Strange how people see things.
The only thing going against 1-4th level characters is their lack of revival spells such as revivify.
Otherwise, if you know your math, all levels are pretty much the same except, of course, that higher level characters have more tools. But mathematically, if the DM plays fair (for both enemies and players) the only difference will the length of the fights as the math keeps the game "balanced".
I have seen more TPK in the early 5ed than in any other edition combined. Why? Would you ask? Simple, this is the first edition where the rabble keeps a chance to hit mid to high level characters. I do not build encounters with mono groups. You encounter 10 goblins? You can bet that there will be a priest or some other casters. Just one priest in a group can make a huge difference. Take the goblins, on bless will make it so that a simple +4 to hit is now +6.5. This means that an AC of 20 only needs a 13 or 14 to hit. Make the goblins into hobgoblins and now you get simple rabble that will hit an AC 20 for a mere 15 on the rolls (sometimes as low as 13) and hit for 1d8 + 2d6 +1 of damage (average 13 of damage). Put these on archers and now your "rabble" can hit the casters in the rear. Add in a hobgoblin iron shadow and it is more than possible that the casters will be hit with the full 13 damage almost every shot as the Iron Shadow will get in hand to hand combat with the casters. And casters still have the weakest AC. Add in some optional rules such as Flanking, and even an AC 20 will be hit way more often that what is healthy for any character level. This is this simple fact that led to so many TPKs. The underestimation of the rabble.
In previous editions, the math was more instinctual. As the character would progress, so would the type of opponents. From goblins, to hobgoblins (orcs) to gnolls, ogres, trolls and giants. The only difference was the length it would take to neutralize your opponent. The rabble was quickly left in the dust behind the characters. Now, with 5ed, the rabble keeps a good chance to hit as the max AC is 20. (I know you can raise it higher with spells and such, but these are not available to everyone) and depending on the level, the rabble can become ogres, orogs and so on. Just one priest or wizard can make a huge difference.
So no. The level 1-4 are not truly training levels. These are just "deadlier" levels where the characters do not have access to revival spells. The game plays the same for all its length but as you grow in levels, so are tools. This is this "augmentation" of tools that makes the game much harder to play for DM and Players alike. The more choice you have, the bigger the chance you are to forget something. It is this level of complexity that most players (and DMs) are not ready to face and thus, most campaigns ends around level 10.
Creating adventures for 10th level characters and up takes a lot of time, much more than the first few levels. And taking great care that character can and will be challenged all the way through takes experience and dedication to learn both your players, their characters and the motivations of both. Multiply that by the number of players and it begins to take quite a toll on the DM's time. For me, to make my players play 1st or 20th is almost the same as I know them and their character almost by heart. But not all DM are ready to spend that amount of time.