(Bloomberg Opinion) -- People keep saying not to worry so much about the Covid-19 surge in Florida, because it isn’t killing a lot of people. I’m not buying it. And I’m sorry to predict that by early August, deaths in Florida will be almost as bad as New York at its worst.

Florida and New York have similar populations (21.5 million and 19.5 million, respectively), so it’s possible to compare the two states using raw numbers rather than rates. New York’s worst daily case count was 11,434 on April 15, and its worst seven-day average was 9,909 on April 10. Florida had 15,300 new cases on July 12, and a 7-day average of 9,957. In other words, by this measure, Florida is as bad as New York ever was.

But testing in the U.S. is inadequate, and hence misses a lot of cases. The percent of tests that come back positive offers a sense of how many we’re missing: The higher it is, the more we’re probably missing, because people with mild or no symptoms aren’t getting tested. Florida’s positive test rate was 18.6% as of July 12, which is pretty high but still short of New York’s 44% in the darkest days of the outbreak.

The difference in positive test rates suggests that New York’s maximum daily case count was more like 24,000 in Florida terms (meaning that if New York had been testing as much as Florida is now, it would have found about 24,000 cases). So Florida’s actual case count of about 15,000 probably isn’t quite as bad as New York in April. Still, it’s very, very bad.

What will this mean for deaths? That depends on how long fatal cases take to die. New York didn’t start counting cases until well after the epidemic had started, so recorded cases and deaths rose almost simultaneously. Florida has been counting cases for longer, so it’s easier to get a sense of the relationship. The time lag appears to be nearly a full month, in part because younger people – the ones who tend to go out and socialize in bars and so on -- have been getting infected first.

A real prediction requires a number and a time stamp. So here goes: I forecast that by Tuesday, August 4, Florida’s seven-day average of daily deaths will reach 600. And every week with a high case count –- an average of about 10,000, assuming the positive case rate stays at about 18% -- will be followed about four weeks later by another 600-plus-death-per-day week. Of course I hope I’m wrong, and that my forecast proves to be wildly high. This is a case where there will be zero satisfaction in getting it right.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

Cathy O’Neil is a Bloomberg Opinion columnist. She is a mathematician who has worked as a professor, hedge-fund analyst and data scientist. She founded ORCAA, an algorithmic auditing company, and is the author of “Weapons of Math Destruction.”

©2020 Bloomberg L.P.