More XP

I was puzzling over the DnD Next XP charts today trying to figure out their system for assigned XP values to monsters.  This is really important for a bunch of reasons, primarily because those XP values are presumably useful to generate fair encounters.  In theory you can use the DM guide for encounter building and simply pick out monsters with an XP total equal to the 'normal encounter' amount for their group and go.  It turns out this is not the case at all.  

For example, a gnoll is 40 XP.  This works out semi reasonably at low levels where encounters with gnolls are tough but winnable.  At level 9 a character has a budget of 700 XP though which works out to 18 gnolls.  Those gnolls shoot for 6 damage hitting about 20% of the time so those 18 gnolls deal 21.6 damage per round which kills the character in 3 rounds.  There is no way a character can deal with 18 targets spread out like that - this is not a normal encounter but rather a death sentence.  It is even worse if you have a four person party because the gnolls can easily focus fire and kill a character every round.  Hell, you can't even run away from them because longbows have utterly absurd ranges.  Encounters built with high level monsters work reasonably though because high level monsters were budgeted with high level characters in mind.

It seems clear that what they did was design a nice looking XP chart and then built monster XP values to match the chart.  That is not an appropriate way to approach this.  What they should have done is to assign a level 1 monster an XP value and then figure out how many of those monsters it takes to challenge a character of a given level.  If a level 9 character can beat 7 gnolls then they should have an XP budget of 280 and high level monsters should be costed appropriately.  This is a classic example of building the game to match a chart pulled out of somebody's ass instead of building the chart to make the game work!  Don't get too attached to your fancy charts and algorithms when they interfere with the game working, I say.

Heroes By Trade was built with that idea in mind.  I did a bunch of simulations to figure out how hard monsters should be to offer a tough challenge and then kept iterating on my formula until the Encounter Strength of the monsters actually reflected their chances of beating appropriate parties.  It took a bit of doing but I think the system is fairly robust right now.  I do have the advantage that HBT is not designed with the ideal that characters can wade through thousands of chumps at high level - the scaling is much less which makes the formulas much easier.  

Fundamentally the problem is that Next focuses all defensive scaling on HP.  Offensive scaling is very substantial but when you have six times as many HP at level nine as level one but enemies still hit just as often and just as hard you simply can't take on eighteen of them!  If you want to build a system where characters can massacre infinite legions of losers you have to implement serious defensive scaling and they have not done that.