What the Betting Markets Tell Us About The Second Debate

In the wake of a videotape of Donald Trump making crude and vulgar sexual remarks about women and the second presidential debate, political analysts are rushing to guess how these major campaign events will affect the race. The two scientific post-debate polls say Clinton won (CNN by a wide 57–34 margin while YouGov by a smaller 47–42 margin). However, the CNN poll also found that Trump exceeded expectations. Additionally, the Luntz focus group suggested that Trump came out ahead.

One way to analyze the impact of the 2nd presidential debate on the race is to follow the money.

How do Betting Markets Work?

The political betting market I use most often is PredictIt. Launched in late 2014, this site allows political junkies to buy shares of potential political ‘events’ at various prices. One event is “Which party will win NC in the Electoral College?” The current price for Democrats is 75%, suggesting there is approximately a 75% chance that the Democrats carry North Carolina. An investor can choose to buy a share for 75 cents — if the event resolves to a ‘yes,’ the investor makes one dollar. If the event resolves to a ‘no,’ the investor gets nothing. The investor can buy and sell shares at various prices as the market reacts to recent political news and polls. The idea is that supply and demand will give us a pretty good idea of where the race stands probabilistically.

Chances of Winning the Election

First, let’s take a look at how the #TrumpTape and the 2nd debate affected Clinton’s chances of winning over the weekend. In the following spreadsheet, I compare the probabilities of Clinton winning on Friday morning (before the #TrumpTape was released), on Sunday evening (before the second debate), and on Sunday night (after the second debate). I collected data from three PredictIt markets, Predictwise, and Betfair. On average, Clinton’s chances of winning increased significantly (from 75% to 82%) because of the #TrumpTape, while the 2nd debate didn’t change the odds at all.

Chances of Winning Key Swing States

Next, let’s look at how recent events have impacted Clinton’s chances of winning six key states in the Electoral College. Although Clinton is well ahead of Trump in most of these states in recent polling, they are considered most important because they are the most likely states to put Clinton over the 270 Electoral College vote total required to win the election. The data here shows Clinton’s chances of winning each state at the three intervals.

In half of these swing states (Ohio, North Carolina, and Florida), Clinton was narrowly favored before the #TrumpTape, increased her chances significantly following the #TrumpTape, and roughly maintained that probabilistic level of support after the debate.

In the other half of swing states (Pennsylvania, Michigan, and Wisconsin), Clinton was heavily favored before the #TrumpTape and roughly maintained that probabilistic level of support over the entire weekend. This makes sense, since it’s harder to increase her chances in states where she is already the heavy favorite.

Chances of Winning Traditionally Red States

Let’s also take a look at how Clinton’s chances of upsetting Trump in traditionally red states have changed over the weekend. These are states where Trump has generally been considered the favorite.

Here we see that Clinton’s chances in many of these red states skyrocketed after the #TrumpTape before falling back down to Earth after the second debate. While states like Arizona and Georgia showed a reversion to Friday morning’s probabilities, other states like Missouri and Utah showed a significant increase in probability of Clinton victory vis-a-vis Friday morning. While the debate looks like it might not have much effect on who wins, Trump’s middling performance may have been enough to reduce the likelihood of a Clinton electoral landslide.

Conclusion

Overall, the betting markets suggest that the #TrumpTape significantly increased Clinton’s chances of winning the race overall. This is consistent with news over the weekend of numerous Republican elected officials jumping off the Trump Train and rumors that the Trump campaign might be finished. It looks as if Trump’s performance in the second debate may have stopped the bleeding, with his probabilities of winning overall and in key swing states leveling off and Clinton’s chances of pulling off stunning upsets in red states falling back down to Earth.

While the Trump campaign may be satisfied that the bottom is no longer falling out, Trump is in a pretty perilous situation. He is currently losing in two states he absolutely needs to win — Florida and North Carolina. The FiveThirtyEight election now-cast has Clinton 6 points ahead of Trump nationally with an 86% chance of winning the Electoral College. This is a pretty significant gap this late in the campaign — my guess is that Trump needs a major external event like a very damaging Wikileaks document release to have a shot. Luckily for Clinton, many people are saying that more damaging Trump audiotapes may be on the way.