Self-adjusting Loot Based on Popularity

[Garth2079]

I don't know about others, but I tend to see the same adventures being completed all the time (and the same ones ignored). Roaring Bull and Black Knights are the current favourites, because they offer the best value (now cheap to buy, 2 lootspots, low losses). This will always be the case - as the balance changes, other adventures may become more attractive, so the variation in adventures played is low.

I don't expect BB to keep monitoring and tweaking to try to balance the adventures and get more people playing more adventures. Why not get the code to do it for you?

Here's what I would do:

- Count the number of adventures available. For the sake of simplicity, let's say there are 21 different ones.
- Create an adventure with an initial popularity of 0. Each time someone plays it (and completes it) the popularity rises by 20, and each other adventure falls by 1 (so basically, the adventure takes a popularity point from every other adventure).
- Use this popularity figure to adjust the available loot. As the popularity goes up, the loot decreases very slightly. So, for instance, 500 granite in SFTR might become 499 granite.
- Conversely, unpopular adventures find their loot goes up. SOTV would increase significantly in the early days.

That's it - as simple as that. After a few days, Roaring Bull would be less attractive as the loot dropped, and SOTV would become more attractive. As people move away from RB, the popularity figure would drop and the loot would increase. People would naturally vary the adventures, for instance they might suddenly discover that Secluded Experiments was paying 2500 granite, word would get around, more people would complete it, then VtV would become the best adventure.

The best thing of all is that it requires no further interaction from the developers. It is self-balancing, like prices in the trade office. No need to accurately predict how much loot to offer. Just get it about right and let the popularity count fine-tune it.

It's also pretty simple logic to code. Make your base figure "1" and increase/decrease it by 0.01 (or 0.001, I don't know how often it is played per day) each time, then you just have to multiply the random loot e.g. 500 granite by the popularity figure, then round it.

This principle could be introduced elsewhere in the game too.

[Peajay]

I think you'll find the popular adventures can be completed in around an hour (with enough generals) and have around 1000r losses, where the unpopular adventures cost a lot more troops and take much longer because of the multi wave camps. Loot is the second reason why some adventures are favoured over others. I have suggested a rework of loot tables and maybe even removal of some camps in some adventures, but it looks like nothing will change.

Lastly, at end-game its still exactly the same, only 1 fairy tale adventure, roaring bull and BK are popular at level 50 (possibly traitors and IOP too)

[Dorotheus]

Which adventures get played is dictated by need to some extent. If you need granite to upgrade your buildings then your not going to be playing adventures with no granite drop when given a choice.

There are also external factors which drive the popularity of adventures such as the cost of the adventure in trade or the price of a loot spot or just the fact that it allows 2 or 3 players to complete their guild quest. This last on my server has made demand for IOTP outstrip supply.

[King-Fero]

The goal of Garth's idea can be reached without the elaborate mathematics for loot adjustment. Now that the new provision house is still half empty, you could add items, a bit like the collectibles, but more valuable in there. The materials would be dropped from adventures on top of the loot. Let's, for now, refer to them as tokens.

So the provision house items are valuable enough to be desired by at least the richest players. They are willing to pay for the tokens to produce the items. But if an item costs 20x VtV token, 20x BK token and 20x Gunpowder token, there's an imbalance on how many of each token is in the market. BK being a popular adventure will have more tokens around than the others. But those tokens are worthless unless paired up with the other materials! Thus the market price of the VtV and Gunpowder tokens rises. Higher value of token = higher value of total loot.

For simplicity's sake, let's estimate BK and VtV losses as roughly the same, but loot value of VtV is 200gc lower (doesn't have to be true, this is an example). Now, several lv50 players already have 20 or even 100 BK tokens but no VtV tokens. They might switch from playing BK to playing VtV in order to get the tokens, or they could be paying a high price for the VtV tokens in the market. This price would add up to the 'base' value of VtV loot and thus it would close the gap between it and the more popular BK. If a VtV token's market price went up to 200gc, the imbalance between the profitability of the two adventures would be completely fixed.

There's two reasons why I find this approach appealing. For one, it is technically more simple to implement. You add an item to each adventure's loot without the need to tamper with the rest of the reward system, and you add prizes to the rarity provision house (they could even be existing ones, like floating res, rainbow bag etc) for those items. No new functions or loot adjustment formulas needed, just add this bit to the already existing system. Secondly, players can't plan on their game play as efficiently if the loot drop keeps changing. In this alternative the loot stays fixed, it is the market value that changes. What's the difference though? I could check the TO to see what's being offered for each token when thinking about starting an adventure, and it would give me an idea what the value of the loot will be. If the loot was changing due to popularity, and the popularity would obviously be changing all the time, it would be impossible to keep tabs on each adventure's loot drop.

Either way, a balancing factor between the worth of adventures is more called for than ever. Earlier you might have decided to play VtV in stead of BK just because it was practically free to buy, whereas BK cost 1500gc. So even if VtV had cheaper loot and higher losses, you could balance it out with the lower adventure price. Now, any adventure can be acquired for 200-300gc, so there is no longer anything to be gained by going for a less profitable adventure.

[Bluesavanah]

The ideas put forward have merit, I think the only way your going to make these adventures popular is to drastically change the loot tables.

I'll give vtv as an example column 1 remove the titanium, increase granite and logs by 30%, column 2 100% increase on marble and planks, column 3/4 50% increase in weapons dropped, column 5 change to 200/400/600 settlers, column 6 increase gold coin to 200/400, gold refill 500, iron refill 2k wheat refill 2k. Then add small chance to drop column 7 for ornaments.

Suddenly it becomes a lot more appealing, what you could do with the others I don't know because tbh the time investment and troop losses you'd be better off doing a FT adventure.

[Gerontius]

Simple and effective

But will never happen because it implies there is something wrong with the current loot tables

[Sharpielein]

Somehow, they should make it so that adventure loot is proportional to troops lost when clearing out every single camp with regular troops.

That is, for each settler that would be lost, there should be at least a half settler's value (currently roughly 0.6 coins) for the average loot.

[Garth2079]

The ideas put forward have merit, I think the only way your going to make these adventures popular is to drastically change the loot tables.

I'll give vtv as an example column 1 remove the titanium, increase granite and logs by 30%, column 2 100% increase on marble and planks, column 3/4 50% increase in weapons dropped, column 5 change to 200/400/600 settlers, column 6 increase gold coin to 200/400, gold refill 500, iron refill 2k wheat refill 2k. Then add small chance to drop column 7 for ornaments.

Suddenly it becomes a lot more appealing, what you could do with the others I don't know because tbh the time investment and troop losses you'd be better off doing a FT adventure.

But the core problem remains the same... unchanging loot tables means people work out the most efficient adventure and repeat it. You can only get a balance by monitoring how many adventures are being completed, judging popularity based on that metric, then readjusting (which is basically what you are suggesting, I think). Why not let the code do that for you? Sure, some people might not like loot tables that go up and down... but don't they effectively do that already, with the random element? I'm just suggesting to combine the random element with a popularity-based element, in order to get us playing different adventures.

[Dorotheus]

There are two flaws with your view of popularity. First not all adventures are as easy to obtain. Some can easierly be found with explorers, others are only availble in the shop and cost map fragments or the even more expensive gems. How often have you seen Nords2 in TO, or players offering it's loot spots. Secondly players are not doing adventure X because they want to, they are doing it because they have a daily or guild quest which demands they do it. When the guild quest just asks for any adventure they are often choosing loot spots or adventure's with low troop loses to preserve their troops for the guild quests which demand particular adventures. Why should a player see deminishing returns because the popularity vote has decided worse loot because of the games constant demands to do a particular adventure.

[King-Fero]

Nords 2 hasn't got a lootspot FYI.