Level Independent XP Awards

[Anubis]

Anubis,

I missed your last response and forgot all about this issue. I was just looking at it this morning. And I realize that I didn't really understand your suggestion in your post from May 05.

I think the power factor idea is a good one. If I understand it correctly, each creature would be worth a certain amount of Raw XP, according to the formula Raw XP = (CR^2)*300, and this raw XP would be divided evenly between players. Each player then divides by their ECL (level plus racial hit dice plus LA) to calculate the XP they get from the encounter. This lets lower level characters "catch up" to higher level creatures.

With regard to your second point (from May 08), I don't think I've ever seen a creature listed with a negative CR- usually they are fractional. I think the idea is that eight 1/8 CR creatures is a CR 1 encounter. That would have a raw XP value of 300, so divide by the number of critters to get the raw XP value of each one; in this case, 37 xp.

Now I want to see if the power level thingie works. I'll use the example from page 185 of Grim Tales, which uses UK's system:

Suppose a party of three characters, one 20th level Smart hero,one 20th level Fast hero, one 18th level Strong hero, and one 15th level (with a +3 LA) Dedicated hero (a werewolf) confront a Vrock demon (CR 13).

In UK's system (and in Grim Tales) you take the total CR of the party (76), determine (table 14-1) that this is EL 25, subtract 4 for their being 4 members (table 14-2), so EL 21. A CR 13 encounter is EL 15 (table 14-1), so the difference is -6. Table 14-4 gives such an encounter as being worth 37.5 xp per level. So 750 xp for the 20th characters, 675 for the 18th level, and 562 xp for the 15th level werewolf (xp is per hit dice, not ECL). Total xp is 2737.

In the system I propose here (raw xp= (CR^2)*300, divide evenly, then divide by character level) the Vrock is worth 50,700 raw XP. That's 12,675 raw XP per character. The 20th level characters get 1/20 the raw XP, or 633 XP each. The 18th level characters get 1/18 this raw amount, or 704 XP each. Total XP awarded is 2674.

The two systems give a similar amount of XP, but UK/GT doesn't allow low level characters to catch up with high level characters. In this system lower level characters get more XP. In fact, they can get ridiculous amounts of xp. A 20th level character accompanied by 3 first level characters would still get 633 xp, but his companions would get 12,675 xp each; they convert raw xp to xp on a 1-1 ratio. They automatically go up at least one level.

Now is this a bug, or a feature? Can you really advance low level characters merely by slaying a vrock in their presence? If not, how could you fix it?

BTW, I misstated the rule about xp for characters with a LA- Grim Tales says that xp is awarded per character level (I think this is unchanged from UK's v5) But what if you had a monstrous character who had racial hit dice and a LA, but no character levels? No xp? I bet the author of Grim Tales (Wulf Ratbane) meant hit dice, not character levels.

[edit] The original method would award (50,700/76)=667 xp per person, or 2668 total. You take the raw xp total of (13^2)*300 and divide by the total party level of 20+20+18+18=76. That's what each person gets. See the following post for more examples.[/edit]

Actually, you're not using my system correctly. Sorry for the slow response, but I haven't been around the boards as much lately. Anyway, on to the explanations.

First off, yes, characters with lower levels SHOULD get more XP for the same encounter. This is for balance and realism both. First, it allows characters who create items to eventually start catching up, albeit never all the way. Second, less experienced people, logically, would learn more from a given challenge than more experienced people because they have more to learn.

That said, here is how the system works:

First, each monster has a set amount of Raw XP. This is " (CR^2)*300 ". This makes finding Raw XP numbers easy by just having a chart for each CR. Anyway, yes, there ARE negative CR values. I'm using the Imortal's Handbook method for determining CR/EL, remember. Although it DOES translate into a fraction and then into an EL, the first number is a set value based on the factors; this doesn't really matter, though, because you only use the base CR value if it's 1 or more, while for values below 1, you want to go with the fractions. Basically, ignore the thing about negative CRs. Just simply go with the base number if it's greater than 1 or the fraction if the base CR is less than 1.

Moving on, the next value needed is the Power Factor. Each member of the party has his or her own Power Factor. Simply put, this is " LV*#Members ". These two numbers came from the condensing of the formula from earlier. Note that I DO NOT use ECL per the standard rules, but as per the Immortal's Handbook, meaning ECL and CR ARE EQUAL. You do use the ECL for this formula, though. Basically, ECL is the overall modified CR.

The final step is finding the actual XP. This is found by simply diving the Raw XP by the Power Factor PER PERSON.

I'll give an example. Say you have a party of four. The party members are ECL 15, 15, 14, 14. Their Power Factors are thus 60, 60 (both 15*4), 56, and 56 (both 14*4). This is best done before the session and just noted, as it only changes if someone gains a level or someone joins or leaves the party. Anyway, let's say they face two CR 13 creatures. CR 13 creatures are worth 50,700 XP each, meaning the encounter is worth 101,400 XP total. To get the XP, simply take the Raw XP and divide by each Power Factor individually. That means the gains are 1690, 1690 (both 101,400/60), 1810, and 1810 (both 101,400/56). As you can see, the gains aren't grossly outrageous.

If you put low levels with high levels, though (which shouldn't happen in the first place, but still . . .), well, I have a house rule that takes care of that. You see, I give XP per encounter and allow level gaining in the middle of the adventure. This is to simulate "learning on the job" to give a bit more realism. Still, this house rule works either way. At no time can any character EVER gain more than his or her level times 1000 XP. So if you're Level 1, you can only gain 1000 XP max; that's 2000 max at Level 2, 3000 max at Level 3, etc. Bascially, if the EL (which is pretty much never used except by the DM to pace encounters in my system) is +8 or more, your XP is gonna be pretty much capped out.

After EXTENSIVE use in my campaign, I can certify that this system works flawlessly and has absolutely no bugs in it. Everything works precisely how it's supposed to, and it finally gets rid of the problem in using EL that I mentioned before. I only hope UK puts it in the Immortal's Handbook. I can assure everyone this is the most balanced and realistic method. Sorry if I sound egotistical, I'm just really proud of coming up with the originating formula. By the way, Cheiromancer, thanks a TON for condensing it to the more managable formula.

 

[Cheiromancer]

What do you mean I’m not using your system correctly?

What I was doing was divide raw experience evenly, then divide by the character’s level. In your example I would divide the 101,400 raw xp evenly (so 25,350 per character), then divide either by 15 or 14. The 15th level characters would get 1690 xp (25350/15) and the 14th level characters get 1810 xp (25350/14). Which is, of course, the correct answer.

My only doubt about your system is that it makes the players have to divide their xp. That’s not something they do naturally. I mean, you can say “everybody get 120 xp per level” and they won’t forget to multiply, but if you tell them “everybody get 1200 raw xp” they might forget to divide. It’s just human nature. Though I could do it for them, of course.

As for me, I would calculate raw xp (101,400), divide by the total party level (58) and tell everyone they get 1748 xp (101,400/58). One problem with this is that folks who fall behind will never, ever, catch up. And I kinda think they should.

But the biggest problem with *my* method (one I've just now noticed) is that it makes the total xp from an encounter independent of the number of characters in the party; it is dependent only on their level. If I have twice as many characters (but the same level) the total party level doubles, so everyone gets half as much xp. But there are twice as many characters, so the total xp is the same.

(The same is true of your system, my friend. Twice as many characters means that the power factor is doubled, so everyone gets half as much xp. But since there are twice as many people, that means the total xp is the same.)

However, in the IH doubling a party will double the CR and double the number of combatants. This will increase the party EL by +4 for the increase in CR, and decrease the party EL by -2 for the increase in numbers (tables 2-1 and 2-5 in v4, or tables 2-1 and 2-2 in v5). A net increase of +2, which means the difference between the encounter level and the party level drops by +2, which means the *total* xp awarded is halved. (See table 2-8 in v4, or note that party size is counted twice in v5; once in table 2-2, and again in table 2-5).

So I think we need a new formula!

For 4 characters of level n, the formula for total xp is

xp = (CR^2)*300/n

Highlight to see the derivation of this formula (which we already know and accept):

Spoiler

First, we know from UK's tables that doubling a creatures CR increases its EL by +4, and that quadruples the xp awarded. So xp is proportionate to CR^2. The xp is also a function of the party's level (and maybe other things); call the function f. Then we want to figure out what f is in the formula

xp = (CR^2)*f

Now, we know that 4 characters of level "n" need to face 13.3333 equal CR challenges to go up a level. For an nth level character to go up a level, he needs to get n*1000 xp. That's n*4000 xp for the group. Divide by 13.333 and that's n*300 xp per encounter. So if CR=n in the previous equation, then

xp = n*300 = (n^2)*f

and we can solve for f. In other words, when there are 4 characters of level n,

f = 300/n

And so we conclude that, for a party of 4 nth level characters,

xp = (CR^2)*300/n

Let p be the number of characters in the party. Then the total xp awarded is a function of p. Call this function g(p). g(4)=1, and, generally, g(2x)=1/2 * g(x). Clearly g(p)=4/p In other words, the total xp awarded for an encounter is

xp = (CR^2)*1200/(p*n)

The expression p*n is just the total level of the party.

----

Now, what to do when the members of the party are of different levels? What UK does is award people xp on a per level basis. In this case you divide the total xp into a number of shares equal to the total number of levels in the party. Each party member gets 1 share per level. In which case

xp per level = (CR^2)*1200/(total level^2)

Now since for characters I am using "level" when I really should be using CR (since ECL and CR is the same in UK's system) I can say something quite neat, namely that the xp award (per character level) is directly proportional to the sum of the squares of the CRs of the opponents, and inversely proportional to the square of the sum of the CRs of the characters. That is quite as elegant, imho, as the pythagorean theorm.

If you divide xp equally (as I am wont to do), you need to determine a power factor for the group, equal to the number of players times the total level. (Or you could say the number of characters squared times the average level of the character.) Now the raw xp is (CR^2)*1200, and you divide by the power factor to work out the individual awards.

To give more xp to lower level characters (as you seem to like), each character has a power factor equal to the number of characters squared times *that* character's level.

After EXTENSIVE use in my campaign, I can certify that this system works flawlessly and has absolutely no bugs in it. Everything works precisely how it's supposed to, and it finally gets rid of the problem in using EL that I mentioned before. I only hope UK puts it in the Immortal's Handbook. I can assure everyone this is the most balanced and realistic method. Sorry if I sound egotistical, I'm just really proud of coming up with the originating formula.

All this may be true, but your formula departs greatly from what UK has in the Immortal's Handbook whenever the party size is significantly larger (or smaller) than 4. So did mine, of course. Would you care to revise your statement of flawlessness, bugfreeness and balance, or would you prefer to argue that UK should change his system?

 

[Anubis]

What do you mean I’m not using your system correctly?

What I was doing was divide raw experience evenly, then divide by the character’s level. In your example I would divide the 101,400 raw xp evenly (so 25,350 per character), then divide either by 15 or 14. The 15th level characters would get 1690 xp (25350/15) and the 14th level characters get 1810 xp (25350/14). Which is, of course, the correct answer.

My only doubt about your system is that it makes the players have to divide their xp. That’s not something they do naturally. I mean, you can say “everybody get 120 xp per level” and they won’t forget to multiply, but if you tell them “everybody get 1200 raw xp” they might forget to divide. It’s just human nature. Though I could do it for them, of course.

The part you got wrong was in doing two divisions. By using the Power Factor, there's only one division. Raw XP divided by Power Factors gives the proper XP.

As for me, I would calculate raw xp (101,400), divide by the total party level (58) and tell everyone they get 1748 xp (101,400/58). One problem with this is that folks who fall behind will never, ever, catch up. And I kinda think they should.

But the biggest problem with *my* method (one I've just now noticed) is that it makes the total xp from an encounter independent of the number of characters in the party; it is dependent only on their level. If I have twice as many characters (but the same level) the total party level doubles, so everyone gets half as much xp. But there are twice as many characters, so the total xp is the same.

(The same is true of your system, my friend. Twice as many characters means that the power factor is doubled, so everyone gets half as much xp. But since there are twice as many people, that means the total xp is the same.)

This is not a problem, it's how it's supposed to be. Twice as many people means half as much XP. The Raw XP isn't supposed to change. For any given encounter, XP should be halved if party level OR size doubles. That holds true in my system. What's the problem?

However, in the IH doubling a party will double the CR and double the number of combatants. This will increase the party EL by +4 for the increase in CR, and decrease the party EL by -2 for the increase in numbers (tables 2-1 and 2-5 in v4, or tables 2-1 and 2-2 in v5). A net increase of +2, which means the difference between the encounter level and the party level drops by +2, which means the *total* xp awarded is halved. (See table 2-8 in v4, or note that party size is counted twice in v5; once in table 2-2, and again in table 2-5).

Whether the initial XP OR the final XP is halved, the end result is the same. The PROBLEM with UK's system is that it uses EL and EL goes over several levels. The total Raw XP isn't supposed to EVER change. In his system, it does, and it causes a problem. A Level 8 character in his system gets less from an encounter than a Level 9 character in the same encounter as I showed before, meaning not only will people never catch up, they'll get further behind.

So I think we need a new formula!

SNIP all the rest

Why? The Raw XP is flawless. You have yet to show a PROBLEM. So what if the Raw XP changes? The way to find out if the system works is to take a party of four of any level, tack on 13-1/3 encounters of the same CR, and see if that gives them the XP needed to gain a level. It does under my system, meaning it works.

You are making things too complex. Monsters are worth a set amount of XP, adjusted for the power of people going against them, and that's shown in the Power Factor. In practice, the end result is exactly the same, except my system requires less calculating.

Your first problem is looking at EL. IGNORE EL. It's irrelevent. EL covers too many different CRs at higher levels to be of any use. EL is merely a tool for DMs to gauge encounters and should have no bearing on XP. Level 100 and Level 108 are the same EL, but one is obviously harder than the other, even if by just a little bit. My system takes that into account, as it should be.

Anyway, as I said, in practice, the results come out the same. My system gets you there faster with a formula that the average player or DM could grasp. That's why I thanked you for compressing my formula, in fact, to make it more friendly to average people. Still, that goes to show that my method works.

All this may be true, but your formula departs greatly from what UK has in the Immortal's Handbook whenever the party size is significantly larger (or smaller) than 4.

No it doesn't. More or less people mean a higher or lower Power Factor, which in turn adjusts the XP. A party of eight should get half as much XP as a party of four. CR 15 is worth 67,500. A party at Level 15 gets 1125 XP per person, and a party of eight at Level 15 gets 562 per person. I don't see the problem.

So did mine, of course. Would you care to revise your statement of flawlessness, bugfreeness and balance, or would you prefer to argue that UK should change his system?

Hmmm . . . Seeing as my system IS flawless, bug-free, and balanced, I won't be revising that statement. You haven't shown any problem with it. Come up with any example and you'll see that it works. You gotta go with final numbers, as that's the ONLY thing that matters.

Yes, UK should change his system. EL doesn't work, period. Those with a higher level in the same EL gain more XP, which is the opposite of what should be. XP gains for any particular encounter should go DOWN as you gain levels, not up. Yours may do that, but mine does the same thing a lot more easily and accurately. UK's doesn't do it at all. So yes, that's why I'm hoping he adds this to the Immortal's Handbook.

 

[Cheiromancer]

A party at Level 15 gets 1125 XP per person, and a party of eight at Level 15 gets 562 per person. I don't see the problem.

The problem is that a party of 4 at level 15 gets a total of 4,500 xp, which is exactly what a party of 8 at level 15 gets. The award per person should not just be cut in half- it should be cut into quarters. At least, that's what the IH appendices say.

You note that if a party's CR doubles, the total xp awarded is cut in half. The power factors assures that. But if the number of members in the party doubles, the total xp awarded by your system stays the same.

That's my problem!

[edit]I checked with Wulf over here. (subtle brilliance thread) and he says that the total xp is supposed to be constant. I.e. the two IH appendices are incorrect. The central idea is that 13.3333 moderate encounters are enough to advance a person a level.

So no problem after all.[/edit]

 

[Cheiromancer]

Also on that thread I've floated a proposal that the xp be related to the square of the party's levels, not just the square of the CR. I.e., you square the levels of the party members and add them up. Call that Y. Then the xp per character level is

xp = (CR^2)*300/Y

It's a generalization of the formula we were using for the total xp of a 4 person group, all of whom were the same level:

total xp = (CR^2)*300/n

where n is the level of the group. Divide by the total number of levels to determine the xp/character level. If you divide by 4n, the denominator looks a lot like the sum of the squares of the CRs of the characters, which is nicely parallel to the CR^2 in the numerator (that, of course, is a sum when there are multiple opponents.)

I think the Y formula is kinda elegant. I wish there were an equally elegant way of helping characters catch up in xp, but I don't know of any. The power factor method that Anubis (replace n by the product of the individual's character level and the number of people in the group) doesn't seem as pretty to me.

 

[Upper_Krust]

Hi guys!

My heads not really up to speed on the CR/EL system at the minute, what changes (from the v5.1 in Grim Tales) are you suggesting.

I understand the details of the new formula and so forth but I am just trying to gauge the big picture here.

 

[Cheiromancer]

Well, aside from some details about how the experience should be divided, the difference is that you could use a couple of formulas instead of the EL and XP tables.

For each monster you could compute their "power" equal to its CR^2. It could be in the monster entry, or you could generate them directly from the CR. You would also compute the raw power of the characters in the adventuring party; the sum of the power of the characters is Y. A DM would have to calculate that, and change it when characters level up, and so on.

Then by comparing the power of an encounter to Y, you can see how much of the party's resources should be used up. If the power is 1/4 of Y, then it is a moderate encounter. If it is equal to Y, it is a 50/50 proposition that the party will get slaughtered, etc.. Divide by Y and multiply by 100 to get the percentage of resources used up.

And you use the formula from my previous post (or one like it, depending on your philosophy of dividing xp) to determine what the xp award is. A piece of paper and a calculator, or a few cells on a spreadsheet. A monster entry could have its "raw xp" of (CR^2)*300 which is divided by Y to get the xp award per character level.

Situational modifiers will affect the relative power, of course. If the party is handicapped so that their EL is two less than normal, that corresponds to their power being cut in half. In another situation a monster might, for tactical reasons, have a power 50% higher than normal, and thus be worth 50% more xp.

If you look at a table look-up as being equivalent to a calculation, there is probably no difference between the two. But a THACO formula is equivalent to a table lookup as well, but nobody seems to prefer the table.

Anubis thinks his variant is perfect, but I think there are still a few refinements to be made. Perhaps he'll chime in!

 

[Cheiromancer]

A few other things- the example with the Red Dragon and the skeletons works a lot better- you don't need to disregard the skeletons. Recall that was a CR 62 Great Wyrm Red Dragon with 600 CR 2/3 human skeletons. Fractions are funny, but I think what you do with them is treat them as 400 CR 1 opponents. Then the raw xp from the encounter is given by

(62^2)*300 + 400*(1^2)*300
=3844*300+400*300
=1,153,200+120,000
=1,273,200

Over 90% of the experience comes from the Great Wyrm. The 600 skeletons might occasionally interfere with charging attempts, could attempt to grapple spellcasters, and a few actions might be wasted in dealing with them. Perhaps the skeletons came in waves over a few minutes, and so burned up some of the 1 round/level buff spells, or maybe a turning attempt was used before the numbers became apparent. Anyways, they do relatively little, but are not entirely negligible.

The encounter has a power of 4244 (vs 3844 for the dragon alone), so 4 35th level characters (power 4900) should find it a tough fight if neither side is prepared or has any particular advantage.

In your system four 35th level characters are EL 25, vs an EL 24 dragon with an irrelevant skeleton horde; also a pretty tough fight. So the results are pretty close; the advantage to the formulaic approach is that you don't have to arbitrarily discard the contribution of weak allies.

The formula approach also handles unbalanced groups a little better. Consider the case of a 20th level Ranger with a trio of first level halflings in tow. The EL of a 20th level solo character is 18; with the hobbits the EL of the party is 14. If the hobbits hide in the bushes while the Ranger (still not King) defeats CR 14 bad-guys. CR 14 would be a moderate expenditure of resources for a 20th level character fighting by himself, but he gets 4 times the xp because of the hobbits with him (due to the shift of 4 in EL)

A DM could just disregard these shenanigans, of course, but such arbitrariness is unnecessary if the Y formula is used. The power of a 20th level character is 400. That of a 1st level character is 1. Together the group has a power of 403. So a CR 14 encounter (raw xp 58,800) earns him

2940 xp solo (58,800/400 * 20 levels)

and

2918 xp with hangers on (58,800/403 * 20 levels)

The hobbits each earn 145 xp by observing his prowess in battle. (58,800/403 * 1 level) After all, they are at much more risk than if they had stayed home in their hobbit holes. And they could be some small help to their high level friend- stabilize him if he slew his opponent in the same round as dropping to -7 hp, call out a warning if he fails a spot check, etc.

 

[Kavon]

xp = (CR^2)*300/Y

Ok, so if I understand you correctly...

Let's say we have a party of 4, with one level 6 (36), two level 5 (25*2), and one level 4 (16) characters (Y=102).
They win a fight with a CR of 5, which is 7500 raw XP.
7500 / 1200 = 73.53 (do you round down for this? I'll assume it does for the following, since it's irrelevent for the example).

The Lvl 6 character gets 438 XP, which is about 13.7 encounters to level up.
The two Lvl 5 characters each get 365 XP, which is also about 13.7 encounters to level up.
The Lvl 4 character gets 292 XP, which is also about 13.7 encounters to level up.

It will take the members of the party an equal time to level up, irrelevent of what level they have. So, the level 4 character will always remain 1 level behind on the level 5 characters, who will always be 1 level behind on the level 6 character.

Did I get it right?

If so, I don't really see how this is an improvement on Anubis' formula, since the characters don't even out in the long run.

Edit: wow, how could I not see that newer reply? *pokes browser*

Hmm... That side of the system (talking about the the Lord of the Rings orientated explanation) does sound good, but the fact that advancement is equal for each party member, irrelevant of level difference, does make it sound less good. The Ranger needs to fight the same fight (a little under) 7 times to go to level 21, and the hobbits will go to level 2 at the exact same time. When the Ranger hits level 22, the hobbits hit level 3, etc...

Edit 2: Hmm... Though, after compairing the two results, Anubis' formula seems to hold the other extreme, with the Ranger needing a little over 27 such fights to hit level 21 (solo: a little under 7), and the hobbits each gain enough XP to rocket them into the high end of level 4...

 

[Wulf Ratbane]

Hmm... That side of the system (talking about the the Lord of the Rings orientated explanation) does sound good, but the fact that advancement is equal for each party member, irrelevant of level difference, does make it sound less good. The Ranger needs to fight the same fight (a little under) 7 times to go to level 21, and the hobbits will go to level 2 at the exact same time. When the Ranger hits level 22, the hobbits hit level 3, etc...

There's a very simple fix, however.

Total all of the party levels to calculate the total XP award, then just divide by the number of characters in the party.

This is the "Equal share for inequal contribution" method, which allows the hobbits to soak up most of the XP actually earned by the ranger.

Whether or not you think that method is flawed in and of itself is down to DM preference. Personally, I think it's silly to give the hobbits a greater proportion of the XP than their own proportional contribution to the fight.


Wulf