This model creates a probability we are in a recession using four monthly coincident economic indicators: non-farm payroll employment, industrial production, real personal income excluding transfer payments, and real manufacturing and trade sales. These are four of the six indicators that the NBER uses to determine if we are in a recession.
Historically this model doesn't have a high false positive rate once the probability is above ~10%. Generally when this model has three consecutive months above 80% we are in a recession, while three consecutive months below 20% generally mark the start of a new expansionary cycle.
This model isn't forward-looking. It just evaluates the present probability of a recession. We can use leading indicators to forecast the risk of a future recession. For what it is worth, the leading indicators suggest a future recession is likely.
The actual FRED release isn't updated yet. Note this is a smoothed probability model, so previous months can change with the new releases.