D&D 5e’s CR system is… bad. In fact I will take it one step further: it is a disservice to DMs. The problem is the whole system is built around a single design assumption and this assumption simply no longer describes the most common way people play D&D. 

That assumption is you should have 6 to 8 encounters per day and about half of those should be combat. But is that what DMs actually do? What if I want to have one big fight per day? 

The reality is we need a mix of answers for a proper CR system, I propose 4: 

1. One fight per long rest against 1 strong enemy 
2. One fight per long rest against lots of weak enemies 
3. 3-4 fights per long rest, this specific fight is against 1 strong enemy 
4. 3-4 fights per long rest, this specific fight is against multiple weak enemies 

DMs need this mix of answers to try and build a variety of combat options for their player. This flexibility will also allow DMs to respond to what players naturally do in game. The CR system is fine for doing 3 to 4 combats, which works great for a classic dungeon crawl. It sucks at 1 or 2 combats more common in more narrative focus play. The problem here is 1 and 2 is what players and DMs naturally lean into. It is what they inherently do. Yes, a thousand forum posts point out that you should do more encounters per day, but to that I have to say… 

Bullshit. 100% Bullshit. I see it on every CR forum thread and this thought process, this counterpoint, is completely and fundamentally wrong. 

I encourage everyone who thinks “you should do more encounters per day” to reread your favorite fantasy story or rewatch your favorite fantasy movie. And as you do, actually try to count out the combat encounters per day. Yea, heck of a lot more one fights per long rest then 3-4 fights per long rest aren’t there? Why? Because that is GOOD STORYTELLING. 

Yes, from time to time a dungeon run with multiple encounters is great. That is part of the game and any system that did not do that well would also be doing a disservice to D&D. But storytellers have for centuries relied on an episodic structure of big events, move/reset, then another big event for telling epic stories. Throwing that out is trading The Odyssey, The Arthurian Legends, and most of the works of Tolkien to get improved dungeon balance. Are we sure that is a good trade? I think it is an awful trade. 

If your party needs to travel 7 days from one town to another, are you going to have to fight 28 random encounters against bears and wolves? That is a lot of combat and real world man hours to go to the next town and talk to an NPC or do a basic fetch quest. Most DMs don’t do this. There may be 0-2 encounters on that road along the way, but those will have a long rest in between them. This is also where your ranger or druid can shine by helping you dodge an ambush or pacify a fight. 

Want to do a big battle between two warring factions? Yea, your PCs will do everything they can to have a long rest before that happens… which is exactly what real generals did with their real armies before real battles. If you try to stop your players from doing that.. it will actually harm the immersion of the game. 

The first thing any useful CR system needs is a way to calculate for one big battle per long rest. You know, like in The Hobbit when they meet the trolls on the road, or the goblins inside the mountains, or the spiders in the forest, or breaking out of the elf jail, or the dragon in the mountain, or the big battle at the end. That right, everything in The Hobbit was one encounter per long rest. D&D 5e’s CR system can’t properly balance a recreation of The Hobbit or even a single chapter of The Hobbit (/facepalm). 

Ignoring it altogether is bad design. Now while we are at it, let’s go ahead and address the other counterpoints in the room. Is your group playing D&D to play D&D or are they trying to tell/experience an epic fantasy story through D&D. 

This matters because it changes a single word: should. Should you have 6 to 8 encounters per day? Well, if you are playing D&D for the sake of D&D and that is what the rules say, then yes you should. But if you are telling a story via D&D then you should set story flow (encounters) to match the needs at hand and then bend the difficulty to match that flow. Maybe that is 4 smaller combats as intended, but maybe that is 1 really big combat or 8 tiny combats or 1 medium combat followed by 1 larger combat. 

So I did an experiment on this. In my game we were coming up on the mid campaign big bad and I knew they would be doing so after a long rest… because my players are smart so of course they are. So instead of setting all the fight details, I kept parts of it to be determined. I determined his damage and AC, but then built out a list of nasty tricks/abilities and instead of removing health I just tracked how much damage my party did and then when it was down to the wire decided to let the next hit kill him. 

The party was four players at level 8, one player at level 9, and a good mix of abilities. The battle was the big bad, a pair of low CR guards at the start of the fight, and 10 low CR guard equivalent enemies at the end of the fight divided into 2 different hunting parties. In the middle of the fight I had the big bad use some, but not all, of the nasty tricks I had built. When all was said and done I did my best to reverse engineer what I did to what CR that actually was. It was CR 16 for a party of mostly CR 8 characters. I did not expect it to require that high a CR to challenge my group. To retest this I had a fight against a dozen Hun-like riders and archers. This time I explicitly set it a CR of 12 as they were just a group of bandits on the road. They were taxed, but I think they could have handled another fight of the same size. 

Assuming this scales, and I do not know if it does, that would give a new standard of: 

• 1 encounter per day: twice their level in CR 
• 2 encounters per day: +50% their level in CR 
• 4 encounters per day: their level in CR (current default) 

There are a few additional issues to take into account when doing one big fight. First, long rest centric classes will be stronger vs. their short rest cetric counterparts. For my party they were a good mix but take into account your party as needed. Secondly, there are multiple healing options in 5e vs.older additions, but the new healing dice and healer’s kits are not viable so a healer with full spell slots is more critical then normal. Third, my combat design was centered on having 3 “phases” to the fight, given that this was centered on one big fight I basically treated it like it was 3 small fights back to back. It took place on a giant map of the whole town, with different elements being pulled in at different phases and containing a polymorph of the big bad at the start of the 2nd phase and ending at the start of the 3rd phase just to hammer home the differences between the phases. My CR 12 fight was a 2 phase fight with additional calvary swing in about half way through. 

The hard truth is I just don’t think the current CR is built to be used like this. I can kind of force it too… but I make a lot of little edits along the way. D&D is very consistent in terms of AC, bonus to attack, and dice used per magical attack. Pushing the CR like this both adds risk for a run-a-way TPK and enemies that will never miss or never be hit. On the table top, I can fake this behind the screen but I currently DM on roll20 so those rolls are in the open. 

The design assumptions behind CR in 5th edition are just wrong. I think it works great for a 1980’s convention hall dungeon run. I don’t think it works for an episodic fantasy tale.