Sitting like a Log - Manipulation of the PvP Point System

[Sparton_LOTB]

Followup up from my last comment above... my preliminary math says no, having big gains for defense holds and very little losses for defense losses would not stop mediocre or decent players from just paying to take the top. As the amount of attacks approaches 2x what a "good" competitor would do, the amount of points they'd gain overall exceeds what players with better win rates could do.

If we make the assumption that they're at the top of the leaderboard (avg 12 point attack wins, -20 point attack losses, 25 point defense wins, -5 point defense losses):

• "Hyper-Garbo" player with 1000 attacks (75% win rate) and 1000 defenses (5% win rate) gains 4000 points from attacks, loses 3500 from defenses. Total gain of 500 (yeah, these guys still aren't competing)
• "Hyper-Mediocre" player with 1000 attacks (90% win rate) and 1000 defenses (5% win rate): +8800 from attacks, -3500 from defenses. Total gain of 5300 (a decent chunk)
• "Hyper-Good-o" player with 1000 attacks (92% win rate) and 1000 defenses (15% win rate): +9440 from attacks, -500 from defenses. Total gain of 8940 (after accounting for starting at 1900, this would almost break our current top VP record)
• "Good" player with 500 attacks (95% win rate) and 500 defenses (20% win rate): +5200 from attacks, +500 from defenses. Total gain of 5700 (narrowly better than "Hyper-Mediocre"; that means with slightly more participation, "Good" loses to "mediocre who plays a lot")
• "Great" player with 500 attacks (98% win rate) and 500 defenses (30% win rate): +5680 from attacks, +2000 from defenses. Total gain of 7680 (Better than "Good", but worse than "Hyper-Good-o")
• As a fun bonus, a "Lazy Great" player with 250 attacks (98% win rate) and 250 defenses (30% win rate): +2840 from attacks, +100 from defenses. Total gain of 3840 (worse than more engaged "Good" and "Great", but also miles below "Hyper-Garbo" and "Hyper Good-o"

(Don't mind my name archetypes, they're purely for discussion purposes. If you put in 150+ battles each week, you're awesome and engaging at a healthy level in my books.)

In short, the hyper-engaged players trump players with far better win rates if they have the time, patience, and Ironite to get a lot more matches. Of course, if the "Good" and "Great" players start spending as much Ironite and time to get to close to the amount of attacks the less-good hyper-engaged players, they will make better gains, but that's going to happen regardless, and happen more under something like the current system, where defense win rate has far more of an impact on your score (or more accurately, how often you lose will pull you down more).

As a fun experiment, I took a look at what would happen if defense losses didn't take away any points... naturally, this favours the hyper-engaged to a far greater degree than players with less battles but far superior win rates ("Hyper-Good-o" gains a staggering 13,190 points, and even "Hyper-Mediocre" gets 10,050 points and edges out the "Great" player's 9,430 points).

As for the strategy of being super patient and only taking matches that give +20 points... well, definitely let us know how that goes, Nicko, but that does look like it's competitive under the current system if we don't account for the Saturday rush (booooo). However, it stands no chance under the system where defense wins give far more points... or in a world where you just lose a bit less points for defensive losses, which I hope we can move to soon.

(Also, it's worth keeping in mind that all but #1 is not going to get 12 points for a win unless they're leaps and bounds ahead of all other competitors... you guys will have a better sense for this than me, but I suspect an average of 13 or 14 is probably more realistic, if you assume rarely getting someone above you in points and occasionally getting someone around/a bit below you in points.)

[slauki]

Point System

Why don't you create a contest for the community to build a better point system. If you tell us what you want to archieve, with the different arcetypes i bet someone could come up with something usefull.
Swarm intelligence for the win

[Patrice-1201]

@Sparton i think there aren't any solutions as long as everybody is mixed up. IMHO you should create a league system where top players with your current point and reward system per division and add a promote to higher division and demote to lesser division once a week.

Another way would be to segment players in conferences a la NFL kind of thing and ultimately compete for the SUperbowl arena once a quarter or something like that.

[Tritium]

This whole discussion is great! I thoroughly enjoyed seeing the community debate together in order to figure out the best way to change the current PVP system. Sparton's input is also a huge plus as well; I am glad that he is as communicative as he is. I hope that, as slauki mentioned, all of our inputs together can help refine the PVP system in such a way that it still properly rewards the best players, without egregiously punishing those who don't have the time to put in hundreds of PVP fights a week. This whole thread has been refreshing. Thank you, Nicko, for making it happen!

[Enti]

This might not be a well-thought suggestions, but the idea came to my mind during shower and I'm in a hurry to get to work:

Wouldn't a flat rate of 12 points on defense loss solve this whole thing?

PoV of low rank players: they lose 12 points anyway from people 300+ VP over them
PoV of high rank players: instead of going backwards with 70% of their attacks, they just don't advance in case of a loss, which can still be countered with participation.

Yes, this would open up again that more attacks grant you higher rank, but not to the extremes we witnessed in the first three months

[Sparton_LOTB]

Why don't you create a contest for the community to build a better point system. If you tell us what you want to archieve, with the different arcetypes i bet someone could come up with something usefull.
Swarm intelligence for the win

I'm 100% open to suggestions, like what Nicko has shared here; even if the proposal he mentioned isn't something I think would be viable after crunching some numbers, there's definitely merit to it, and maybe a better solution that is along the lines of better rewarding defenses could work.

@Sparton i think there aren't any solutions as long as everybody is mixed up. IMHO you should create a league system where top players with your current point and reward system per division and add a promote to higher division and demote to lesser division once a week.

Another way would be to segment players in conferences a la NFL kind of thing and ultimately compete for the SUperbowl arena once a quarter or something like that.

League systems are interesting, but they tend to create fiefdoms and people playing in weird ways to try to be the kings/queens of their own fishbowls (look up strategies for competing in King of Thieves, for example). For large scale, always-around, individual vs individual affairs like our Arena, it's not something I'm a big fan of, but we may do something like that for future competitive features.

While obviously some people will do funny business even within the current kind of point system (such as putting in a purposefully weak defense team to tank their score so they can get easier wins), those are usually pretty binary affairs of "trying to compete or trying to farm easier", and that doesn't have much of a negative impact from most people's perceptions.

This whole discussion is great! I thoroughly enjoyed seeing the community debate together in order to figure out the best way to change the current PVP system. Sparton's input is also a huge plus as well; I am glad that he is as communicative as he is. I hope that, as slauki mentioned, all of our inputs together can help refine the PVP system in such a way that it still properly rewards the best players, without egregiously punishing those who don't have the time to put in hundreds of PVP fights a week. This whole thread has been refreshing. Thank you, Nicko, for making it happen!

I agree. I have to say that I also greatly appreciate the receptiveness and patience you guys have had with us as we continue to refine the system and discuss ways of making it better (and chipping away at fixing the bugs...). My thanks to you too, Nicko!

This might not be a well-thought suggestions, but the idea came to my mind during shower and I'm in a hurry to get to work:

Wouldn't a flat rate of 12 points on defense loss solve this whole thing?

Hm... that is also an interesting idea. Running more numbers against the player archetypes I posted above, that definitely does favour people with good attack and defense win rates more than those with lots of attacks but mediocre/bad win rates... but players with a good attack win rate but a mediocre defense win rate can still get a lot of points, even if those with either or both win rate being poor are just hooped. Since it takes some of the sting out of losing against weaker players, that could encourage mid-ranking players to pick on weaker players to get more consistent wins from a high win rate, which doesn't sound ideal, though.

I'll have to sleep on that. Good suggestion.

[Tritium]

Sparton, would it be possible for us to have the option of choosing our defense team to always go first when testing it? Winning the coin toss can be rather annoying when all we want to do is see how our defense performs when it goes first.

[BillLion]

Thank you Nicko for launching a needed, healthy conversation!

Well, as I noted in the week 11 thread, that is part of the reason why we're looking at adding in logic that reduces point losses for defenders when they lose, as intentional or not, lack of participation can at least feel rewarding in some ways.

I missed this in the other thread but this makes me very encouraged. While enjoying many aspects of it, from the beginning I've been like a 2-string guitar re: PVP critique: 1) Reduce defensive losses & 2) Improve rewards. From pieces I've read it seems both are on the way. These things are huge to me as you attempt to strike the right balance all around.

One additional suggestion is to not introduce any more new talismans until we can also get some of the bugs resolved. It is dizzying to have to constantly change defenses because an update bugs passive on artillery dog, new OP talismans essentially make you have to bench the troll & void talismans, etc. The variety of characters, talismans and combination possibilities is already quite rich. This is good. But repairs should precede new implementation IMO.

[Enti]

One additional suggestion is to not introduce any more new talismans until we can also get some of the bugs resolved. It is dizzying to have to constantly change defenses because an update bugs passive on artillery dog, new OP talismans essentially make you have to bench the troll & void talismans, etc. The variety of characters, talismans and combination possibilities is already quite rich. This is good. But repairs should precede new implementation IMO.

Yup, I depleted my skill shards for reactive changes on my defense and whatever might come up next I won't be able to keep up with. I can't even imagine the struggle of newly joined players who wish to participate in PvP. If they want to be competitive they have to shard characters essential for Arena and not what they like to play with. Not to mention the difference in the amount of shards obtained.

[Jofer16]

Well it seems my attacks have finally stabilized some after my early week experimenting binge. Didn't wake up to 40-50 attacks and I'm starting to stabilize in rank. I just wish I could play more. Perfect solution for me would be one that doesn't penalize participation yet doesn't allow mr garbage money bags player like Sparton points out to easily pay to win. Now to figure out how to get there