I Hate the Market: a Random Roll Log
By
Aliana Blue,
Posted on: 2011-03-10 24:39 in The Market and Inventions
I'm just back from a vacation with no internet [surprisingly interesting, real life can be] so I don't know, but for a long time, Kin Combat had two recipes worth [crafted] around 100M that went for one AM each. Lots of stuff crashed, that may have as well.
It's all about what you like, what you'll suffer through, and what you are willing to pay to avoid.
Mini-guides: Force Field Defenders, Blasters, Market Self-Defense, Frankenslotting.
So you think you're a hero, huh.
@Boltcutter in game.
Quote:
For a week and a half or so, the crafted were going for 100-150 million. Easiest billion I ever made, since I was running tips anyway.
|
Originally Posted by Daemodand
So I checked it out and the 20 RR ToHit/Recharge recipes (the best of the 3 pieces, looks like) seem to be going for about 40 million, crafted for 60 million.
|
Having Vengeance and Fallout slotted for recharge means never having to say you're sorry.
Ah, well, looks like I missed the party on the Rectified Reticles. I'll console myself with...
TODAY'S RANDOM ROLL!
ROLL THE FIFTH:
Blessing of the Zephyr: Knockback Reduction (4 points) (50 Recipe). 9,082,005
Sovereign Right: Acc/Dam/End (50 Recipe). 22,505
Performance Shifter*: Rech/Acc (50 Recipe). V9,995
Mako's Bite: Dam/Rech (50 Recipe). 15,000
Mako's Bite: Acc/End/Rech (50 Recipe). 90,005
THIS ROLL: 9,219,510
5 ROLL TOTAL: 249,463,590
5 ROLL AVERAGE: 49,892,718
COMMENTARY: A marginal recipe, 3 pieces of junk including a vendorable, and a so-so Mako's. Another bad roll. I think that makes 3 below-target rolls at this point and my average is now 17 million behind target. This is bound to pick up over the next few rolls. At least I hope so.
QUESTION: How many random rolls does it take to reach statisitcal significance?
Quote:
Welcome back!
|
Originally Posted by Fulmens
I'm just back from a vacation with no internet [surprisingly interesting, real life can be]
|
The best comics are still 10�!
My City of Heroes Blog Freedom Feature Article: "Going Rageless?"
If you only read one guide this year, make it this one.
Super Reflexes: the Golden Fox of power sets!
WARNING: I bold names.
Quote:
Don't have the math to know, but my gut says in the triple digits. Of course, that many recipes hitting the market over a short time would likely have an impact on prices, further compromising significance. |
Originally Posted by Daemodand
QUESTION: How many random rolls does it take to reach statisitcal significance?
|
Quote:
It depends on what you are trying to prove or disprove. |
Originally Posted by Daemodand
QUESTION: How many random rolls does it take to reach statisitcal significance? |
If it is "My random roll will net me less than 65mm inf", probably a lot, especially since conditions change over time. I'd guess several thousand off hand, since we don't exactly know what the drop rates are for random rolls, there are so many recipes that can drop, and because prices on recipes will ebb and flow based on a lot of reasons. I just know enough statistics to make me dangerous to people who actually know statistics, so hopefully a real expert will posit on that!
If it is "The RNG hates me!", then just one. I, myself, have proved that one many, many times!
Ooh, a sarcasm detector. Oh, that's a *real* useful invention. http://www.telegraph.co.uk/technolog...t-sarcasm.html
Quote:
My premise is random rolls will earn an average of around 67 million inf each in the long run. Right now, I'm well below that but I guess I have at least another 1,995 rolls to make to be able to tell for sure.|
Originally Posted by Yomo_Kimyata
"My random roll will net me less than 65mm inf", probably a lot,
|
The best comics are still 10�!
My City of Heroes Blog Freedom Feature Article: "Going Rageless?"
If you only read one guide this year, make it this one.
Super Reflexes: the Golden Fox of power sets!
WARNING: I bold names.
Quote:
I'll help in a few days once I get a rogue back to hero so I can convert merits! But I won't sell at one -- I'll normally craft and take my chances. I'll post the lowest of the last five sales though unless I vendor. I think this is good work on your part, by the way!|
Originally Posted by Daemodand
My premise is random rolls will earn an average of around 67 million inf each in the long run. Right now, I'm well below that but I guess I have at least another 1,995 rolls to make to be able to tell for sure.
|
Generally in stats, it's a lot easier to support or most likely reject a premise by saying it is greater or less than something, rather than it is equal to something. I didn't remember 67mm versus 65mm, so my bad on that!
Ooh, a sarcasm detector. Oh, that's a *real* useful invention. http://www.telegraph.co.uk/technolog...t-sarcasm.html
And now, today's random roll!
ROLL THE SIXTH:
Positron's Blast: Dam/Acc/End (Recipe). 4,500,006
Edict of the Master: Damage (Recipe). PENDING
Touch of Death: Acc/Dam/End (Recipe). 180,005
Decimation: Acc/Dam/Rech (Recipe). 900,068
Obliteration: Chance for smashing damage (Recipe). 14,220,005
THIS ROLL: 19,800,084 PENDING
6 ROLL TOTAL: 269,263,674 PENDING
6 ROLL AVERAGE: 44,877,279 PENDING
COMMENTARY: A lot of mediocre recipes making for a pretty poor roll. I'm now 22 million out from my target. This method is looking pretty bad right now, but we'll see.
The best comics are still 10�!
My City of Heroes Blog Freedom Feature Article: "Going Rageless?"
If you only read one guide this year, make it this one.
Super Reflexes: the Golden Fox of power sets!
WARNING: I bold names.
Quote:
|
Originally Posted by Daemodand
<Edit snip of the prices--they were too big> |
Your sell prices are seriously making me cry. If you had crafted these, you could have made several times what you made off of it.
The crafted makos/lockdown's alone can sell 5-8 million, Touch of Deaths and Posi blasts go for 8-15 million, and I'm pretty sure that GotA: Run speed goes for more than that (so I assume you're just lowballing everything for minimal marketeering time?).
If you don't AE for easy rare salvage accumulation, I would still say it's worth the investment to craft these suckers to get the full return on your A-Merit purchases.
As he has said for half his posts on this thread, that isn't the point.
Quote:
He is simply sharing his experiment that he is conducting to prove or disprove his hypothesis.
| My hypothesis is that while random-roll-craft-sell is the optimal approach, random-roll-sell is "good enough" as an alternative, especially given the lower overhead and lesser time requirements, not to mention avoiding an in-game activity I don't enjoy, and minimizing contact with the market to boot. |
Quote:
Ah. I see. Well then, never mind me. I'll just go sit in that corner over there. |
Originally Posted by Dr_Occult
As he has said for half his posts on this thread, that isn't the point.
He is simply sharing his experiment that he is conducting to prove or disprove his hypothesis. |
My lowball statement stands, though. It causes me pain to see a posi proc sell for that little, knowing I've paid millions for the recipe alone.
Start placing lowball bids to scoop them up for yourself
Quote:
In the first post of this thread I laid my rules out. I sell for 1 inf, and vendor junk. Avoiding crafting is part of the experiment, but your advice that the random-roll-craft-sell method would yield superior results is most likely accurate, though I don't think anyone has actually proven that statistically. At any rate your advice is good advice, it's just outside the parameters of this experiment.|
Originally Posted by Stone Daemon
(so I assume you're just lowballing everything for minimal marketeering time?).
|
Quote:
|
Originally Posted by Stone Daemon
Ah. I see. Well then, never mind me. I'll just go sit in that corner over there.
|
Quote:
|
Originally Posted by Dr_Occult
Start placing lowball bids to scoop them up for yourself
|
The best comics are still 10�!
My City of Heroes Blog Freedom Feature Article: "Going Rageless?"
If you only read one guide this year, make it this one.
Super Reflexes: the Golden Fox of power sets!
WARNING: I bold names.
Quote:
http://boards.cityofheroes.com/showt...ro+merit+rolls|
Originally Posted by Daemodand
In the first post of this thread I laid my rules out. I sell for 1 inf, and vendor junk. Avoiding crafting is part of the experiment, but your advice that the random-roll-craft-sell method would yield superior results is most likely accurate, though I don't think anyone has actually proven that statistically. At any rate your advice is good advice, it's just outside the parameters of this experiment.
|
Organica's 30 herit merit rolls in one month, crafted(mostly) yielded about 2.3 billion.
That is equivalent to 76.6 million per roll. Considering Organica was converting reward merits into hero merits, you could add another 600 million onto that total to reflect the 30 conversion fees of 20 million.
2.9 billion or so from 30 rolls is equivalent to 96.6 million inf per roll, over 30 days. Some of this is of course could change based on who was doing the posting, and their experience in listing items, etc. But it is a good set of data to analyze how profitable crafting recipes can be.
Quote:
That's a great thread, but statistically speaking, 30 data points most likely isn't significant enough to draw any conclusions from. Okay, you and I know her method is probably rock-solid, and one would be wise to follow her method sans conversion fees, but I still maintain there is no statistical evidence for or against it. I'm being very nitpicky there, but that is what I said. |
Originally Posted by Captain_Karate
2.9 billion or so from 30 rolls is equivalent to 96.6 million inf per roll, over 30 days. Some of this is of course could change based on who was doing the posting, and their experience in listing items, etc. But it is a good set of data to analyze how profitable crafting recipes can be.
|
The best comics are still 10�!
My City of Heroes Blog Freedom Feature Article: "Going Rageless?"
If you only read one guide this year, make it this one.
Super Reflexes: the Golden Fox of power sets!
WARNING: I bold names.
And if one were to demonstrate it nonstatistically, by (e.g.) making 50 billion inf buying Pool C recipes and selling Pool C enhancements, that would prove nothing. Right?
Move along, nothing to see here.
Mini-guides: Force Field Defenders, Blasters, Market Self-Defense, Frankenslotting.
So you think you're a hero, huh.
@Boltcutter in game.
Quote:
If you don't enjoy crafting, then that's certainly relevant. Heck, on the occasions when I farm AE Bronze rolls, I don't always craft them. Sometimes I'll knowingly sacrifice 200 or more million because I just can't be bothered to craft. As someone on this board recently pointed out, it's less annoying to craft in stacks of ten (as in marketeering) than it is to craft ten different recipes.|
Originally Posted by Daemodand
That's a great thread, but statistically speaking, 30 data points most likely isn't significant enough to draw any conclusions from. Okay, you and I know her method is probably rock-solid, and one would be wise to follow her method sans conversion fees, but I still maintain there is no statistical evidence for or against it. I'm being very nitpicky there, but that is what I said.
|
For that reason, I for one am interested to see how your little experiment pans out, but there really isn't any need for an exhaustive statistical analysis of crafted versus uncrafted listings: where the crafted IO is better, it's better, and since people who craft their listings generally check the recipe's price beforehand, there isn't any case to be made that vast swaths of the playerbase are losing money by crafting IOs for sale. You're basically asking whether x + y is greater than x, where y can only be a positive number.
You can't divorce the player's common-sense discretion from the analysis, nor can you quantify it. Is crafting everything always going to be better than crafting nothing? Probably not. Is crafting everything that's worth crafting always going to be better than crafting nothing? By definition, the answer is yes.
Quote:
|
Originally Posted by Iggy_Kamakaze
Nice build
 |
Quote:
Yeah, and the list of stuff that's worth crafting is pretty long. In my experience, pretty much everything that's not vendor garbage (hello, Ruin!) or super-high-end (damage set purples) consistently sells for a heavy profit crafted, even after you factor in salvage costs. There are occasional exceptions where the crafted and the recipe will hover around the same price, or the salvage shoots up and makes the profit margin on the crafted really thin until prices adjust, but it's pretty widespread.
|
Originally Posted by Obitus
You can't divorce the player's common-sense discretion from the analysis, nor can you quantify it. Is crafting everything always going to be better than crafting nothing? Probably not. Is crafting everything that's worth crafting always going to be better than crafting nothing? By definition, the answer is yes.
|
Having Vengeance and Fallout slotted for recharge means never having to say you're sorry.
Quote:
Statistically speaking, yes. That is not proof. It's an anecdotal indicator, and most likely points in the direction of reality, but the number of data points for statistical relevance is quite high. Collect 1,000-2,000 rolls, and we'll talk about proof.|
Originally Posted by Fulmens
And if one were to demonstrate it nonstatistically, by (e.g.) making 50 billion inf buying Pool C recipes and selling Pool C enhancements, that would prove nothing. Right?
|
Quote:
|
Originally Posted by Obitus
For that reason, I for one am interested to see how your little experiment pans out, but there really isn't any need for an exhaustive statistical analysis of crafted versus uncrafted listings: where the crafted IO is better, it's better, and since people who craft their listings generally check the recipe's price beforehand, there isn't any case to be made that vast swaths of the playerbase are losing money by crafting IOs for sale.
|
The best comics are still 10�!
My City of Heroes Blog Freedom Feature Article: "Going Rageless?"
If you only read one guide this year, make it this one.
Super Reflexes: the Golden Fox of power sets!
WARNING: I bold names.
And now, today's random roll!
ROLL THE SEVENTH:
Obliteration: Acc/Dam/Rech/End (Recipe). 56,099,104
Numina's Convalescence: +Recovery and +Regeneration (Recipe). 55,800,005
Touch of Death: Dam/End/Rech (Recipe). 1,350,005
Lockdown: Chance for +2 Mag Hold (Recipe). 95
Devastation: Acc/Dam/Rech (Recipe). 2,070,005
THIS ROLL: 115,319,214
7 ROLL TOTAL: 384,582,888
7 ROLL AVERAGE: 54,940,413
COMMENTARY: Best roll of the seven! Got a good upward correction and am now 12 million out from my target. Hoping this trend continues. I knew I should've vendored the Lockdown, but there were bids out and the last 5 were above 10k, so I gave it a shot...and lost. 1 inf comes back to bite me once again.
I had 7 Carbon Rods listed for 500 that sold overnight for 1.5 million. The salvage market is crazy these days.
The best comics are still 10�!
My City of Heroes Blog Freedom Feature Article: "Going Rageless?"
If you only read one guide this year, make it this one.
Super Reflexes: the Golden Fox of power sets!
WARNING: I bold names.
Quote:
Not to derail, but - source, please? There's an entire set of methods within statistics dealing with power analysis (ref. Cohen 1992 "A Power Primer"), which can be used to make estimates of the number of observations required to ensure that one will find an effect if one is present. Unless you're talking about small effects (and my "anecdotal indicators" from observing the market say that we're not talking about effects that would be quantified as particularly small), combined with extremely high levels of power, I can't think of a statistic that would require 1000-2000 observations to provide a fairly high degree of confidence that any true effect would actually be identified.|
Originally Posted by Daemodand
Statistically speaking, yes. That is not proof. It's an anecdotal indicator, and most likely points in the direction of reality, but the number of data points for statistical relevance is quite high. Collect 1,000-2,000 rolls, and we'll talk about proof.
|
That being said, I don't dabble in this stuff like I used to, and am honestly curious where these large numbers are coming from and what statistics you're basing them on.
My postings to this forum are not to be used as data in any research study without my express written consent.
Quote:
I'm going on the answer I got to the question earlier in the thread. I asked about how many rolls would be needed for a valid statistical analysis of random rolls, and it apparently would need a couple of thousand at minimum.|
Originally Posted by EarthWyrm
That being said, I don't dabble in this stuff like I used to, and am honestly curious where these large numbers are coming from and what statistics you're basing them on.
|
Now, if you can prove mathematically that that is unnecessary, I'm all ears and always willing to learn. In fact, I've been hoping for something like this, because threads like this become meaningless unless those thousands of rolls are made. I'd love to hear it takes only a couple dozen, or even just a hundred or so.
But given how large the recipe pool is, I doubt such a small number of rolls could be statistically considered proof, but I'm certainly open to and even hope that that is the case. It means a lot less work for me.
So far, those who say they have some expertise in statistics are telling me it takes a large number of rolls. You're telling me it possibly doesn't, so ok, I'm listening.
P.S. He didn't say, but in Fulmens' above-mentioned 50 billion, I'm assuming that's all 35-39 Alignment Merit random rolls that were crafted and sold, so I'm guessing based on that number he's made around 500 rolls. A decent sample, but is that enough to be statistically meaningful? And even if it is, it says nothing about the result of just selling the recipes. No one here is doubting crafting will get you more money. The question is, will just selling the recipes get you enough whatever you consider enough to be (for me it's 68ish million per roll).
The best comics are still 10�!
My City of Heroes Blog Freedom Feature Article: "Going Rageless?"
If you only read one guide this year, make it this one.
Super Reflexes: the Golden Fox of power sets!
WARNING: I bold names.
Quote:
|
Originally Posted by Stone Daemon
I'm pretty sure that GotA: Run speed goes for more than that (so I assume you're just lowballing everything for minimal marketeering time?).
|
Crafted GotA: Run Speeds were going for 20 mil for a long time, but the bottom dropped out of that niche a couple of weeks ago and they now go for about half that. Oh well. Was good while it lasted.
And now today's random roll!
ROLL THE EIGHTH
Malaise's Illusions: Chance for Psionic Damage (Recipe). v9,995
Mako's Bite: Chance for lethal damage (Recipe). 9,005
Perfect Zinger: Rech/Acc (Recipe). v9,995
Decimation: Acc/End/Rech (Recipe). 900,005
Trap of the Hunter: Acc/Immob/Rech (Recipe).v9,995
THIS ROLL: 938,995
EIGHT ROLL TOTAL: 385,521,883
EIGHT ROLL AVERAGE: 48,190,235
COMMENTARY: My best roll is followed by my worst. 3 vendorables and a fourth that should have been vendored. One half-way decent recipe. Random is a harsh mistress! The good news is the next roll practically has to be higher than this. I'll tell you though, so far this method is looking like it's not panning out. One wouldn't have too difficult a time using the direct-buy-craft-sell method to beat 48 million per AM, heck, one could probably direct-buy-sell and beat that. Oh, well! Random is random!
The best comics are still 10�!
My City of Heroes Blog Freedom Feature Article: "Going Rageless?"
If you only read one guide this year, make it this one.
Super Reflexes: the Golden Fox of power sets!
WARNING: I bold names.
I'm not too impressed. I'm getting about 60 million right now with zero crafting, though I can see the appeal of going after this, I personally wouldn't.
The best comics are still 10�!
My City of Heroes Blog Freedom Feature Article: "Going Rageless?"
If you only read one guide this year, make it this one.
Super Reflexes: the Golden Fox of power sets!
WARNING: I bold names.