Riven Trading & Toolbuilders: Phase 1

[korndolorous]

11 hours ago, [DE]Rebecca said:

SIMPLE EXPLANATION:
There are 8 fields to start things off in Phase 1:

itemType:  What type of Riven is it?
compatibility: Which Weapon is the Riven for?
rerolled: Has it been rolled or unrolled (true/false)?
avg: Average value of trades with this Riven Type.
stddev: The average variation in the prices that the Riven trades for.
min: Lowest amount of Platinum per trade of this Riven Type.
max: Highest amount of Platinum per trade of this Riven Type.
pop: Popularity of this Riven Type being traded within all traded Rivens.

Can we have Median, Standard Error of the Mean in the next phase? since it would be better for riven pricing

I think price of riven should base on riven score, because roll or unrolled doesn't matter to me, for example 

riven has [ + dmg to grineer , + damage to corpus , + damage to infested ] should have 0 or 1 score. (junk tier) and [ + crit chance , + crit damage , + multishot, - damage to infested ] should have 10/10 or 9.5/10 score (god tier), while both of them are unrolled the first one will give you 0 - 100 (or more depend on popularity of weapon), the second will give you 200 - 5000 or more base on popularity and weapon stats

so the price is base on stats tier and weapon popularity

• price of junk / common tier usually base on number of rolled 0 / 1-4 / 4-8 / 8+
• price of  decent / good / god tier base on stats of riven  

** there are 3 tier of stats, and 5 tiers of riven

Stats Tier 

• High tier / will give weapon more damage like(damage, multishot, critical chance, critical damage, slide crit , - zoom, - dmg to infested, etc. )
• Medium tier / will give weapon utility like ( element damage, physic damage, - recoil, etc.)
• Low tier / will make weapon go bad ( damage to faction, - damage to faction, magazine size, - damage, - multishot, etc.)

some time element / physic damage and utility stats are high tier like - recoil on grakata

Weapon Tier / ( + positive stats , - negative stats)

• S or god tier      / +3 high -1 high  ,  +2 high -1 high for some weapon ( 2000+ ) 
• A or good tier    / +2 high -1 high  ,  +3 high ( 1000+ )
• B or quite good / +2 high + 1 low or medium  ( mostly 400+ )
• C or common   /  +1 high 
• F or junk tier     /  +0 high 

Conclusion : so the data that we collect should be based on score on each riven

Note :   price are base on stats on each weapon for example + critical chance are good on crit weapon, but useless on status weapon

 

 

 

[Graavarg]

Thanks DE!

Riven trading has been the shoddy underbelly of Warframe for too long. Any body talking about a "free market economy" in regard to Warframe riven trading is either an imbecile or a crook: "A free market is a system ... in which the laws and forces of supply and demand are free from any intervention by a government, price-setting monopoly, or other authority".

Another interesting fact might be that "free markets" cannot exist without quite heavy rules & laws, including active policing and inclusion in a justice system with both economic risks (appropriation) and jail time for misuse. And it is not only national legislation either, just check the amount of NYSE rules (http://wallstreet.cch.com/nyse/rules/).

The basic commodity in the riven market is of course rivens, and their supply is decidedly not free, it is heavily regulated by the authority in question (DE), and with good reason. What have been needed for a long time is that the regulation would extend to using the system for immense personal profit, which affects the game community and players negatively. As far I as know there has never been any intention of creating "plat millionaires", or allowing them to influence the game & the community.

The limit of 90 rivens does nothing to stop the riven mob, since they use a multitude of accounts. Multiple accounts are also needed to get around the "trade-with-yourself" ban. And of course such "insider trading " can be used to influence prices as well. The riven mob system goes much further than just fooling "new players by buying their rivens "cheap" and selling them for "much plat" and investing immense amounts of plat to corner the market for "hot" rivens (Stradavar and Tipedo, looking at you...) in order to achieve a price increase. There has been persistent rumours about "kuva farming for pay", that is whales & the riven mob paying players in plat to farm kuva for hours on accounts they then use to roll rivens, before internally trading them to other accounts for sale/resale.

I find it hard to believe that DE has ever had the intention of creating and/or supporting systems like this, but like many "authorities" they now have the task of regulating the current misuse.

There is also a legal and moral aspect, as the in-game currency has a clear link to real-world money and a certain amount of the player base is underage (according to the laws in respective countries). As far as I know most developed countries do not allow underage citizens to engage in "free market" trading (or trading with real money at all). While buying a small amount of plat for your weekly allowance to get some stuff or boosters in a game you love to play is completely ok, allowing underage players to be potentially abused by misuse of an internal trading system is definitely not ok. In some countries it might even be illegal (as anyone knows who have followed such developments during the past few years). Comments like "if anyone is stupid enough to pay thousands of plat for a riven, well there is a price to pay for stupidity" doesn't really work anymore when the one being "stupid" is only 13 years old and has just emptied the savings account in order to buy a mob-inflated riven, believing that it is "godly" and worth it...

I am not trying to pile on DE here, on the contrary I am trying to defend and applaud DE for tackling this problem. Informing players with pricing data is only a first step (as stated), there is a clear need for more to be done, including more regulation and tougher sanctions on the riven mob for turning a basically good system into something negative. 

Edited March 30, 2019 by Graavarg

[Sharpears]

Bravo! As a newer player, it's lovely to see >ahem< Riven trading moving out of the stone age... To quote a certain discussion.

Maybe I'll actually start trading them now, rather than considering them more for their (small) endo value.

More and more open value data is only a good thing for the vast majority of players, I'm sure.

[kelllen]

2 hours ago, ReaverKane said:

It should update weekly according to Reb's post, i'd love it to be daily at least, that'd be awesome.
Since the update interval is one week, you don't really need a date or date range. Just use the HTTP HEAD request to get the date field (which should be the date it was updated), should be enough for synchronicity purposes. You can also use it to display a date range for the trades, just convert it to a date data-type for whatever language you're using, and do some maths on it.

You are correct on all points, however it's a cheap way to provide some additional reliability to consumers of the data.Otherwise I need to checksum and maybe bake some assumptions into the fetching/update code. I can do all that...just more convenient if they could provide it.

If this data is more around market trend than determining a "best price", then most of it's value comes from week-over-week aggregation. Guess I just don't trust myself enough to assume I can pull it reliably without any problems ever.

[JCD89]

Nice, finally a step in a right direction.

[(PSN)RenovaKunumaru]

Sheesh, the amount of hate is massive. You can leave feedback without calling someone's work "useless".

I'm quite neutral to all of this. As long as it doesn't tank the PS4 market which is significantly lower than PC's, I'm cool with it.

Thanks DE

[ReaverKane]

27 minutes ago, kelllen said:

You are correct on all points, however it's a cheap way to provide some additional reliability to consumers of the data.Otherwise I need to checksum and maybe bake some assumptions into the fetching/update code. I can do all that...just more convenient if they could provide it.

If this data is more around market trend than determining a "best price", then most of it's value comes from week-over-week aggregation. Guess I just don't trust myself enough to assume I can pull it reliably without any problems ever.

Well, depending on how you're parsing it, it might be "cheaper" in terms of network to check the header code for the date.
I use this allot on mobile apps, since it's less data intensive than parsing even just the first line of JSON.
For this, though, since the file is small, i'd just parse from the API instead of storing and regularly updating.

Edited March 30, 2019 by ReaverKane
More info

[BrenBlaze]

Awesome. Are there any plans to add a filter for stats in the future? It's probably over the top, but just curious.

Edit: I guess what the practical form of what I am asking is the data from all rivens individually. Name, MR, Rolls, All Stats and Price it was sold for. For the purpose of seeing what stats are selling for what on each riven after some filtering.

Edited March 31, 2019 by BrenBlaze

[BrenBlaze]

Curious as to why people are comparing this to a stock market and other real markets. Not an expert, but any comparison of the riven market to the stock market is kind of nonsense. The purpose of buying them is different, one is entirely an investment with no physical product, while the other has a usable product and is an investment. Stocks are a portion of a companies future value, a riven is not a part of anything but builds. Two stocks for the same company have the same price and change together, and are heavily based off of predictions. None of that is true for rivens except when new primes are announced, so very niche. I could keep going if anyone needs, but instead of comparing the riven market to the stock market, literally a chuck e cheese is a closer comparison. 

[Jangkrik]

Thanks for the data! Is it also possible to show the quartiles (or deciles if possible), median, and mode?

As it stands, it can be seen that for some items, the variabilities are very high (e.g. Mios and Sigma & Octantis), this usually implies that the price ranges are very big.

For example, Mios' trash unroll price is 60-150, but a good unroll might sell for 400. This is also reflected in the mean (200) and standard deviation (700). Hence, to read the data for such rivens, the mean and SD alone are not enough to draw a conclusion on the price.

 

Also, would it be possible to also release the stats of traded rivens along with their sell prices? I'd love to do some predictive analysis of the data sets.

Edited March 30, 2019 by Jangkrik

[Guest bloodtron40000]

Thanks for this preliminary data!

In order to avoid giving some false impressions to players please also provide this information:

- Note in provided api or documentation whether the data we have is lifetime or limited to a specific date range.

- Is it lifetime? Then please also provide an additional data set with a limited date range, weekly or monthly, so we can more closely observe shorter-span market developments with, say, the announcements of new weapon variants, buffs, or nerfs.

- 2nding median, mode, and std error of mean being added to the data.

- Show trade history over limited data range, with included specific riven stats. (EX: This week was traded Cronus +cc/cd for 50p, +dmg/+heat for 100p...) This will also help us recognize data outliers such as: false trades to transfer plat, joke sales, perfect rivens, data manipulation, so on. Maybe there is a genuine reason why someone traded a Fang riven for 4.9k. Maybe there isn't. It'd be nice to be able to estimate why. 

For reference on the value of having specific trade history data, here is Semlar's own data presentation with point notation that would be great to be able to replicate in future community projects:

Spoiler

In light of the new disposition adjustment schedule, having a reliable source for confident data evaluation is more important than ever for both those looking to purchase and/or sell. Very much looking forward to future developments.

Edited March 30, 2019 by bloodtron40000

[000l000]

Il y a 17 heures, (PS4)xtharbadx a dit :

One question thou, what stops a riven mafia gang from repeatably trading amongst themselves to influence prices? Most simple scenario, 2 guys keep trading a riven between themselves for 1k plat hundreds of times, wouldn't that influence the metrics? Even harder to track if it's a larger and more organised group.

I don't think they care much cause they aren't necessarily exploiting players lack of knowledge, they're basically monopolizing the whole market. Why would they care about any price since they choose the prices themselves once they've bought every single popular riven ?

We don't need analysis, one just needs to trade a day or two to understand what's going on there - We need regulation. As long as only one guy will sell half of the rivens on only one riven market page, we'll experiment higher and higher inflation. They don't even necessarily sell all of this expensive rivens since buyers aren't a lot but since the prices are going crazy it prevents most players from having access to them. I don't have any issue with selfish players becoming rich, but impairing other players experience is another story DE should take care of.

And @DE, no, rivens aren't optional. They're mandatory as any other mod in the whole game. Mods aren't optional...

[Pikaapooo]

Thanks for the JSON! That’s the best format to use.

Please include a few more fields:

Count: the number of trades happened during that week for that specific type

Count of Unique Users: so we can know if some person is trying to skew the data.

Median: the median trade value. This will give an indication of the skew, so say if I just want to trade a melee riven for 100k plat,  your average will be totally destroyed.

5th percentile: Filtering out outliers on the lower end.

95th percentile: Filtering out outliers on the upper end.

And, if possible, include the following as well:

25th percentile, and 75th percentile. Those are going to be used for doing the vanilla outlier calculation.

Final request would be, can those data be at a finer granularity? Like by a day? Or if not, can we have weekly data from the past? Not too much to be done for just 1 week of data, but I can see if I have like half a year of weekly data, then there are better ways for analyzing something. 

Edited March 30, 2019 by Pikaapooo

[Kontrollo]

I've posted a script to save those files to disk in this thread before:

Spoiler

 

But I've figured it's easier for mere mortals if these were saved in CSV format (comma separated values), which you could then open with e.g. Excel or Calc. So here's a slightly extended version of that script, again in Python:

 

from urllib.request import urlopen
import json
import csv

riven_data_dl = [
    'http://n9e5v4d8.ssl.hwcdn.net/repos/weeklyRivensPC.json',
    'http://n9e5v4d8.ssl.hwcdn.net/repos/weeklyRivensPS4.json',
    'http://n9e5v4d8.ssl.hwcdn.net/repos/weeklyRivensXB1.json',
    'http://n9e5v4d8.ssl.hwcdn.net/repos/weeklyRivensSWI.json',
]

for url in riven_data_dl:
    file_name = url.split('/')[-1]  # use what comes after the last slash as file name
    file_name = file_name.replace('.json', '.csv')  # proper CSV file suffix
    with urlopen(url) as riven_data, open(file_name, 'wt', encoding='utf-8', newline='') as riven_file:
        riven_data = json.load(riven_data)
        fieldnames = 'itemType, compatibility, rerolled, avg, stddev, min, max, pop'.split(', ')
        csv_writer = csv.DictWriter(riven_file, fieldnames, dialect='excel')
        csv_writer.writeheader()
        csv_writer.writerows(riven_data)

 

_{(Note: I'm putting this into the Public Domain. If not applicable where you live, license: MIT.)}

 

Edit: Screenshot of what the open dialog looks like with OpenOffice Calc when using one of these files. (Make sure to uncheck 'Merge delimiters' if it's active here.)

Spoiler

Edited March 31, 2019 by Kontrollo
Note: if you have trouble running the script, then it's probably DE's forum software's fault. sometimes some invisble chars get added when loading, selecting or copying? in any case, Python will fail. can't do much about that. maybe try another browser :(

[(PSN)cleefsentence]

Good start but the folren's riven bot on discord is better because he tells you the value based on the stats you put.

This is the link for those who may want to see it https://discordapp.com/invite/rnzSW6k

[KettleTicket]

How will your system prevent possible "stat padding" of a possible group of people trading the same riven back and forth at an exorbitant amount of plat to boost the average? 

[fellow1]

Thank you (and for drop rates data too).
Would it be possible to make more occasional updates (instead of being weekly)? Like daily?

Edited March 31, 2019 by fellow1
question

[---.CCC.---TradeFrame]

Can we also see how many times that riven was traded?
+

How do you handle trade data with more than 2 rivens in the trade?

[Darkvramp]

why not just make a better trading system and avoid another problem like this again?

[SmallRat]

Dear Rebecca,

I saw the word 'average variation' is used in item 'stddev'

May I want to know which calculation method is used, average deviation or standard deviation?

Thank you.

[iNew_Vexx]

I see many complaints and zero solid arguments, kkkkkkk, good work DE.

 

Buuuut, i dont know if this is the solution. Making kuva more acessible with bosses rewards on spiders and eidolons, or maybe just a new event content like thermia fractures, ghouls, etc, maybeeee, IMO, can let new players come more often to "endgame" content and make it more easy to reroll your riven. Farming kuva today is boring as hell, maybe chaning that is to important as trade chating rivens and toolmakes.

Adding some new mechannics on riven like "save this status i like for more kuva on the next reroll" can make more nice rivens, and getting prices low.

idk, just my opinion, i play this game about 4 years now, and i see a lot of people complain about rivens rolls. The rng factor is to bizarre for some players to invest their time on kuva farm. Change that pls, make kuva more fun or add some events with it.

CREATE A LOCK SYSTEM ON RIVENS!

 

Edited March 31, 2019 by Khaeji

[Melanholic7]

reep riven mafia?=(

[SordidDreams]

This is an excellent start, however its practical utility is basically nil until the stats of each riven are included in the data, thereby allowing them to be filtered by stat combinations.

[(XBOX)Kuva Queen]

On 2019-03-29 at 9:58 PM, Soulless_Spartan said:

I like this quite a bit. Although I have one comment (not specifically for DE, but just a general comment.)

To the person who coughed up 16,500 plat for an Artax riven... damn. Just... damn. You do realize you just spent $1,100 USD for that, right? (Assuming 75 plat ~= $5 USD)

It's because you cannot get those rivens easily. They were out for less than a week then taken out of the game so the chances of finding one are so so low. 

[THE_ZEEK]

SO this is great news.. How will you account for the following :

Manipulating the market with this system is more easy than ever..

When wanting to manipulate the market, the only thing you have to do is set up a trade with your friend and trade one riven between yourselfs for rediculous prices.. this will if you trade 27 times *assuming MR27*  between a clan or even a few people, balloon the riven price of that particular riven

 

Edited March 31, 2019 by THE_ZEEK