There have been volumes written on how the presidential debates might impact the election. With their conclusion last week the polls have shown a significant impact.

Our simulation model which uses the latest state wide polls has been run each week since the beginning of summer. Its estimate of the probability of the President winning reelection peaked at 99.9% on September 28 when the electoral college math showed there was no likely way that Governor Romney could garner the necessary 270 electoral votes.

But the campaigns continued, the Benghazi attack was still making headlines, the stock market dropped, and the debates began.

With the first debate came performance critiques of the two candidates and new state polls. By the end of that week the race had tightened. Our model predicted that Governor Romney had gained in several key states primarily Florida, Virginia, and North Carolina. With this surge his probability of winning also rose – moving from 0% the week before to 11% on October 5.

The following week provided an opportunity for the Vice Presidential candidates to debate. As the debate was late in the week they could not have had significant impact on that week’s polls. Instead the new polls were more likely a continuation of the results of the first debate. On Octaber 12 our model showed a continued decline for the President with his probability of reelection dropping to 82%. At the same time Governor Romney’s probability of winning rose to 16% and for the first time in several months the probability of an electoral college tie reached significance at 2%.

The effects of the Vice Presidential debate and the third Presidential debate could be seen in the following week’s polls. As of October 19, Governor Romney’s chances continued to rise reaching a probabilty of 25%. Similarly President Obama’s probability of winning continued its slide dropping to 73%. The probability of a tie held at just under 2%.

Now the debates have been concluded and the candidates are back to the criss-crossing the country giving stump speeches in the battleground states. But with this return to the traditional campaign has also come a return to President Obama’s reelection hopes. As of last week our model shows the chance of reelection reversing, climbing from the previous week’s low back up above 92%. Similarly Romney’s chances have declined to about 7% with the probability of tie again dropping below 1%.

Why the reversal? A noticeable change has occurred in Florida. Romney had taken significant leads in the Sunshine State over the past several weeks. While still leading Florida has again tightened. Similarly Romney’s Virgninia lead has dropped to a tie. And in Ohio, a key for a Republican victory, the polls show that while Obama had been able to retain a slight lead throughout most of October, his lead has increased over the last week.

While the model does not indicate that reelection is a certainty, it does indicate good news for the President. Our model currently predicts an electoral college vote of 301 for President Obama and 237 for Governor Romney – a buffer of only 31 votes – a slim margin in electoral college math. But with only a week to go, and early voting and absentee voting already in progress, our model indcates the GOP may have a significant challenge to get those 31.