Correcting for the Conservative Nature of Protection

Using all of my suggested corrections for half-point, you will on average undercount your hand by about .35 points. This is about .40 for hands less then 10 HCP and .30 for hands larger than 10 HCP. So technically, you can add .35 to your hand for using this system. I am painfully aware that it is awkward to add 1/3 point to your hand. If you add 1/2 point to your hand, you are systematically overvaluing your hand by .15 HCP. I estimate this is not a problem. On the other hand, my intuition is that the 12 HCP opposite 12 HCP that is 50% for 3NT does not use this final correction.

If you and your partner have 23 HCP, the opponents have 17. If you stop short of game because you have 23 points counting half-points, but you have 25 points before half points, then the opponents have only 15 HCP. So, on the face of it, you are no stronger in the second situation than you were in the first. However, the opponents are weaker.

This is potentially a small but real problem. In practice, this has not been a problem for me. I think the resolution is that the half-point corrections take into account that the opponents have more. Obviously, I cannot know for sure. But there was nothing scientific about saying the correction factor was 1/2 point. That was just a very convenient number.