As the Republican Party prepares for its convention, a look back at Election 2004 reveals an amazingly stable race for the White House. In fact, the monthly polling totals show virtually no movement in the campaign other than a minor bounce related to the Democratic National Convention.

In February, the month that Senator John Kerry became the Democratic Party frontrunner, President George Bush attracted 46.3% of the vote in Rasmussen Reports polling. Kerry, with 45.2% of the vote, trailed by just over a single percentage point.

For the full month totals in March, April, May, and June, the candidates were never more than half-a-percentage point apart. During those four months, Kerry's support stayed in a very narrow range from 45.1% of the vote to 45.6%. Bush also stayed in a narrow range from 45.5% to 45.9%.

In July and August, Kerry opened up a one-and-a-half point lead. July data generally reflects the period following the announcement of John Edwards as Kerry's running mate. August reflects the period following the Democratic National Convention. It seems fair to conclude that Kerry received a two-point "bounce" from these events (he was down by 0.5 percentage points in June, up by 1.6 in July and 1.5 in August).

The stability of the race is highlighted by the fact that for the entire seven-month period, Bush's vote total has varied by less than a single percentage point--from a low of 45.5% (May) to a high of 46.3% (February, August).

Kerry, even with his VP announcement and convention bounce, has stayed within a three point range for the entire seven months (a low of 45.1% in May to a high of 47.8% in August).

It is worth noting that Kerry has consistently polled better in the 16 key battleground states than he has in the national poll. During the month of August, Kerry leads 48.5% to 45.5% in the battleground states. The 3-point advantage is twice as big as his lead in the national polling.

Data for August reflects only the first 20 days of the month and will be updated in early September.

Rasmussen Reports interviews 500 Likely Voters each night for a total of approximately 15,000 Likely Voters per month. The theoretical margin of sampling error on a monthly basis is less than one percentage point. However, as with all surveys, other potential sources of error exist.