There are lots of inconsistencies
I'll try to address some of them.
The multipliers are important.
For example, spells like Conjure Animals don't consider the multipliers. This means, according to the spell, one CR 2 beast is equivalent to eight CR 1/4 beasts. According to the Monster Manual, one CR 2 beast is worth 450XP. A single CR 1/4 beast is worth 50. Eight of them are worth 400, or 1000 using the multipliers. If those modifiers are accurate, then you are summoning 2.5 times the challenge for the enemy. Which do you think most players usually summon?
Similarly, you need to consider the multipliers when gauging an encounter or you may end up with more challenge than intended.
CR is a bit broken
Unfortunately, it's as if all institutional knowledge with respect to challenge and experience points was lost with 5th edition* and they just created a completely new system that is, in my estimation, fairly broken. There are a number of creatures that are rated to high for their level of challenge and a few party killers rated too low, like the Will-o'-Wisp.
All CR since its inception in 3rd edition has failed to accurately capture challenge, especially when it comes to special traits and features that can dramatically affect the challenge. The 5th edition DMG has rules for creating a monster, but they do not reflect the monsters in the Monster Manual. Whether that's because they didn't have the DMG rules when determining CR or just ignored them is unknown. The blog of holding did an excellent analysis on this and created a table based on the Monster Manual numbers. This is more useful for creating monsters, but also for analyzing existing monsters.
Reprints
Also, if the monster shows up in Monsters of the Multiverse, it's likely it has been adjusted to closer match the Challenge Rating, so you may want to use the newer version.
To the rescue
Trekiros, a gaming youtuber created a pretty cool tool, battlesim which I've used to gauge encounters. It ignores CR entirely and simulates a battle. I don't put the party's characters in, just 3-4 basic party members and if the players win in four rounds with no more than one dropping, I have a pretty reasonable encounter for most folks.
Advancement by the numbers
The other very bizarre result of 5e's experience system is, as you've pointed out, the multipliers aren't rewarded, which means if you had a deadly encounter with a singe foe, or a deadly encounter with eight, you'd earn 2.5X less experience for the latter, meaning, you're characters will need more encounters to level when using multiple opponents. I've never understood that logic, but that's not how I determine level. Why? Because even ignoring the multipliers, it doesn't make any sense. For example, according to the Player's Handbook Character Advancement progression and the Dungeon Master's Guide XP Thresholds by Character Level, you have this strange outcome of the number of medium encounters to level starting at about 6, doubling at 3rd level to 12, then 15, but then, it starts going down again, with a notable exception of level 10 (which requires 17.5 encounters), then around 9-10 from there on out. What??? Really bizarre and as mentioned, encounters with multiple creatures will be rewarded less than their challenge, driving up the already awkward number of encounters to level.
Inconsistent tables
There are several tables that combine for challenge, experience and character level.
For a system intended to be easier, this is both not simple and not internally consistent. I've touched on the first item, where you determine difficulty using the multipliers, only you don't award the higher value. This will mean your characters will need to overcome more encounters to level for the same CR, if the encounter is made up of multiple creatures. The second inconsistency, also touched on, is the rate of encounters. However, there is a further inconsistency, because the CR table and the XP Budget tables are different.
Level |
# Equal CR encounters |
Medium encounters to level |
2 |
6.0 |
6.0 |
3 |
5.3 |
6.0 |
4 |
10.3 |
12.0 |
5 |
13.8 |
15.2 |
6 |
16.7 |
15.0 |
7 |
15.7 |
15.0 |
8 |
15.2 |
14.7 |
9 |
14.4 |
15.6 |
10 |
12.8 |
14.5 |
11 |
14.2 |
17.5 |
12 |
8.3 |
9.4 |
13 |
9.5 |
10.0 |
14 |
8.0 |
9.1 |
15 |
8.7 |
10.0 |
16 |
9.2 |
10.7 |
17 |
8.0 |
9.4 |
18 |
8.9 |
10.3 |
19 |
8.0 |
9.5 |
20 |
9.1 |
10.2 |
I understand the desire to have nice round numbers in the tables, but I would rather have consistency.
Number of encounters to level
However, we can generalize, about two easy or medium, two hard and one deadly encounter make up an Adventuring Day, analyzed here. You can then generalize advancement by number of adventuring days, keeping it at one for rapid advancement, or two or more days for slower advancement. You can also have it increase by Tier, such that one adventuring day is required to level at Tier 1, two at Tier 2 and so on. This sort of system is what I've been using since I started playing 5e.
Rest
The amount of rests taken between encounters has a tremendous impact on encounter difficulty. In general, the DMG assumes the entire adventuring day before a long rest, with perhaps two or more short rests. In practice, I generally see many fewer encounters per day as a player.
As DM, I do try to use the guidelines, but I enforce them with the story - essentially, you can take short rests as often as you like, but you will run out of Hit Dice after two or three short rests, and then I require a "save haven" for a long rest. You're welcome to take them whenever you want, but you often have to retreat back to a safe area, like a town or nearby grove of allies.
Milestone advancement
You can, and I think many DMs do, eschew experience entirely and use story milestones for advancement. My suspicion is this is the most common method, especially when using commercial adventures.
* 3rd edition Unearthed Arcana presented Level-Independent XP Awards to address making challenge and XP more aligned and introducing the idea of mixed encounters simply adding up XP, what 5e calls Building Encounters on a Budget. This became the basis for the standard Pathfinder 1e experience system, including their CR-Equivalencies. In 5e, without the appropriate equivalencies, they had to introduce the clunky multiplier system to make up the difference.
If enemy challenge rating values were such that CR X + 2 was double the award for CR X, then the multipliers would not be necessary and budgets just work. XP to level could simply be a static number or encounters per level, or even follow the weird rate above, and that could be multiplied by the CR for that level and you'd have a dead simple system with two tables, one derived from the other and all the rest of the nonsense would be unnecessary. If you wanted a medium encounter, you would use the budget for equal CR and you could split that up between multiple enemies without any fuss. You can even generalize equivalencies for even simpler estimation, which pathfinder did here.
This doesn't fix creatures that are sized inappropriately (too high or too low CR), but it does greatly simplify the rest. Why 5e came up with a brand new, inconsistent system is a question for the designers, but in my estimation, it was a mistake. If I wanted to use experience points, I would probably eschew 5e's system and just use the XP for CR table and one of the advancement columns from Advancing your Character table from Pathfinder 1e.