New Moon/ Full Moon

Post some type of testing that backs this. I always get good results myself when I rub my lucky rabbits foot.

Mikesjustice wrote once that after having leveled every single craft to 100, he didn't notice any statistically significant difference regarding direction (might have also said day, but definitely said direction).

Since Mikesjustice is probably the most epic crafter ever in FFXI, I tend to take his word on stuff more than I might with other people.

As far as Moon Phases, my own feelings from Fishing are that New Moon is better for skillups, and Full Moon is better for bite rate. I once did 4 or 5 levels of skill in one New Moon while fishing. Granted that was at low levels, but still, that's a ton of fishing skillups.

For Crafting, I agree with Darksday for HQ and Lightsday for Success, but I'm still torn on the moon phases.

Just sayin, people do this. I've seen people cast barspells on themselves and pretty much everything else I listed in my rant. Whatever helps YOU feel you are crafting better or getting skillups faster than do it. But don't tell other people to do it like this until you have substantial evidence that proves its right.

The fact is there is NO hard evidence to say whether or not HQ are more frequent during a full moon or a new moon, or on this day or that day, its all speculative. I personally have found over the last 5 years of bone crafting that crafting on the day of the crystal I'm using during a New moon seems to yield my best results, but this could be complete luck, and more then likely IS since this method seems to work for me but not for a lot of others.

I didn't tell anyone to do anything. I said it may be superstition, but I...

Ok going to put all my marbles on the table. I have 100smith, 100Leather, 86gold, and 91bone, each one has all others to 60 cept smith which doesn't have bone finsihed, i don't count food or fish lvls. I have a combined 1,765 levels of crafting. So i have a little experience in skilling up, and i wish to share with you some of my knowledge. I apologize in advance if i ramble.

There is and never was any solid evidence to back-up any theory or idea of how to better your craft EXCEPT to increase your level. The generally accepted theories are based solely on peoples observations. A crude but easy example:

A crafter doesn't try skilling up until they collect 50 synths worth. Today they got 50 and it just happens to be New Moon. Said crafter always records there batches, and they average out 2 full levels of skill. Today they got 6! Way more then they ever got before, so this time when they have another 50 they wait for New Moon again. This time they got 4 levels. All of a sudden new moon is the hot ticket.

This was very crudely thrown together BUT nonetheless it shows how these theories came to be.

As skill ups go i have my own little ideas and what i have found to work for me. I have had almost as many equally bad batches as i have good batches using the same day/moon. Some times it was hot and some times it was not. Nothing in any of the current theories out there have proved themselves. Prior to 60 i didn't see any consistency with day/moon/direction. Post 60 i found one thing to be consistent and that is if i get 3~5 skill ups within my first 5~8 synths of the batch the next 10~20 would have 0~1 skillup. I've had this happen to me frequently and when it does i will stop skilling for a few hours. Now did I lose a possible skilluo festival? Maybe, and maybe not. I want to make this very clear that this is my niche and in no way am i selling this as more then what it is.

As for HQ i hold zero reservations and have currently found nothing to be my a prime target day/moon/phase. I have always done my crafting in large batch sizes like last week i did 16 haubergeons with 1HQ and 15NQ and an overall net profit of ~500k. I have done hauby batch of 15 that produced 2 broken D.ing and 13 NQ haubys, this batch cost me 2m in net loss. Also to note the 16batch was done on Thundersday, 69% Waxing Gibbous, and my 15batch was done on Windsday, 2% New Moon (suposed prime HQ day)

To sum up the key points of this thread. Crafting is the great unknown in FF11, and it would be prudent to treat it as such until there is fact. Find your own niche, craft in larger batches, and above all else remember that HQ crafting is a roll of the dice and more times then not you will lose.

Crafting is streaky, like any other chance-based thing. You can't expect to necessarily see the trend over the course of a few stacks. However, over the long-run, you can find certain trends.

I believe this is still the standard quoted source on HQing, at least last time i checked (it's from 2004! how can you call yourself a crafter and haven't read this yet?):

http://ffxi.allakhazam.com/forum.html?forum=24&mid=1093897190388898652

It's based on data from over 2,000 synths at the "31 levels higher" HQ tier, and the basic results were (in percentage of HQ):

Full Moon 18.49%
Quarter Moon 21.88%
New Moon 39.58%

and (with a wind crystal synth--remember ice > wind > earth):

Darksday 38.89%
Earthsday 18.06%
Firesday 29.86%
Iceday 39.58%
Lightningsday 24.31%
Lightsday 16.67%
Watersday 24.31%
Windsday 21.53%

So the best HQ results were achieved on Darksday or the day that is strong to the crystal you're using, on a New Moon. The worst HQ results were on a day that the crystal is strong to, Lightsday, and on a Full Moon:

Darksday New Moon 58.33%
Iceday New Moon 56.25%
Lightsday Full Moon 12.50%
Earthsday Full Moon 12.50%

Darksday New Moon 58.33%
Darksday Full Moon 27.08%
Lightsday New Moon 25.00%
Lightsday Full Moon 12.50%

At the 51+ HQ tier they found no significant difference between days and moon phases. Synths less than 31 levels lower than your cap weren't tested, as far as I know.

Question answered. With Science. Next question.

That is pretty general and lots of variable included. Not what we're looking for to be honest.

What's general about it? The variables were controlled when they could be. The level above the cap was constant. The crystal was always Wind. The direction faced was always the same. No weather. Night/day should average out, if you're synthing all day. You could probably mine it from the data if you wanted, but they didn't.

Eruntalon's (the author of the article which I cited) testing wasn't meant to answer how to get HQs 100 percent of the time, or to answer what factors affect skillups. All he was trying to do was determine the effect of Moon Phase and Day on HQ percentage. That was the OP's question on this thread, and in that respect, the experiment seems pretty valid to me. Give me some specific reasons why it wouldn't be and we'll talk. If you're just going to shrug it off as "too general," whatever that means, then don't bother replying.

Well I'd say full moon/ I also hear new moon made drops shitty. But some testing and I have noticed this during an event that drops nothing really "dynamis" or "sky farming". Bad day to farm anything on really.

If you look at the data from that post it appears that the different days affect the synth not just the moon phases. In order to rule out any inconclusive evidence you'd need to do these tests during the same moonphase on the same vana day to eliminate the possibility that days are affecting the HQ. You could generally say that the moon phases are affecting the synth, however you could also say that its the day of the week that's causing the increase.

Example:
Player 1: New moon produces best synths! Look I got 56% HQ!!! (Iceday)

Player 2: Oh really? I think New Moon sucks *** as I only got 25% (Lightsday) New Moon clearly is better as I got 31% HQ rate (Darksday)

My suggestion in order to get an accurate data is to do the wind crystal synth on Lightsday only during those specific moon phases. We need lots of synths (preferably 500+). The data presented is only 48 synths per day per moonphase which is a very small and inconclusive dataset.

On a side note, it would look like Iceday yeilds the best results for wind crystals in all 3 moon phases, however 48 synths per Iceday per phase is again a small dataset.

The dataset could be larger but you can say that for any sample if you wanted to.

As Khorbin pointed out that post is 5 years old. During that time it's been read by almost every serious crafter in the game and reposted onto every FFXI forum worth mentioning. People have done millions and millions of synths to test it. In that time no-one has managed to provide a dataset that contradicts it and many have provided others to support it.

Given the extremely argumentative nature of the people that lurk on places like BG, KI and Alla I'm sure if there were a way to bash this it would have been done. Peer-review is a very good test of any theory's validity and this one has been reviewed more than any in FFXI I'd guess.

Of course this still won't satisfy some people and if they want to ignore it they can. At the end of the day it's your game - play it how you want to. If someone asks what the majority consensus among crafters is on day/moon though this is it.

Personally I don't believe in direction at all as no-one has ever shown a link there as far as I know. Even so on big-ticket items (tier 0, multi-million gil, etc.) I still follow the compass because it costs me nothing to turn my toon a little and you just never know if that might be the difference between a 20 mil profit and a 3 mil loss.

ive ignored it but omg i make +1 sole sushi like a mad woman on new moon wind/darksday

O.oa Isn't it 384 per phase? O.oa

Thats a mistake, I meant 48 per Iceday. Editing my post.

One of the best parts about statistics is that you don't need big numbers to start to see preliminary trends. Often in scientific/medical studies the preliminary tests start out small to see if there is any merit to the hypothesis. If this were a scientific study, the conclusion would likely be that there is enough evidence for the hypothesis that "crafting during a New Moon positively impacts HQ percentage" to warrant further testing.

Sure, the data set might be relatively small, but other than that there is nothing wrong with the methodology of that experiment when taken as an initial experiment. 48 synths were done on each unique combination of Day/Moon Phase, and if you compare the data from each unique day on the different moon phases, the data points line up very nicely. For example, for each day, look at the percentage on Full Moon, then Quarter Moon, then New Moon. Just as an example (every day pretty much lines up just like this):

Lightningday New Moon 35.42%
Lightningday Quarter Moon 20.83%
Lightningday Full Moon 16.67%

Darksday New Moon 48 28 58.33%
Darksday Quarter Moon 48 15 31.25%
Darksday Full Moon 48 13 27.08%

On every day, you will see that New Moon has the highest percentage of HQs on that day, almost always followed by Quarter Moon, and on the low end, Full Moon (except on windsday and lightsday, where Quarter and Full moon produced the same results, but New Moon was always the highest percentage). Sure, you are only looking at 48 points of data per moon/day combination, but combined it adds up to over 1100 data points. It is possible, but not likely, that the data would line up that nicely if the variation was due only to random chance.

Another great thing about scientific experiments is that they're REPEATABLE! So instead of picking this experiment apart with arguments that basically amount to saying "that doesn't sound right to me," go do your own experiment. Find a way to fix any issues that you see with this study and tell us your results.

I'd rather have a small-scale study backed up by REAL data, especially where the results are pretty clearly shown than listen to all the people who are just "noticing" things. Confirmation bias is a ***. I'll take a little bit of hard data over observations that are worth nothing.

come on now...stop bashing on the data.

the fact that someone spent the time to create a data based on 48 synths per game day on the different game days and on the different moon phase. Its insane. considering the total synth needed to compile this data. i think the data is relevant enough.

of course people will say that this is a small sample size but what of it. most market research are based on small sample size also and people follow it. Like the previous posts had stated, it is the choice of the player to follow or not to follow it. just don't be bash on the data that someone spent their time to compile to show people. if you don't think the data is relevant enough then do your on test. thats all there is to it.

but give the guy that posted his research some credit.

Ever synthed multiple HQ in a row?
Ever broke multiple synths in a row?
Ever not HQd in a crafting session?

Thats why. 48 synths is not enough to be conclusive. You can see some type of data trend yes, and it is a good start. This data is also 5 years old. Who knows what SE has changed on top of that.

I'm a 100+5 cooking crafter. I went through 8 stacks of mats on Sole Sushi to not HQ a single synth. Based on my dataset I could say that its impossible to HQ sole sushi. Would I be right? Hell no. Small sample sizes show incorrect data. I was unlucky and that is it.

i may not be a 100 crafter but i do understand the feeling.

but you’re only taking a snapshot of your worst synth session which makes it’s a bias comment. i bet if you add up all the HQ and NQ synth you have done on that one item. most likely it will average out to be about + or – 5% of the given data.

I bet you are right. You know why? Large sample size. That was exactly the point I was proving.

However, my percentage of crafting over the course of my lifetime on one single item doesn't prove whether new moon, full moon has any bearing on the outcome. For that, you need control and a large sample size.

Control being facing the same direction, on the same vana day, under the same moon phase, with the same skill, no weather, etc etc, see my previous rant. Any variables that are changed will make the data tarnished and draw speculation.

I don’t disagree with you about the larger sample size giving a more accurate picture.
But come on dude are you going spend that time?
I sure as hell won’t do it.
I’m not complaining about the reasoning.
I’m just defending the dude that spent the time to put that report together.

If you’re so adamant about disproving the given data, then prove it by doing the test yourself and post it.
This is a game…no one wants to write a damn thesis report on synthing.
As everyone has stated over and over again, it’s in the eyes of the beholder.
Take the information and compile your own conclusion but don’t bash on someone else’s results.

I fully agree with you except for the bashing on their results part. I'm saying that while the results are a step in the right direction they are not conclusive proof of anything.

The problem is noone has tested the ideas I've presented with real controls and large datasets yet people seem to know know for sure without a doubt what the best days, moon phases, etc for crafting are. This is why I call BS and say its all speculation. I don't have any reason to spend the time to test it as I personally I don't believe in any of it. All I'm suggesting is people stop telling others when to craft without real proof of their superstitions.

I have to say I agree a lot with what Motokoson is saying above, however for those of you reading what he says, hes an experienced crafter...
RANT ON..
Crafting in large batches is fine, not everyone can do this, and until you have the gil to back it up, I suggest much smaller batches of different items, say 7 different items 3 times each. And for the LOVE OF GOD, please don't put up 16 like items in the AH at once.. you destroy the market!
RANT OFF
Sry =) back to the current discussion...

btw as far as EC's work on skill ups, umm there is nothing scientific that can be proven here. This game is evolving with every update, you do not have the same game you did 4 years ago, you would need to do that study all over again, and btw, you would have to factor in any possible element that would or could change the study, for example the gear your wearing, time of day, zone your in, weather in that zone... so on.. anything that could be considered to change the hypothesis, so Ec's studys doesn't prove HQ on just moon phases due to inconclusiveness... "Question answered. With (out) Science. Next question."

I think some people need to look up the meaning and purpose of control groups.

You don't have to exclude all other variables if you're comparing to a control that's experiencing the same variations. In this case we don't even know all the variables so his methods are fine. It's just not practical to carry out the whole test 8 times, once in each fixed-weather zone between noon and 1PM game-time just to rule out two factors that are covered by the control group anyway.

Gonna chime in here as far as direction is concerned:

Ive done bonecraft to 100 leathercraft to 60 and goldsmithing to 52 so far.

Ive always tried to skill up while crafting 8+ levels above my current cap.

Now when I blow through 4-5 stacks of mats not caring about profit just straight skill ups I find that direction does come into play.

When I use 2 stacks in 1 direction and not only cant make 1 damn synth without failing, and I turn into the crystal's direction and now nothing is breaking its painfully obvious that direction accounts for SOMETHING.

Be it skill or resist rate or whatever is still up for debate, but direction helps.

I usually find that standing into the direction of the day of the week if its light or dark day helps with a same day crystal (darkday for dark based crystals etc)

Aonen, your argument is equivalent to saying that all science is invalid because God (or the universe, whichever you prefer) can "change the rules," so to speak, at any time. Descartes wrote about similar topics in his philosophical work re: what we can truly "know." It's not that the argument is without merit; it's that if that were true, we couldn't "know" anything. In the end, we are left with what we can observe scientifically, both in the real world and in this game. In order to "know" anything about how the crafting system works, we must assume that they system has not changed until we find evidence to the contrary. As far as I've seen, nobody has released any data to contradict the study that I previously posted.



Thank you, Boyle. We can be friends now.



First, by way of disclaimer, I don't *dis*believe in directional crafting. However, I won't buy into it until I see some hard data on it (and trust me, I'm working on producing just that).

Second, I am by no means singling you out here. about 80% of this thread is guilty of this same logical misstep, and it is an easy one to make. Now on to the point:

The reason you can't trust peoples' anecdotal evidence, even the anecdotal evidence of MANY HUNDREDS of people, is because of a principle called "Confirmation Bias." One of the best examples of confirmation bias in the real world is the folklore surrounding the Full Moon (wow, almost relevant to the original topic!). Ask someone who works in a hospital if more crazy things happen on a Full Moon, and he or she will probably tell you that it is true. However, studies tend to show that a Full Moon has absolutely no effect on human behavior. So what happens? Well, if the hospital employee has a belief that the full moon affects human behavior negatively, when he or she sees something crazy during a full moon, the tendency will be to attribute that occurrence to the full moon. When it happens on a different moon phase, they will chalk it up as just one of those crazy things and not even think about the fact that the moon is not full.

In other words, we tend to notice evidence that supports our existing belief systems and ignore evidence that is to the contrary.

Another logical fallacy that people tend to make is called post hoc ergo propter hoc, meaning "after this, therefore because of this." In Phalon's case, it's specifically a "regression" type logical fallacy. An example of a regression logical fallacy (ripped from wikipedia, sue me--I'm getting tired of coming up with examples) is as follows:

"The frequency of accidents on a road fell after a speed camera was installed. Therefore, the speed camera has improved road safety.

Speed cameras are often installed after a road incurs an exceptionally high number of accidents, and this value usually falls (regression to mean) immediately afterwards. Many speed camera proponents attribute this fall in accidents to the speed camera, without observing the overall trend."

In the course of leveling a craft (or even several crafts) to 100, things are bound to happen as you described. You will have bad runs, followed by good runs. You notice that you are having exceptionally bad luck, so you change directions. Since statistically over time, your results are bound to regress to the mean eventually, your results get better. You attribute the improvement to your change of direction.

I know I'm being long-winded here, but these critical thinking skills, and evaluating an argument aren't just useful in a game deciding which direction to face while crafting. They're useful in LIFE! Rip-off artists, scammers, and marketers use these tactics all the time. Watch one damn infomercial for some fake "medical" product and you'll see them all over. You'll also see them all the time in alternative medicine. "OMG my back pain was bad, until I got this rock that I put under my pillow (or started acupuncture or hypnosis etc)! Now it still comes back from time to time but when it does I just shift the rock a little at night (or go back to my acupuncturist, etc.) and it gets better! Watch out for this ***, folks, and ask for hard data (and learn to read and interpret it).

There is more fail in this thread than you can shake a stick at. Take everything you've learned here, and disregard it all.

The biggest failure of most of the "scientific" studies of these things are the relatively small sample sizes, and the inability to control other variables.

Those are both extremely relevant if you want an experiment's results to be taken seriously by people who understand such things.

Anyone who has taken Statistics 101 should know that you can make the numbers show just about anything you want if you know how to work the system and manipulate them.

When someone tells me that they have a sample size of 100 synths and the deviation is something like 5%, that is not STATISTICALLY SIGNIFICANT, and that would be under the assumption that all other possible variables have been isolated except the one being tested.

*calculates the critical value of Aramina's avatar against a 2-sided t-test to disprove the H0 hypothesis that she might be lying*

Yeah I barely remember this. lol