about the national polls. But that isn’t how we elect presidents!
Is this a prediction?
No. Not at all. At least not yet. Everything here is essentially an “if the election was held today” snapshot. But the election is not today as I’m launching the site in 2014. It is a long way away. Things will change. Events will happen. The two parties will actually select their candidates. Maybe as we get into the Fall of 2016 the data here can start to be reasonably interpreted as a prediction. Before then, it is not. It is simply a description of what things look like today, which can still tell you quite a lot.
Why a five poll average? Why not six? Or ten?
This is actually fairly arbitrary. It was chosen to be high enough that you could get a reasonable average out of it, but low enough so that the time frames covered by the average are short enough to be responsive to changes in the campaign… at least closer to the election in the close states… when we are far from the election… or for uncompetitive states right up through the election… the average can cover so much time it won’t respond much to change. But what happens is that as the election approaches, close states are polled more and more often, so by the time of the election, the five poll averages in the close states generally only cover a few days.
But why a number of polls? Why not an amount of time?
Well, once you include the backfill from old elections, you can always fill out the average. With any timeframe, you will often hit the situation where you have no polls at all in your chosen timeframe. Then what do you do? There are obviously answers to that question, but you have to start having special cases. Having an “X poll average” is just simpler and more straightforward.
How about some fancy curve fitting or smoothing instead of an average?
Oooh! That is fancy! Initially I didn’t consider this because I was calculating everything with a simple Excel spreadsheet and it was a little daunting to think about. Now that I’ve automated all sorts of things, it wouldn’t be that hard to plug in whatever logic I wanted for computing the “average”. It has occasionally tempted me, but then I remember that one of my points in doing this is to show how very simple logic can do just as well as the complicated models, so I decide not to add complexity if I don’t need to.
OK, but how about a median instead of a mean? Wouldn’t that be better?
Maybe. Sam Wang at PEC uses medians. They are more resiliant to outliers. They definitely have some advantages. I have thought about it, although there are probably downsides too. In the end though, I’ve used means for the last two election cycles and it has worked pretty well, and I’d like to be able to compare results between election results to some degree, so I’ll stick with the mean.
Fine. So what about these categories? Aren’t they completely arbitrary?
Yes. Yes they are. I consider any state where the margin is less than 5% to be “weak” and thus subject to the possibility of going either way. This could be some other number. 5% just seemed like a reasonable amount that you could imagine either polls being wrong, or for a last minute movement in the race that could happen too fast for polls to catch. After the last couple of elections, I considered narrowing this a bit, but ended up sticking with 5%.
Do you do any sort of simulation that gives odds of winning?
No. At the moment I just provide the “possible” band and the tipping point margin. The band between the best cases roughly means that anything inside that band is within the realm of the possible, and anything outside of it is not. It of course isn’t that simple. It is far less likely that every single close state goes to one candidate than the states are roughly split in the way indicated by who is ahead in each state.
I have considered modeling this and doing a Monte Carlo simulation and generating “Win Percentages” and such. But to do so, I have to basically toss out the core methodology of just categorizing the states into broad bands, and instead have to start looking at modeling uncertainty based on the available polls and other factors.
This is of course a valid thing to do, but it is a fundamentally different way of structuring things than the basic premise of my breakdown. If I did this, I’d probably want to reassess almost everything else about how the site’s analysis is done. So for the moment I have decided not to pursue this direction. If you want that, there are plenty of other sites doing exactly that. (Or there will be by election time anyway.)
How do you decide what polls to include?
I basically am not picky. I include just about everything I find. I’d have to have a really good reason to exclude something. I don’t try to evaluate pollsters by their methodology, previous accuracy, or anything like that. I just throw everything into the mix and see what comes out.
Doesn’t this take a lot of work?
Yes. Hopefully it will not be quite as much of a slog in the run up to the election now that I’ve automated a large part of the grunt work. But just scanning the web for polls, entering them, and writing commentary when things change will still suck a significant number of hours. But referring back to the first question, it is fun for me, so that isn’t that bad.
Aren’t you starting this a bit early?
Well, again, it isn’t a prediction. Predictions this far out probably are fairly useless. But the first 2016 polls actually appeared BEFORE the 2012 election. And there have been new polls dribbling out pretty much constantly since then. They may still be pretty sparse, and it is still a very long time from the election as I officially launch this site in November 2014, but in terms of trying to understand how things are shaping up, it is NOT too soon to be paying attention. The folks that do groundwork and manage and fund campaigns have been paying attention and making decisions based on what they see in the wind… including polls… for more than two years already. No reason why the rest of us shouldn’t be looking too.
Who are you anyway? Are you even qualified to be looking at any of this?
A long time ago I was a Physics major. Then somehow I got into the world of the web, and eventually ended up doing product management at a tech company in Seattle. This election stuff is purely a hobby. I do it for fun. I do not have professional credentials in this area. I can’t claim to be an expert. I’m just an amateur putting this together in his spare time. I think the results have been pretty good so far, and I’m looking forward to the 2016 version. If you ARE an expert and have constructive criticism for me, I’d love to hear it and learn from you. Just don’t be mean. :-)
Shouldn’t you maybe get a life? Maybe go outside or something?
Nah, I’m good.
What do I do if I have questions or comments on all this?
Well, just add a comment to one of the blog posts about the election on Abulsme.com. Or email me at firstname.lastname@example.org. Or tweet at me via @abulsme or the dedicated twitter account for these election updates, @ElecCollPolls.
How do I follow along with what is changing with this analysis?
For up to the minute updates when polls are added, when states change categories, when tipping points change, or information of that sort, follow @ElecCollPolls on Twitter. When changes worth noting more widely occur, I will be making blog posts on Abulsme.com. To see just Election 2016 related posts, look for the tag. My cohost Ivan and I will undoubtedly also be speaking more and more frequently about the election on our podcast as we get closer to November 2016.
Edit 2014-11-06 05:43 UTC – Fixed typo
Edit 2014-11-07 21:50 UTC – Fixed incorrect word
Edit 2016-05-14 19:55 UTC – Changed first sentence to refer to election graphs.com instead of the older election2016.abulsme.com, added link to Delegate Race FAQ.