If you’re following the latest political polls, you probably think you know which candidate is going to win the 2020 Presidential election.
But as with anything in the realm of politics, you shouldn’t forget about the politics.
When assessing a poll on any candidate or issue, it’s important to know how the polls were conducted, what the parameters were, and so on.
All too often, you must also make the distinction of whether or not a survey constitutes a Democratic poll or a Republican poll or a Biden poll or a Trump poll, etc.
Polls always have an inbuilt bias (or ten), and they always have.
Political polls today are no different than they’ve been in the past, and you need to a hefty grain of salt to make them palatable. A typical poll’s margin of error is often larger than the historic voter gap between parties themselves.
Polls are, at best, questionably useful.
Fortunately, a better metric for assessing the electorate’s trend exists: political betting odds.
Now, if you’ve been following the 2020 election cycle, you’ve seen how much these odds change through the course of a given day or week, as they reflect the electorate (or the betting subset of the electorate) in real time. But this doesn’t make them less accurate – it makes them more accurate.
And though not perfect markers for election outcomes, betting odds seem to have a better track record when it comes to more accurately predicting results.
Why?
Allow us to explain.
The following are the top 11 reasons why betting odds are better than political polls:
11. Average bettors are better informed than average voters.
Most people (i.e. those who respond to polls) consume media – including news media – from entirely within their own bubble.
Generally, consumers consume only what they want, and most people favor those outlets that tend to agree with their points of view. When a person is polled, their responses will reflect this bias (because it is their bias).
Bettors, on the other hand, typically consult more sources, many of which are in significant opposition with one another. One of the first rules of betting is to be as informed as possible, and you can’t do that if you stay inside your own echo chamber.
10. Bettors strongly consider how others will vote.
Politics is a popularity contest, and there are plenty of people who vote with the herd. But that herd is selective: friends, family, acquaintances, etc. Thus, when a voter goes with the group, it’s not entirely organic, as people mostly associate with those sharing their same basic values and beliefs.
Bettors, on the other hand, put more than a little credence into how others are voting across the entire spectrum. This includes strangers, rivals, and opponents.
A voter will pick the outcome they support, not necessarily the outcome they believe has the most support.
A bettor will pick the outcome they think is measurably more likely to occur.
9. There are no demographics in betting.
When someone is polled, they’re chosen at “random” based on a series of criteria that often include age, gender, income, voting history, party affiliation, faith, and – of course and above all – race. The parameters of any given poll use these as markers, and the results can be skewed bigly across any of them.
Assumptions about the voting habits of men vs. women, the rich vs. the poor, Christians vs. non-Christians, the old vs. the young, and whites vs. blacks vs. browns are baked into the polling methodology.
But with betting, things like gender, age, wealth, and religion don’t matter. And the only color anyone cares about is green.
Bettors come from all walks of life, and the lines reflect a consensus that isn’t predicated on any variable other than that consensus.
8. The lines show where the support is.
For pollsters, the entire point of a poll is – allegedly – to show where a candidate’s or issue’s support lies. But all too often, it’s the poll that lies.
The bias of a polling agency or its partner media conglomerate comes through in the questioning, and they’re looking for pre-determined results that fit a narrative.
One of the big knocks against polling is that the institution isn’t used to demonstrate actual support but is instead used to gin up support for an outlet’s chosen position.
As an example, for the 2020 Presidential election, the polls funded or reported by CNN show Joe Biden leading Donald Trump yugely, while the polls funded or reported by Breitbart show a much narrower margin.
Polls are used as a tool to promote a given vote, not as a measuring stick to assess votes in a vacuum.
Betting lines, on the other hand, have no such in-built bias. Vegas election betting sites aren’t funded by PACs and – in actual fact – have no relationship whatsoever with US governance. You get a betting line with no loaded language, and you pick a winner. That’s it.
As such, smart campaigns – and smart voters – will look at the lines for a better reflection of what reality tells us, not what we try to tell reality. That’s a big difference.
7. Betting is nationwide.
Betting action, at overseas election betting sites serving the US marketplace, is almost entirely nationwide. Most sites accept players from the vast majority of states, which casts the widest possible net.
Thus, the moneyline on a given candidate or issue reflects what the nation as a whole (or, at least, its informed bettors) are thinking the most likely outcome will be.
That said, this one is firmly in “double-edged sword” territory, as such bets may better inform campaigns on whom or what is likely to win the popular vote more so than what may yield an actual electoral victory.
But on the other hand, given that bettors are more politically savvy than rank-and-file voters, they likely understand that Presidential elections aren’t won via popular vote, so the argument can be made that the nuance of the electoral college and its machinations are built into the odds.
The main point, for those gauging a particular election outcome, is that betting lines show the most comprehensive possible picture of empirically measurable popularity.
6. Betting samples are massive, polling samples are tiny.
Most polls only survey between 500-1000 respondents, and wider projections are extrapolated from that small pool. While some of the biggest political polls of 2020 have sampled upwards of 40,000 people, these are time-consuming outliers.
The general rule is that polling samples are very small.
However, the odds posted at the best political betting sites are the live, real-time product of hundreds of thousands of individual wagers.
In general, with any statistical science, the larger the sample size, the more veracious the conclusion. By this metric, there’s simply no question about which grouping is more reflective of popular sentiment.
5. Betting is voluntary, polls are solicited.
This should probably be higher on the list, because it’s an extremely crucial difference with fundamentally important implications.
When a poll is conducted, respondents are chosen based on a number of criteria (as outlined above). But beyond that, polls are aggregated from only those who actually participate in the poll, and you can only participate in a poll if you’re chosen to participate in it.
Never mind the fact that upwards of 90% of potential respondents reject a pollster’s solicitations – the fact that the response is solicited in the first place is enough to hopelessly skew the results.
Bettors, meanwhile, actively seek out sportsbooks with which to wager. Nobody is cold-calling them to advertise the betting lines. Betting is wholly voluntary.
At political sportsbooks, there is no “shy Trump voter” or “silent majority” angle to qualify or consider. People simply place their bets for whichever candidate or issue they want to bet for, at the behest of nobody and nothing other than their own bottom line.
4. Betting is live, polls take days or weeks.
The odds on political props bets and election betting lines are live, and they’re updated in real-time. As the money comes in on either side of a given wager, the odds change to reflect the current favorite and underdog (as well as the extent to which they are favorites and underdogs).
If you want a current snapshot of how things stand today – or this hour, or this minute – political betting lines are the only way to get one.
At best, polls take a few days to conclude, and they usually take closer to a week. Bigger polls can take even longer. And as we’ve seen countless times, a whole lot can happen in a week, especially in the information age of 24/7 “breaking news.”
3. People bet on who they think will win, not who they want to win.
This is one of the biggest reasons why betting odds are viewed as more accurate than political polls.
In most cases, when someone is polled, they will answer truthfully about what they want to happen or what they hope will happen, not what they think will happen. (Unless, of course, we’re talking about Trump voters, who are notorious for lying to pollsters in the first place, which is yet another point against polls and in favor of betting lines as a predictive institution.)
Yes, there is fandom in politics as in sports, and there are probably a few bettors who bet with their hearts and not with their heads.
But generally, political betting doesn’t suffer from any home-team skew. No matter how bad the Jets are, the moneyline at The Meadowlands will always pay out less than the same bet in Las Vegas.
In sports, betting is an ingrained part of rooting for one’s team. Political betting, in its modern accessible form, has not been around long enough to establish a similar pattern.
For now, it’s safe to assume that political bettors by and large choose the outcome they think is most likely rather than the one they’re personally pulling for.
This could change in the future as the fracturing of America increases along partisan lines, but for now, most political bettors are looking to make bank, not cheerlead for one side or the other.
2. Bettors are all “likely voters.”
We’ve saved the second best for second last.
Polls have a very difficult time assessing whether or not their respondents will actually vote in a given election. This is why all “credible” polls will explain whether or not their field of respondents are “registered voters” or “likely voters.”
In the United States, there are far more of the former than of the latter for any given cycle, as most people are registered to vote when applying for mundane things like driver’s licenses and so on. The push to register everyone in America to vote means it’s so simple and automatic to do so that the process borders on “opt-out” status.
“Likely voters” are a subset of registered voters, and different polls establish this in different ways. Sometimes, the benchmark is whether or not you’ve voted before (which is on record). Other times, it’s just something a pollster asks and takes at your word (which is not on record).
New and first-time voters are also difficult to nail down for pollsters.
That said, when it comes to political betting, all bettors are likely voters, and most of them are guaranteed voters.
Why?
Because when you’ve got money riding on an outcome that you can actually control in some way, you’re going to follow through to do everything you can to ensure that your bet hits. With political betting, that means voting.
Additionally, given that one must be 18 or older to bet on politics at any of the top election gambling sites, every wager posted has been posted by someone of legal voting age.
Betting lines ensure almost automatically what polling agencies work very hard on divining out.
1. People put real money on the line.
Ultimately, of course, betting lines are more accurate than polls for one very simple reason: Bettors are putting their money where their mouth is.
You can be as invested as you please in an idea, but when the rubber hits the road, if you’re not putting money behind it, your support is unreliable.
As stated previously, political bettors are not apt to put money on the outcome they don’t actually think is more likely. But the fact that they’re putting money on the outcome at all is enough to show that bettors are more serious than polling respondents as a general rule.
Campaigns, for their part, know this intuitively. It’s why they announce with great fanfare their quarterly donation figures. It’s why “internal polling” has a lot more to do with those donations – with what the money says – than with simple questionnaires thrust upon a marginally welcoming cross section of the populace.
Publicly, though, the only indication anyone has about where the money’s going is what the betting odds say.
To wrap up, it’s true that polls have their place, but they’re an admittedly imperfect science.
So, for that matter, is gauging election outcomes on betting alone.
Betting by itself is not a totally reliable indicator of how an election will go, but it’s a better and more reliable indicator than polling.
And there are caveats.
In the modern world of the 24-hour news cycle, bettors are starting to rely more on polls than they used to. That’s unavoidable, but it’s not yet saturated.
Indeed, future elections will be even better metrics of how much stock gamblers give to polls. In 2016, bettors favored Hillary Clinton, though not quite as strongly as the polls favored her. Those polls were proven comically unreliable.
If they prove similarly unreliable this time around, betting lines will become even more accurate as bettors put less credence in polling standards across the board.
Conversely, if the polls for the 2020 Presidential election turn out to be more accurate than last time around, polls and betting might blend to the point where no useful edge can be gleaned from the odds, as the latter will simply reflect the former in toto.
For now, though, political betting odds still seem to hold more water than polls in an apples-to-apples comparison.
Follow the money.
And you can take that to the bank.