BP officials have told the Wall Street Journal they should be able to stop the flow of the disastrous Gulf oil gusher by July 27, weeks earlier than had been projected. But BP shouldn't be surprised if the collective reaction is: Just tell us when it's done.

BP has lost its credibility on promises and projections. The company still hasn't adequately addressed a question that should have been answered at the start: Just how much oil has been spilling into the Gulf?

Given the fishy estimates so far, it's a wonder anybody believes anything about the ongoing disaster.

The first estimate after the Deepwater Horizon explosion on April 20 pegged the oil flow at 1,000 barrels a day. Not long after, the estimate rose to 5,000. It was raised again to somewhere between 12,000 and 19,000, then 25,000 to 40,000 and, currently, 35,000 to 60,000.

That final tally is subject to change — for the worse, if the past is any guide. Some skeptical independent experts peg the output at 100,000 barrels.

It's easy to imagine the technical difficulty of measuring a runaway well under 5,000 feet of water, but come on. At one point, BP reported capturing more oil each day than supposedly was escaping.

Such obvious baloney engenders public anger and cynicism. Americans need information they can trust from BP and their government. Both parties have disappointed.

The damage goes beyond mere disillusionment, because to some extent the flow rate influences the cleanup. It's a crucial input for oceanographic models that predict where oil will arise. How much will wash up onshore, and where? What's the best way to deploy the fleet of skimmers? Knowing the flow rate can help clear up those mysteries.

The first 1,000-barrel estimate came from the oil giant, which had a practical reason to lowball it: The amount of oil spilled typically determines the size of federal fines and damages in related court cases. The second estimate of 5,000 barrels, from the Obama administration, depended on aerial observations. It fell way short, giving the American public a false impression of the disaster's vast scale.

The larger official estimates came from the Flow Rate Technical Group, a collection of scientists the federal government belatedly appointed to analyze undersea videos, sonar data and pressure measurements that increasingly have become available. Its latest report is scheduled to appear shortly, though the bottom-line numbers aren't expected to change anytime soon from 35,000 to 60,000 barrels.

Eventually, government-appointed experts will make a final determination of the amount spilled. That can't come soon enough, and the Obama administration should push harder for a precise and plausible reckoning. Federal watchdogs also need to audit BP's claims about how much oil is being collected: It would be in the company's interest to inflate those recovery figures.

Unfortunately, the early mischaracterizations of the spill's magnitude stand to undermine the credibility of any future tally.