I am currently writing to explain the energy differences Pauling was trying to explain with his theory of electronegativity. There is a paradox one must deal with. According to Pauling theory, bonds contain ionic and covalent character. If polar bonds are stronger than covalent bonds, NaCl has a high heat of formation and a high melting point, we can think they are strong. Yet, ionic bonds are easily broken by dissolving in water. We may think hydrogen bonds are weak, yet you can drive your car on ice. Ultimately, bond strengths are going to come to charge and distance considerations. When that happens, bond types disappear. A covalent bond to t-butyl iodide may be weaker than an ionic bond to calcium fluoride. Protons are positive and electrons negative. If a proton can come in close proximity to an electron pair, its attractive force can be greater. Presumably this is at least in part what is happening in ice. It is also forming a matrix of bonds. A matrix can be much stronger than an individual bond. A bridge of toothpicks can be much stronger than an individual toothpick.
I can imagine an enzyme-substrate complex being more like a matrix than measuring an individual bond. If so, the combined effect of many seemingly weaker bonds may result in great affinity. Also, even seemingly weak hydrogen bonds (or any other interaction) becomes stronger with the inverse square of the distance.
The affinity cited in the referenced link are more like irreversible inhibitors. Even the substrate should inhibit the enzyme. However, if the reaction proceeds, the transition state can become the product. The binding of the product must be much lower in order for the enzyme to turn over the substrate. That must happen.