Change How Level-Up Stats are Allocated
|
Posted 2021-11-11 08:32:25
|
 Zea #27549 |
|
Posted 2021-11-11 08:35:25
I don't mind the occasional misfire for Chasers as much, but my Finishers have consistently been given honestly crap stats in Strength, no matter how carefully I try to make sure it doesn't happen. It's just frustrating, and it's really starting to kill my enjoyment of the game. To the point of me going "why even try? I get screwed over anyway" |
 Ihmislehma #1850 |
|
Posted 2021-11-11 08:47:58
Alternately, there could be some kind of pity threshold - a certain percentage of stats is automatically allocated to the wolf's favored stats (idk, like 30%), and then the remaining stats are distributed by weighted RNG as they are now. Chance would still play a significant role, but it would no longer be possible to get a completely terrible level up. |
 Lionel #34199 |
|
Posted 2021-11-11 08:48:48 (edited)
You can find it HERE! Dzanek went through a lot of math and probability to explain that we shouldn't get annoyed at outliers - and that overall, hunters should gain stats that behoove their roles, or at least, that stats that are the highest. And that the good level ups should balance out the bad ones. So I ran the numbers for my G3 finisher HERE. She was born with 174 strength - 81 points higher than her second highest stat! As she was leveling up, she was a finisher her entire life (I made her a mentor at level 20, so that shouldn't affect her level up gains). Then, I ran the parser on my user log to see where all her stats went during level up. Overall, she got 123 stats in strength, and 125 in Agility. Why? When strength was her highest stat for her whole life? I think it should have allocated more to strength, not only for practical reasons (to make her a better finisher) but also lore reasons (level ups should be an accumulation of the wolf's experiences in their life). You are making the argument that it eventually balances out and the bad level ups screenshots are just us cherry picking bad RNG. However, this case shows that for some wolves, the level-up algorithm consistently overemphasizes the second highest stat. While that isn't terribly detrimental to chasers and stalkers, who use their second stat anyways, it can be disastrous to finishers who are left in the dust when the game decides they should be wise or fast instead of strong. |
Zea #27549 |
|
Posted 2021-11-11 09:39:16
 Just out of curiosity, I decided to run stats on six of my level 20 career finishers. Five out of six managed to get strength equal than or greater to their secondary stat, and the sixth was close. Vice Kyovin 34% str, 27% spd, 15% agi, 11% wis, 12% smarts Castleberry 33% str, 28% spd, 14% agi, 11% wis, 14% smarts Lunar Dynamo 30% str, 27% spd, 15% agi, 15% wis, 14% smarts Moondew Jackal 30% str, 30% spd, 14% agi, 13% wis, 13% smarts Jackfruit Cider 34% str, 26% spd, 12% agi, 12% wis, 16% smarts Cai Kyovin 29% str, 13% spd, 31% agi, 13% wis, 13% smarts That seems pretty okay to me, overall. Of course I wouldn't mind more strength (who would?) but it doesn't seem objectionable. It would still be nice to have some kind of RNG mitigation, especially in early levels, where a random +3 wisdom level up can ruin an aspiring NBW chaser's career. |
Lionel #34199 |
|
Posted 2021-11-13 05:23:41 (edited)
44.29%. Let that sink in. Granted, he is only level 8 right now, but he has been carefully Strength-trained as a pup and adolescent, bred for Strength before that, and is now working as a Finisher, and has worked as such always. His Strength got only 25.71% of all level up stats in it. Hard numbers are +18 Strength and +31 Speed, his secondary stat. Now, yeah. The leveling up may be actually working as intended. It is trying to balance the two stats. But here's the deal: All other roles where stats matter have two stats. Ergo - player will assume those two stats get most level ups, right? But Finisher only has one stat. It implies you do not have to care about any other stat, but this is not true. You do have to care. You will want to cheese it, and bring up the second highest stat to be as close to on-par with Strength as you can.... or this happens. The bigger the discrepancy between Strength and second highest stat, the more stats you will lose at level up to the second highest stat - because the game is playing catch up with Strength. Why is this a problem? Outside of punishing people for going for max Strength and ignoring everything else... it's kind of false advertising. It has not been made clear to players without a lot of technical understanding, or without asking in modbox, that this is how leveling works. A secondary problem will be that if people do start to breed for Strength + another stat... that's a four others to deal with. You will have too many combos for effective breeding. What could be done about this? The options are basically to either edit the leveling up for Finishers specifically (backend math and formulas), to change leveling up in general (backend math and formulas), or to simply give Finishers a secondary stat to care about. I think the last might be easiest technically, but I am no developer. I'm also doing a test soon with a wolf and roles, but that will take time to finish. |
Ihmislehma #1850 |
|
Posted 2021-11-13 20:45:25
my poor Windseeker had a whopping 44.29% of all level up points go into his second highest stat oof yikes I would be livid Hopefully his later levels are less disastrous? I focus on strength only for my finishers, and they have much more strength than their second-highest stat, but their level ups have turned out decent overall. If other people chime in with more data on their level 20 finishers, it would be great - I'm wondering if I just got lucky, or if my results are typical. |
Lionel #34199 |
|
Posted 2021-11-13 22:02:15 (edited)
And here's the ones that are getting me really annoyed. To put it in gentler terms.
He's the son of an NBW Leaderboard finisher that I leveled from 1 to 20 and grinded his mom's stats EVERY. SINGLE. DAY until the last heat.
https://www.wolvden.com/wolf/3316692
Born at 163 str, + Pup training, + full adol training
Starting adulthood at 201 str
This bastard has had 31% of his stats go into WISDOM so far. As a FINISHER. He has actually gained more wisdom than he has strength. This is what's really annoying me and making me want to just stop trying. What's the point?
|
Sɪʟᴀ #37396 |
|
Posted 2021-11-14 02:00:35
This is killing the want to play, simple as that. It's killing the motivation to work on something. That's not how this should go, really. |
Ihmislehma #1850 |
|
Posted 2021-11-14 09:08:50
Party 1: 99% synergy, 199/233 stalker, 200/195 chaser, 316/361 chaser, 329/341 chaser, 460 finisher. Displayed hunting chances: critter trail 81%, small trail 47%. Party 2: 100% synergy, 269/316 stalker, 303/310 chaser, 239/253 chaser, 295/271 chaser, 392 finisher. Displayed hunting chances: critter trail 74%, small trail 42%. The better party, Party 1, has lower synergy so I'm gonna disregard it. If we calculate the average of the party, where the average of stats for stalkers and chasers (don't confuse the average of the party with the average of stalker/chaser stats, these are different) are taken into consideration (i.e. 199/233 stalker has an average of 216, etc.), we have Party 1 with 309.4 stat and Party 2 with 304 stat of average. So Party 1 is better with higher chance when counting averages, okay. If, according to Sepulchre's claim, the sum of stats for stalkers and chasers are counted (i.e. 199/233 stalker has a sum of 432, etc.), we have Party 1 with 526.8 stat and Party 2 with 529.6 stat of average. In this case, Party 2 is supposed to be better when counting sums but it's different. As I collected more data, I realized that even my theory of averages is not completely correct. When the average of two parties is the same, the chances are still different. Maybe luck factor messes up with chances. However, according to the data I provided, I'm inclined to believe that, despite close averages, the party with better finisher has the better chance of success. I've just sent another party, which is Party 1 but I changed a chaser and got him to 100% proficiency, the same biome and season. Party 3: 100% synergy, 242/283 stalker, 266/254 chaser, 349/388 chaser, 232/248 chaser, 496 finisher. Displayed hunting chances: critter trail 81%, small trail 47%. Hunting chances are the same but the average (counting averages of the stalker and chasers) is 325.4 (551.6 if counting sums for those curious). The finisher is better though. In conclusion, I think finisher currently has the greatest impact on hunting success among hunter roles and doesn't require further buffing. Relating to Ihmislehma's post, you claim we're punished for going into max strength as we didn't know before that leveling up favors the secondary stat. But knowing this, would you train your finisher differently? Or would you breed/purchase a different finisher? Your finisher gains 2 stat points daily so you wouldn't train the secondary stat instead of strength to make leveling up favor strength because you'd end up with lower strength for any attempt to focus on the secondary stat. Would you pick different stud which has the same secondary stat as your finisher even with lower strength than your original choice, to maximize the secondary stat to favor leveling up strength? Currently best finisher, Bellatrix, has smarts as the secondary stat with a significant offset from any lower stat. Her father had speed as the secondary stat. It means that secondary stats were disregarded when breeding a finisher who became the best. No finisher that was bred with compatible secondary stat has beaten her. Pup training is the only aspect that could've done differently if we knew. It seems it's optimal to train the secondary stat, instead we used to waste 18 or 27 points into lower stats which matter even less. However, this 18-27 points of difference doesn't make that much difference of strength, though I'm not certain how much exactly, it can be 2-3, it can be less. There's a lot of info hidden from us. We used to not know that breeding T3s with the same base genetics gives the highest chance of breeding T3 pups and still there was a lot of players wasting so much cones to send requests to expensive T3 studs despite different base genetics. There's a lot of RNG factors of values we don't know and investing with expectation of higher profit usually turns out to be a failure, like gene checking NBWs or stuffing guarana on future stud. The mechanic of balancing two best stats is a minor. Giving a secondary stat for finishers to care about doesn't seem to be a good idea. First, it might debuff hunting chances if the average of stats has impact because finishers would end up with less average if the secondary stat is lower than strength. Second, it'd be disadvantageous for current finishers who don't have this particular stat as their secondary. Third, it'd ruin the uniqueness of training finishers, which is not a template of stalker/chaser training. |
 Dżanek #24018 |