Overnight, the number that actually moved on our board wasn't the bid count. It was the gap between what agencies think a job costs and what contractors are willing to do it for.

Low bids came in an average of 3.7% under the engineer's estimate across the 1,000 results we're tracking. A day earlier, that same average was running 5.5% under. That's a two-point swing in one overnight cycle, and it's moving in the direction of higher prices relative to what agencies budgeted.

“The floor came up. Bidder counts didn't move. Somebody priced closer to the number than the trend line predicted.”

Bidder counts didn't move — still 9.1 bidders per job on average, same as yesterday. The spread between low and second bid held flat too, at 16.3%. So this isn't a story about fewer contractors showing up or the field getting tighter. It's a story about the floor coming up. Somebody, somewhere in this last batch of results, priced closer to the agency's number than the recent trend line would predict.

Here's why that matters if you're building an estimate this week: for months now, this desk has been telling you low bids are running well under what agencies pencil in — a gap that's good news if you're an owner watching a budget, and a signal if you're a contractor deciding how aggressive to get. A 5.5%-under environment rewards sharpening your pencil. A 3.7%-under environment gives you a little more room to price in your actual risk and still land competitive.

Two points isn't a full trend reversal — one more cycle in either direction will tell us whether the market is genuinely tightening or whether this batch of results just happened to include a few jobs nobody wanted to chase hard. But if you've been shaving margin to stay under engineer estimates on public work, this is the first read in a while suggesting you don't have to shave quite as hard. Watch the next few cycles before you bet a bid on it, but if you're pricing this week, don't assume last month's discount is still the market.