Fight Rewards Explained (in General)

Synge[Memento Mori]February 24 2008 12:10 PM EST

I've been crunching a few numbers this morning, and I can present my findings here. I'll have to think a little bit about how to present any actual numbers that were used to back up this data, though, since they don't easily transfer into these forums. It is there, though, and I think I can cut and paste small sets into here if questions come up.

Anyway, here are the key points:
(1) There is a concept of a "base" reward for each opponent that is independent of any bonus. Note: This base reward does not include any of the printed bonuses: challenge bonus, a printed clan bonus, or N*B bonus. The printed bonuses modify this base reward.
(2) The base reward for each opponent if based on that opponent's PR. (Note: it is NOT based on the attacker's PR, BA cost, etc.) Yes, this means that different people have different reward values, with the higher PR folks giving the better rewards. As previously discussed, this may give Ranger an edge, since he fights only high-PR people, but check the numbers before you complain too loudly. Note that finding (1) shows that this is independent of any printed bonus.
(3) There are bands of PR in which the base reward is the same for each person in the band.
(4) There is a "hidden" 5% clan bonus when a clan member attacks another clan member. This bonus is not displayed in the fight log, and works even on Sunday. Effectively, it modifies the "base" reward by 5% (i.e. increases the reward by 1.05).

Ok, some examples:
Anbhas (PR: 2,773,243) and Godsend (PR: 2,671,166) are both on my fight list. The base reward value (a metric useful only for comparing rewards--for example, if the base reward value for one character is twice that of another, the average money and xp will be twice as large) for Anbhas is 91.2. The base reward value for Godsend is 86.8. However, Anbhas is in a clan and Godsend is not, so if we remove the hidden 5% bonus, we get that both have a base reward value of 86.8. In other words, they are in the same PR band.

Now, that 5% bonus is important when we look at someone else from my fight list: Rigwarl Bristleback (PR: 2,073,698). If we look at Rigwarl's base reward value without the clan bonus, it is in the 83.5 band, which you will notice is lower than the two above. However, the 5% bonus modifies the base reward value to 87.6, which makes it more profitable, even without any other bonus, to fight Rigwarl than to fight Godsend, by about 0.92%.

What does this mean? There is a second, indirect penalty for high PRs aside from lost challenge bonus, in that characters with high PR provide better rewards, and may be attacked more often as a result (clan folks are really the only ones that care about this). Also, it means that the 6/20 crowd should still carefully consider how far down to fight, even without a negative challenge bonus. Lastly, it may be profitable for clan folks to get better rewards by fighting non-clan folks if they chose their targets carefully. For example, Godsend has a base reward of 86.8, whereas Ichigo Kurosaki (PR: 1,968,469) has a hidden-clan-bonus-modified base reward of 85.9 (Ichigo Kurasaki is in the 81.8 band). So, you get higher rewards for fighting Godsend than Ichigo Kurosaki, by about 1%. (Worth it? Up to you since it depends on how much you value the extra clan points).

It would be interesting to find out where the boundaries of the PR bands are. The base reward values also seem to be very flat at the high end of the game, without much difference between 1.5 mil PR to 3 mil PR, so the effect is not giving e.g. Ranger a pronounced advantage, but it would be nice to confirm that. But, that is a project for another day

AdminQBnovice[Cult of the Valaraukar]February 24 2008 12:12 PM EST

Really really nice work man.

Admindudemus[jabberwocky]February 24 2008 12:26 PM EST

aye, very nice. when ranger and i did some messing around with rewards one weekend, we saw that he probably had about a 16% xp bonus advantage over me due to fighting some people that i could not. granted, this was with a certain set of opponents.

i do think people need to realize that getting into the 6/20 range is not enough and if you cannot beat the same people that the top dogs beat, then their growth will still outpace your own, unless a roe is involved. the top dogs are usually in a better position though to use a roe and not have to fight down as much as others. this is the reason, as i have stated before that i retrained and left the robf behind.

i still have to fight down when i remove my tat though so i will likely continue to fall behind more. sooner or later i will be forced to make an ncb run as my only option to stay in the 6/20 range. while this is not horrible and i do like having the option for staying in it somehow, i need to get used to the idea of disposable teams as that seems to be what we have here in cb2...at least for now.

Admindudemus[jabberwocky]February 24 2008 12:59 PM EST

we have also often thought that it was at least boring and sometimes annoying that a so many of the top dogs are in one clan as well. the rewards issue might be even worse than it is now though if the top 3 pr actually fought each other. go br!

DukeFebruary 24 2008 1:17 PM EST

Ill like to see your data cuz even before everything got print like you said i could tell you how much would be your 5 potential rewards again someone, and its did include BA cost.

If you want to check this just look at your rewards when you change BA regen rate you will see a net increase of 20% in rewards and yes clan members give extra rewards and yes that bonus is getting boost by NUB ill never test but i think NCB work too.

Synge[Memento Mori]February 24 2008 5:33 PM EST

Dudemus: It should be trivial to stay in the 6/20 range indefinitely as long as you don't take extended vacations. Yes, the top dogs will grow slightly faster, but you only need to be within 60% of them. The reward difference between anyone in 6/20 and the top dogs is much much less than 60%.

Duke: I have yet to see any evidence of this. If I recall correctly, dudemus and Ranger actually provided some data counter to your assertion.

Rubberduck[T][Hell Blenders]February 24 2008 5:38 PM EST

Interesting. Good work.

I have long believed fighting over double your PR gave a bonus but never really tried to make use of it until tournaments came along.

Sadly it may strengthen the case for paring down your gear as much as possible not only to increase your rewards but to deny your opponents too ;)

Ulord[NK]February 24 2008 5:39 PM EST

Love it. Great work :). I too experienced the difference PR has on awards, particularly while playing tourney. Someone with low PR and same score gives visibly less reward than someone with higher PR and same score.

Admindudemus[jabberwocky]February 24 2008 6:14 PM EST

"Dudemus: It should be trivial to stay in the 6/20 range indefinitely as long as you don't take extended vacations. Yes, the top dogs will grow slightly faster, but you only need to be within 60% of them. The reward difference between anyone in 6/20 and the top dogs is much much less than 60%."

people have told me this before, but what i am always looking at is how much the number one spot, which is what the threshold for 6/20 is based on, is pulling away from the pack and how rapidly.

for ease of calculation, let's say numero uno gets 100 points average reward total. if we add a 20% bonus on that for being able to fight higher than anyone else then we get 120 points. i actually then have to get more than the assumed 60 points to stay as i really need .6*120. that comes out to 72 xp points which in reality is getting 72% of the tops rewards instead of the assumed 60%. if i actually have to fight quite a bit lower than number one, it gets worse.

the real problem is that without some change in the top makeup, someone not fighting all of their ba for example, as the gap widens, the balance tips further.

truly only time will tell if this is an issue and much of my data is based on tracking koy's mpr in relation to mine on a daily basis for several months. the gap was widening, even when i used a roe and he did not. during that time i was very even and sometimes even ahead of ranger on battles challenged as well.

i hope you guys are right and i am chicken little, i do however fear that the sky is falling!

Synge[Memento Mori]February 24 2008 9:23 PM EST

Dudemus: The numbers are much smaller than that. I am pretty sure that no one gets 20% higher base reward than anyone else in 6/20. The level flattens out a lot above 1.5 mil PR. In the example I gave above, moving from about 2 mil PR (Ichigo Kurosaki) to 2.7 PR (Anbhas) was only a 5% increase. Clan bonuses can easily be more important than this.

Admindudemus[jabberwocky]February 24 2008 9:47 PM EST

well when ranger and i were checking that sunday, we got a difference of about 16%. the difference in our clan bonuses are usually under 2%.

over the 3 month period i tracked mpr, even when using a roe, i would more often than not lose ground on him. this is why i use the 20% figure. i am not sure what all is coming in to play, but i do not think the 20% figure is an over-estimated one.
This thread is closed to new posts. However, you are welcome to reference it from a new thread; link this with the html <a href="/bboard/q-and-a-fetch-msg.tcl?msg_id=002MSe">Fight Rewards Explained</a>