I think your points above are very valid, Arc, especially the one you lead off with in your most recent post.
I believe the biggest potential consequence to implementing a combat system with hits and damage thresholds as you describe is a radical narrative change to the feel of the game if you don’t know what to expect. Entire classes of monster can become nothing more than dungeon dressing for the players to wade through, while whole categories of means of delivering damage become substantially less powerful in the game than under ICRPG Master Edition rules as written.
Part of the aesthetic that defines what Hank likes to call “vintage D&D” these days is the rough start. Characters may begin the game with a few boons in their pockets in terms of gear, skills, and abilities, but dungeon delving remains an uncertain undertaking. It’s dangerous. One botched roll in combat against that lucky goblin scout with a captured sword of quality, and your fledging character can die. Even when the odds are well in your favor, an inferior opponent still has a meaningful outside chance of ruining your day.
Let’s look at a starting Warrior character under ME RAW, who could easily be +5 STR +3 CON +4 WEAPON with 
DEFENSE 19 and a Warhammer, facing a lone goblin scout +2 STR +2 CON +2 WEAPON with DEFENSE 14 and that captured heirloom blade that crits on an 18+. Let’s use the goblin’s DEFENSE as the target number. The Warrior hits 60% of the time, eroding the goblin’s DEFENSE on every strike and causing the goblin to skip a turn on 17% of successful hits, each of which accrues average damage of almost 8 points after crits are figured in. Compare that to the goblin, who will hit only 20% of the time but almost always for a critical hit, averaging a killing blow on each successful hit, however remote. Narratively, the fighter is almost assured a victory over his inferior opponent in just two rounds if he can avoid getting hit, but he still has a 36% chance of dying during that same time period if he misses both attacks. It’s high-stakes, exciting, and unpredictable—and the goblin likes those odds, which, for most goblins, are as good as they get, representing at least a fighting chance. This combat will be Hobbesian: nasty, brutish, and short.
Now let’s take the same belligerents, slightly tweaked, and put them into the same combat scenario using a hits/damage threshold combat system. The Warrior is +5 STR +3 CON +4 WEAPON with DEFENSE 19, and the goblin is +2 STR +2 CON +2 WEAPON with the same DEFENSE 14, where hearts are now hits. The fighter has a damage threshold of 9, and the monster has a damage threshold of on 4; this means the goblin—even armed with the potent sword—will only score a hit above threshold about 50% of the time, while the Warrior makes threshold 100% of the time on successful hits. Thus, the Warrior is still hitting well over half the time and stealing a heart each time, while the hapless goblin’s ability to wound the Warrior meaningfully drops to around 10% on each attack. In addition to tilting the odds radically in the fighter’s favor, we have doubled or tripled the minimum turn length of the combat, as even if you are awarding an extra heart (“hit”) of damage on a critical attack, it will take at least two or three rounds for the duel to resolve. There are no one-shot kills. These factors make the narrative potentially less dramatic, as the eventual outcome is more certain, and it takes longer to get there. (Especially when a damage threshold system is put into place (compared to simply counting “hits”), where you still have to roll and calculate hit points of damage before applying them to a threshold “gate,” it seems like it could be a bit of a false economy in terms of saving time and effort through a mechanical hack.)
Now let’s look at two more tweaks. First, we debuff the poor goblin scout by giving him a standard crappy goblin sword made from sharpened bone. (Because goblins.) Then we keep the hits/damage threshold combat system but drop both combatants to . Now, the goblin literally has no way to harm the Warrior except for a natural 20 roll; if he does manage to crit (5% chance, a further quartering of his previous low odds), his average damage will be enough for an instant kill most of the time, but with low rolls a meaningful hit is still not guaranteed even on a natural 20 unless you apply the “extra hit on a crit” optional rule. In scientific terms, by implementing this version of the hits/damage threshold system, the archetypal fantasy goblin is rendered “statistically insignificant” versus a typical starting Warrior under this combat system.
[ETA: Hank’s red-yellow-green chart from his OSE hack post, embracing a “wild power” maximum damage-or-nothing approach to even things out, addresses this aspect of hits and damage thresholds in an interesting and inventive way.]
Sure you could simply have a horde of twenty goblins ambush the Warriors and chuck a fistful of icosahedra banking on that outlier—which would certainly be pretty epic—but it could also be pretty unexciting and would definitely change the entire narrative flavor of your game and your campaign.
I am not saying that any way a GM wishes to resolve combat is “wrong” or “bad” as long as it’s sufficiently fun for everyone, but I am saying that with every hack—and every tweak to every hack—there are trade-offs and consequences, some of which may be neither intuitive nor desirable. I think what can seem like the easy and obvious route to that quicker, easier, more seductive fast-paced table experience we may seek by following a “less is more” philosophy can sometimes actually derail our acquisition of it.
The devil is in the details, and I am an advocate of the details…
