With Labor Day and the traditional start of the fall Presidential campaign looming, Weld reminds readers that national polls showing one candidate or the other ahead are essentially meaningless. The “national” Presidential election is decided in the Electoral College, through the results of 50 statewide elections, with 270 electoral votes needed to win.

With that in mind, we are pleased to inaugurate (pun decidedly intended) this weekly capsule tracking the projected outcome in the Electoral College. This projection is based on various polls in each state, updated according to rolling monthly averages to provide a snapshot of the likely outcome week-to-week. Our results are weighted to account for verifiable shifts in one direction or another and to minimize the impact of “outlier” polls with results that skew illogically against an established trend.

Each week, we’ll award votes in a tiered system that identifies every state as either Locked, Likely or Leaning to one candidate or the other, or else a Tossup. In determining the cumulative weekly vote totals, tossup states will be awarded based on which candidate is leading a given state, no matter how small the margin.

This week’s projected Electoral College count: Obama 332, Romney 206

Locked for Obama (154 votes): California, DC, Delaware, Hawaii, Illinois, Maryland, Massachusetts, New York, Rhode Island, Vermont, Washington

Locked for Romney (117 votes): Alabama, Alaska, Arkansas, Idaho, Kansas, Kentucky, Louisiana, Mississippi, Nebraska, North Dakota, Oklahoma, Texas, Utah, West Virginia, Wyoming.

These states comprise the respective baseline of electoral votes for each candidate, with little or no chance of being remotely competitive in the fall election.

Likely Obama (37 votes): Connecticut, Maine, New Jersey, New Mexico, Oregon

Likely Romney (64 votes): Arizona, Georgia, Indiana, Montana, South Carolina, South Dakota, Tennessee

These states are somewhat competitive, but unlikely to move from one candidate to another. Current trends suggest that Connecticut and New Mexico are shakiest for Obama, while Arizona, Georgia and Tennessee could prove problematic for Romney.

Leaning Obama (56 votes): Michigan, Minnesota, Pennsylvania, Wisconsin

Leaning Romney (10 votes): Missouri

These are states in which one candidate or the other currently has a clear lead, but in which the numbers are volatile enough to change in one direction or the other numerous times before November. Current trends worth mentioning are those toward Romney in Minnesota and Obama in Missouri.

Tossups (100 votes): Colorado, Florida, Iowa, Nevada, New Hampshire, North Carolina, Ohio, Virginia

These states will decide the election. Currently, Obama leads in Colorado, Florida, Iowa, Nevada, New Hampshire and Virginia. Notable trends are toward Romney in Florida, Iowa, Ohio and Virginia and Obama in North Carolina.

Thumbnail analysis: Obama has a margin for error that Romney does not. Indeed, if current trends hold, Romney could carry Florida, Ohio and Virginia and still lose the election 272-266 — though if he also added Iowa, he’d win by the same margin. In short, the presumptive GOP nominee has to hold everything he has and pick several states off the Obama bandwagon. That’s going to be tough.