From Arrow Springs. In short, glass will not retain thermal stress below the strain point (around 840 for COE 104).
http://arrowsprings.com/html/annealing.html
"The cooling of glass is most important between the annealing temperature and the strain point. As explained, glass will develop stress in itself through the cooling process. The strain point is a point in temperature at which any stress that develops below that temperature in the glass through the cooling process is only temporary. Stress that develops in the glass above the strain point is permanent. Once the glass has stabilized to room temperature, temporary stresses will disappear. Because of this fact, you can accelerate the cooling time below the strain point temperature and not worry about this strain causing the glass to break at some time in the future. However, cooling at too fast a cooling rate can still break the glass from thermal shock while still in the annealer. The strain point for glass varies between manufacturers and even between different colors from the same manufacturer. If you use a temperature well below the strain point of all the glasses you use to cool slowly down to before then increasing the cooling rate, you will not need to know the exact strain point temperature of each individual glass. Use 750º F for Satake and 800º F for every thing else."