The problem is with temp stratification and locating the probe in the mash. Lets say the mash gets cool, so the then the pump starts pumping and the heating element heating... by the time the mash gets to the correct temp lets say near the MLT outlet, the top of the mash is too hot, because it takes time for that hot wort to trickle down through the grainbed. I have found that a differential of 4F will give me the desired temp in the mash. If I am mashing at 155F as I am today, I set the HERMS coil to 159F and run the pump during the entire mash to keep the entire mash at the same temp. Otherwise I would be heating the mash when PART of it got cool, and dumping heated wort on top, making the top of the mash say 159F while the temp probe sat at 155F. In terms of this particular system, locating the temp probe in the HLT allows me to monitor my strike water temp and allows me to heat my mash much more consistently than placing a probe in the MLT.