PDA

View Full Version : Spading Snatch



MasterE
02-24-2017, 05:00 PM
Isn't snatch such a fun word to say? Let's find out whether it's actually worthwhile to have and use in this game.

To put this to the test, I've decided to run a regular episode stage and count the drops with and without the use of snatch. This is more ideal than an event run because an event conditions can't be easily repeated and results reviewed anyone wishing to confirm or dispute them. I'm primarily counting normals and rares because they drop in such quantities that differences become more significant. Gold and rainbow drops will be counted as supplementary figures, but I predict differences will be marginal that the results aren't as certain to be replicated by peer review.

For most precise variable control, the same team configuration should be for running the stage x times with and without snatch. Anyone who's interested can do the same thing and provide their own results; all you need is any gunslinger and the willingness to run the same stage repeatedly without snatch, and while using snatch an identical number of times for each run. You'd get the most precise results using a stage that drops more of discrete quantities of items, like a story episode or standard level daily. But the gold and rainbow drops from expert levels would cloud the data by reducing normal and rare drops without adding concrete enough comparison figures.

Unregistered
02-26-2017, 03:29 PM
well i can beat the event with D´arc in the 20 Ap stage with 4-9 snatch stack on boss and got the gold chest like 90% of time
but can´t use her in 30 Ap stage so i barely can stack 1-4 snatch with ex skill on other hero in boss and i must say that the chance of drop gold chest is very low there, like 40% sure there more silver drop in these stage , i wish i could 9 stack on her e.e

Ramazan
02-27-2017, 05:21 AM
4 runs gave me 0 gold chests consequently while snatched boss for 8-9 times and used anzu, once.
While:

http://i.hizliresim.com/R0bl0o.png

To sum up, dmm veterans were right, snatch and anzu aint worth, go for d'art if you want her OPOP debuff, only! Otherwise she's as worse as Solomon. Actually, you still can use anzu to bait people who still believes anzu increase drop rates and use your anzu, thus get you more gems.

Unregistered
02-27-2017, 06:00 AM
To sum up, dmm veterans were right, snatch and anzu aint worth, go for d'art if you want her OPOP debuff, only! Otherwise she's as worse as Solomon. Actually, you still can use anzu to bait people who still believes anzu increase drop rates and use your anzu, thus get you more gems.

Behold! Double Anzu +7 stack snatch~!
imgur.com/JkDgdt2

passerby
02-27-2017, 07:12 AM
Snatch increases drop rate not rarity so whether you get gold chests or not via Snatch is irrelevant. It's still RANDOM. Someone could get way more with or without Snatch compared to another in a single instance or through the entirety of play because it's RANDOM.

Can never understand why some people try to get some analysis of RNG. In this game for example, without knowing any of the actual rates from Snatch as well as from each stage, you can't make any sort of analysis that isn't purely observational.


The point in people trying to increase drop rates is to increase their odds of acquire something they want but I forgot the dumbass min-maxers are everywhere these days.

Ramazan
02-27-2017, 08:49 AM
Behold! Double Anzu +7 stack snatch~!
imgur.com/JkDgdt2

tbh, i expected something like a dia+2 gold chests XD


Snatch increases drop rate not rarity so whether you get gold chests or not via Snatch is irrelevant. It's still RANDOM. Someone could get way more with or without Snatch compared to another in a single instance or through the entirety of play because it's RANDOM.

Can never understand why some people try to get some analysis of RNG. In this game for example, without knowing any of the actual rates from Snatch as well as from each stage, you can't make any sort of analysis that isn't purely observational.


The point in people trying to increase drop rates is to increase their odds of acquire something they want but I forgot the dumbass min-maxers are everywhere these days.

Sorry, the whole post of yours made no sense to me. Idc about an unregistered's opinions anyway.

Unregistered
02-27-2017, 09:21 AM
tbh, i expected something like a dia+2 gold chests XD




Yeah haven't seen a single dia chest. I'll keep trying for double Anzu's anyway though just because. :P

MasterE
02-27-2017, 11:02 AM
Testimonials aren't objective tests, and 'observational study' is rife with cherry picking; both are prone to confirmation bias. I'm going to have to start a new thread to clearly define the objectives and methodology and hopefully cut the spam. You are all sure of your opinions, but they aren't informative even to you.

sanahtlig
02-27-2017, 11:32 AM
Testimonials aren't objective tests, and 'observational study' is rife with cherry picking; both are prone to confirmation bias. I'm going to have to start a new thread to clearly define the objectives and methodology and hopefully cut the spam. You are all sure of your opinions, but they aren't informative even to you.
Your arrogance, towards me and the community in general, is irritating. While you occasionally make valid points, your attitude makes it difficult to engage with you in anything but a confrontational manner.

Allow me to clarify your point here. The problem isn't with 'observational studies', which you seem to have so little regard for. You HAVE identified an important bias, which is termed reporting bias (https://en.wikipedia.org/wiki/Reporting_bias)--in this case, the tendency to report only extreme results (e.g., a streak of no drops or a streak of valuable drops). Correcting for this requires selecting the subjects to be studied in advance, before the outcome (the drops they get) is known. You are correct that failing to correct for this will tend to result in biased anecdotal data that isn't useful. BUT if this is corrected for in advance, crowdsourced data can be a useful asset, and you would do well not to dismiss it.

Unregistered
02-27-2017, 12:03 PM
i just got apocalypse from a platinum box in 30 ap quest with 7 stack of snatch maybe it was luck but i want belive that was for the snacth

Ramazan
02-27-2017, 12:15 PM
Testimonials aren't objective tests, and 'observational study' is rife with cherry picking; both are prone to confirmation bias. I'm going to have to start a new thread to clearly define the objectives and methodology and hopefully cut the spam. You are all sure of your opinions, but they aren't informative even to you.

My main had d'art+too many anzu friends (deteled ALL of them an hour ago)
My alt have sieg, never used snatch before+0 anzu friends
I played since start to "observe" enough and i dont give a damn about your methodology and&or objective things, since my claims were "ALREADY" verified by dmm veterans, who played for more than 9 months, even if im wrong. Anyway it'd be great to see how you explain this despite all that, methodology professor.

- - - Updated - - -


i just got apocalypse from a platinum box in 30 ap quest with 7 stack of snatch maybe it was luck but i want belive that was for the snacth

Yeah, thats the whole point of the matter. And that also reminded me of:

http://i.hizliresim.com/brgqg8.png
http://i.hizliresim.com/X0DVD7.png
http://i.hizliresim.com/Qa4Dgk.png

MasterE
02-27-2017, 12:23 PM
Ram, if it was verified, then show me the data. 'Verified by dmm veterans' only means verified by the ones you cherrypicked to verify what you want to believe. The opinions there, even if they were as unilateral as you claim (you believe there's general consensus,) still aren't an objective test, and no one at discord or here has yet provided factual evidence.

Ramazan
02-27-2017, 12:52 PM
I'm going to have to start a new thread to clearly define the objectives and methodology and hopefully cut the spam.

Can you just skip the comment phase and do the action? We all already debated over this matter at discord before.

MasterE
02-27-2017, 02:48 PM
Can you just skip the comment phase and do the action? We all already debated over this matter at discord before.
This thread wasn't intended to open debate, either. The OP outlined the intended test parameters and explained what contributions could be accepted. You can't discuss objective procedure with anecdotal + bandwagon fallacy, appeals to authority/popularity, appeals to emotion. I'm pretty sure whatever debate you claim has gone on was settled by argumentum ad nauseum rather than supporting fact.

sanahtlig
02-27-2017, 03:14 PM
This thread wasn't intended to open debate, either. The OP outlined the intended test parameters and explained what contributions could be accepted. You can't discuss objective procedure with anecdotal + bandwagon fallacy, appeals to authority/popularity, appeals to emotion. I'm pretty sure whatever debate you claim has gone on was settled by argumentum ad nauseum rather than supporting fact.
These are the components to professionally judging a research proposal (https://grants.nih.gov/grants/peer/guidelines_general/Review_Criteria_at_a_glance.pdf):
Significance: How the findings contribute to the overall body of knowledge and benefit the public. The importance of the question being asked.
Investigator(s): The credentials, experience, and accomplishments of the people proposing the investigation. This predicts likelihood of success.
Innovation: Whether the proposal investigates an unexplored problem, uses and develops novel methods, or attacks the problem from an unexplored angle. The novelty of the proposal.
Approach: Whether the proposal is methodologically sound and doable.
Environment: Whether the investigators have access to the tools, expertise, and/or mentorship they need to carry out their proposal.

A methodologically-sound approach is only one of the criteria for a good research proposal, and it's actually weighed pretty low. Your proposal is methodologically-sound, but lacks significance and innovation. No one cares about how Snatch affects the drop rate of common items from quests. They care about how it affects certain event drops. Snatch might affect both similarly--or it might not. Who knows! Furthermore, there already exists a general consensus that Snatch doesn't make much of a difference. Your contribution would be pinning down exactly what that percentage difference is. You're going to need a lot of power / data for that, which means the study is unlikely to reach a satisfying conclusion. Furthermore, as an investigator you've shown a tendency to ignore constructive criticism, and you haven't displayed any experience or background in statistics, and thus your findings would likely not be accepted by your peers.

For all of these reasons, I rate your proposal as mediocre: unlikely to contribute substantially to the body of knowledge, and likely far more effort than it's worth.

Unregistered
02-27-2017, 03:48 PM
Came someone tell me what this spading is? I've never heard the word used with statistical info. Is it regional jargon or something?

MasterE
02-27-2017, 04:33 PM
For all of these reasons, I rate your proposal as mediocre: unlikely to contribute substantially to the body of knowledge, and likely far more effort than it's worth.Do I need to tell you again that you aren't a qualified observer? Aside from the heavy bias, your assessment is shot through with more logical fallacies than I can name right now, though if you'd like I can take an inventory for you.


Came someone tell me what this spading is? I've never heard the word used with statistical info. Is it regional jargon or something?It's not specific to statistics, it's a general gaming term for discovering the game environment and mechanics. Statistical spading isn't my favorite kind, but if no one else has the will to learn the mechanics, I do.

sanahtlig
02-27-2017, 04:46 PM
Do I need to tell you again that you aren't a qualified observer? Aside from the heavy bias, your assessment is shot through with more logical fallacies than I can name right now, though if you'd like I can take an inventory for you.
Sure, go ahead.

Unregistered
02-27-2017, 06:40 PM
It's not specific to statistics, it's a general gaming term for discovering the game environment and mechanics. Statistical spading isn't my favorite kind, but if no one else has the will to learn the mechanics, I do.

Thanks. Never heard it with any mix max groups that I've seen though. Must be more specific to some gaming groups you interact with that I don't.

Dragonlich
02-27-2017, 09:12 PM
i just got apocalypse from a platinum box in 30 ap quest with 7 stack of snatch maybe it was luck but i want belive that was for the snacth

Lucky you.
Platinum chests is good and I get one with snatch. But it gives me that shitty silver axe.
But I am bad with random, so I got a lot of elemental tablets from platinum raid's chests...

Unregistered
03-01-2017, 05:03 PM
My main had d'art+too many anzu friends (deteled ALL of them an hour ago)
My alt have sieg, never used snatch before+0 anzu friends
I played since start to "observe" enough and i dont give a damn about your methodology and&or objective things, since my claims were "ALREADY" verified by dmm veterans, who played for more than 9 months, even if im wrong. Anyway it'd be great to see how you explain this despite all that, methodology professor.

well dude this is my 2 apocalypse and i have wasted days out of 30 ap map because i was low lv but i can beat her with d art, your veteran players don´t always right
LOL can´t post lickadd the first part>> ://imgur.com/a/Xne2Z

Ramazan
03-01-2017, 06:22 PM
well dude this is my 2 apocalypse and i have wasted days out of 30 ap map because i was low lv but i can beat her with d art, your veteran players don´t always right
LOL can´t post lickadd the first part>> ://imgur.com/a/Xne2Z

And i got SSR weapon without snatch&anzu. So?

Unregistered
03-01-2017, 08:26 PM
And i got SSR weapon without snatch&anzu. So?
it Up % to get it, and i´m sure that u have used tons of pots since the star of the event to get these 1 time luck weapon, your veteran bulshit are wrong about snacth don´t help with anything

President Ramu
03-01-2017, 10:25 PM
it Up % to get it, and i´m sure that u have used tons of pots since the star of the event to get these 1 time luck weapon, your veteran bulshit are wrong about snacth don´t help with anything

I used literally zero pots and quit playing after Sunday, and I still pulled an SSR Eidolon from a chest. I'm not saying Snatch is zero help, but what you're saying right here is flat-out wrong.

Unregistered
03-01-2017, 10:40 PM
Came someone tell me what this spading is? I've never heard the word used with statistical info. Is it regional jargon or something?

It comes from the Bartle test.
en.wikipedia.org/wiki/Bartle_taxonomy_of_player_types
The short of it is, it divides gamers into 4 types of behavior: killers (clubs), achievers (diamonds), explorers (spades), and socializers (hearts).

Ramazan
03-02-2017, 05:20 AM
it Up % to get it, and i´m sure that u have used tons of pots since the star of the event to get these 1 time luck weapon, your veteran bulshit are wrong about snacth don´t help with anything


It comes from the Bartle test.
en.wikipedia.org/wiki/Bartle_taxonomy_of_player_types
The short of it is, it divides gamers into 4 types of behavior: killers (clubs), achievers (diamonds), explorers (spades), and socializers (hearts).

I think its your brain which is the bullshit here, and needs immediate repair. I never used any pots since the event began because im saving my pots for next event which is the lightning&thunder which i have more kh of. Dunno why im arguing with an unregistered tho, probably because of my free time...

Also dunno why all these veteran bitching...

http://i.hizliresim.com/vbgoY4.png

Unregistered
03-02-2017, 08:43 PM
You do realize that those are two different people commenting, right?

- - - Updated - - -


You do realize that those are two different people commenting, right?

He is a turk, give him a break.

Ramazan
03-02-2017, 09:39 PM
http://i.hizliresim.com/R0JOXG.png

Poor unregistered :) couldnt delete your own message are you :)

I took that ss for future reference if another unregistered troll like you shows up.

Unregistered
03-03-2017, 01:36 AM
Poor unregistered :) couldnt delete your own message are you :)

I took that ss for future reference if another unregistered troll like you shows up.

Hello Ramazan, I would like to assist you in understanding how unregistered posting works.

1. If two different people both post with the username set to the default Unregistered it compacts them into the same post with an updated between it.

2. Unregistered cannot delete or edit their posts.

3. Not every unregistered is the same person.

4. The only way to tell if it is different posters is by their writing and posting style.

I hope this helps you since it seems that you are struggling with understanding how it functions.

Discoceris
03-04-2017, 02:46 PM
Hello Ramazan, I would like to assist you in understanding how unregistered posting works.

1. If two different people both post with the username set to the default Unregistered it compacts them into the same post with an updated between it.

2. Unregistered cannot delete or edit their posts.

3. Not every unregistered is the same person.

4. The only way to tell if it is different posters is by their writing and posting style.

I hope this helps you since it seems that you are struggling with understanding how it functions.

He's mad. Let him cool off for a bit. Where's a mod when you need them? This thread went off the rails a few yards back and need to be locked up.

Anyway, for people new to statistics, and game mechanics, here's a short primer.

% chance drop rate increase - It just means exactly that. There are times when you may not have ANY drops (I've observed this a lot in this game already, although I'm wondering in the instances of a raid, where you need to have a minimum amount of turns and/or % damage to be registered for loot table (i.e., people getting in on a boss fight and by the time they defeated the lag monster, are only there to witness the last 5% of boss health to be carved up by the newest 5 members to the raid)

% rare item appearance increase - This means that the rate of appearance for normal loot is suppressed so that the appearance for rare loot can increase. Think of it as this way: the values must always add up to 100%. You cannot have 101%, or more, total (combined) rates.

In every case, without exception, the rates given for anything, whether drops from boss fights or from gachas, do not guarantee anything. It means that for any given event, there is a table governing the appearance rate of any item, and is not connected to any past or future event. In other words, if a drop rate for something is 5%, it DOES NOT MEAN that you are going to get the item after 20 attempts, or 100 attempts, or even a 1,000 attempts. It just means that on any given drop, the chance for that item to appear is 5%. No more. No less.

The exception is on gachas where they promise you at least SOMETHING on particular rolls (Roll 10 and get a 5 star free) You still aren't guaranteed anything, unless it explicitly states it does.

Just understand that for every one of these games, there was a lawyer involved.

sanahtlig
03-04-2017, 11:48 PM
In other words, if a drop rate for something is 5%, it DOES NOT MEAN that you are going to get the item after 20 attempts, or 100 attempts, or even a 1,000 attempts. It just means that on any given drop, the chance for that item to appear is 5%. No more. No less.
What a 5% drop rate means is that, over the course of 100 attempts, on average you'll see the item drop 5 times. If you were to repeat this course of 100 attempts many many times, you'd see a histogram approximating a bell curve where 5 drops was most frequent, followed by 3 and 6, then 2 and 7, etc. It wouldn't be quite symmetric because you can't get fewer than zero drops.


% rare item appearance increase - This means that the rate of appearance for normal loot is suppressed so that the appearance for rare loot can increase. Think of it as this way: the values must always add up to 100%. You cannot have 101%, or more, total (combined) rates.
I think this applies more to Gatcha than battle. 10 draws in Gatcha will get you exactly 10 things--the likelihood of getting one thing is directly related to the likelihood of getting something else (because you can't get both in the same draw). But in battle, the drop rates could be completely independent. You could get any number of chests, and getting one item doesn't necessarily affect your chance of getting any other item in the same battle, or even from the same mob. Or maybe it could, depending on the actual implementation in the game code.

Skulkraken
03-05-2017, 02:48 AM
Question:

If a base drop rate of, say, 10% were to be boosted by 10% with Snatch, would the resulting drop rate be 10 + 10 = 20%, or 10 x 1.10 = 11%?

jazz154
03-05-2017, 03:07 AM
The biggest problem is that game itself doesn't really tell us how much of boost snatch gives us. It's like asking us to believe that this skill does something without delivering any evidence. (as far as I know it might as well do nothing)

sanahtlig
03-05-2017, 08:54 AM
Question:

If a base drop rate of, say, 10% were to be boosted by 10% with Snatch, would the resulting drop rate be 10 + 10 = 20%, or 10 x 1.10 = 11%?
To rephrase the question: If Snatch boosted an item's drop rate to 11% from 10%, would that be a 1% bonus across the board, or a 10% multiplicative bonus to each item's drop rate? Or some variable rate applied to each rarity tier? Does it affect the rate of appearance of platinum chests in event battles? What about the contents? We don't have the data, or even the intuition, to begin to answer these questions, though a fixed-rate multiplicative bonus seems most likely.

Unregistered
03-09-2017, 04:58 PM
In the end it helps... just its not really noticeable... because in the end if you were destined to get a bronze or nothing from the start... then you get a bronze with snatch... how would you even know if were going to get nothing from the start? Playing the guessing game really is pointless lol. It helps. but not in a "Huge" way. It's like using steal in a mmorpg... it doesn't always work... and doesn't mean your gonna get the best item in the game from that mob... wow i got 5% of the gold i'd get for killing him... or 1 piece of shit cloth... yay...............................................