Thanks all for the additional information and opinions. Whether competitive play is at the Matchplay or Medal format, I still think it makes more sense to try and collect as much data as possible in an effort to more closely define anyone's "potential" scoring capabilities, which is what the handicap system attempts to do so I'm all for enforcing the hole by hole score entry, even if some consider it a minor inconvenience that takes maybe another 30 seconds or so.
Ok, this is half-baked, but let me offer an alternative to spur discussion. It's a straw man, so I won't be offended by any criticisms. It's based on the predicate that most amateur handicaps, even those of better players (but mostly of high handicappers) are often distorted by a few bad "disaster" holes that artificially elevate scoring overall. It's also based on a belief that very few players even know what ESC is, much less use it, so hole-by-hole scoring takes that out of the equation.
The concept is simple. Instead of total medal score total up the number of holes a player achieves some number indicating "proficiency", which I'd argue translates directly to "potential". For arguments sake, lets say any hole where a player scores 5 or better (it could alternatively be total pars, or bogeys even) before ESC is applied counts towards their handicap and the other's are strewn to the wind.
In this example, let's say an 18 index has an average round (you could include all scores instead of just the best 8 of 20) of 95 but it's made up of 3s, 5s, 6s, 7s, and 8s and such. The average number of holes where they score 5 or better becomes their index. Let's say it's 4 holes per average.
Compare that with a 5 index who's average round is let's say 80, but who cards 5 or better on an average of 15 holes per round. In a match between these two hypothetical players the higher handicap would receive 11 strokes, as opposed to the 13 they would normally receive. This would in effect reduce the impact of those disaster holes on "potential" for all players as well as eliminate the need for ESC and some other complexities that are neither well understood nor universally applied.
Someone smarter than me can likely easily figure out how to adjust for slope and course ratings to apply to this method.
Darts and arrows welcomed, thanks.