Like Gladys I'm not really a fan of using a model to predict exact numbers and dates, the inputs change too much. Ranges and trends I feel more comfortable with.
But if pushed I'd be predicting below 2000, in fact probably below 1500 but as everyone would know the numbers jump around a lot so a couple of bad days would change the prediction a fair bit (the trend not so much). When does it fall below 1, I'll be an optimist and say before mid September, possible as soon as 7-10 days. But once again, a few bad days would blow this out but probably only by a week or so, I think the general mid-Sept figure is not far off.
But if pushed I'd be predicting below 2000, in fact probably below 1500 but as everyone would know the numbers jump around a lot so a couple of bad days would change the prediction a fair bit (the trend not so much). When does it fall below 1, I'll be an optimist and say before mid September, possible as soon as 7-10 days. But once again, a few bad days would blow this out but probably only by a week or so, I think the general mid-Sept figure is not far off.