More Leniency and Incentive to Reprice Currencies

[Uncle Dang]

(I am mainly referring to the stated price in refined metal for keys with the title, but other tools, cases, and promos would also apply) 

 

It's gotten ridiculous that the sticker prices for basic trading staples like keys are routinely out of range by 10% or more, like keys are at time of posting, and often stay that way for weeks or months at a time. There is a labyrinth of other items that are awkwardly dependent on prices of these commodities being accurate, like if an item is just above or just below being a key, and can be abused (mostly in-server, I'll admit) if people are blindly following a wildly inaccurate rate for keys, or even if bots are misconfigured around an old price. At the absolute least, this leads to stupid situations where a buy order for 44.22 ref is still weighted higher in the buy order list than offering 1 key, despite keys currently being about 50 ref at time of post.

 

This problem mostly I think is mostly chalked up to the much more stringent expectations on when repricing keys is even appropriate, i.e. that you have to be quoting hundreds of sales and the price "has to be stable" (even though it's clear keys almost never are these days). It's still a good thing that the barrier to entry is high to keep out joke or manipulative suggestions on something so important, but from what I've observed, two things should change:

 

1) Getting rid of the inherent expectation of a 1-2 scrap price difference, this used to be reasonable up until keys were like 10 ref, but at 50 ref the difference is bound to be larger (sometimes it's even been around 2 ref spread for days at a time). Ref-key pairing is not stable, probably won't ever be as stable as this unspoken rule would demand.

 

2) More of an incentive to reprice commodities like keys, since you generally have to collect many more sales that are harder to interpret due to the massive amounts of metal flung around alongside other items. As it stands, it's a form of enhanced interrogation to put together a suggestion for keys that stands any chance of passing, as I can personally attest to. Even when it's a valid suggestion, another price spike in keys can just throw the hard work out the window if it's not accepted within a few days. Compare the effort required with suggesting to reprice a 1 of 3 Unusual, where there often is only a handful of sales that need to be quoted and which is rarely going to invalidated from a new sale popping up, and you'll see why the key price is left out to rot.

 

What the form of said incentive should be I'm leaving up in the air. Counting as a few successful suggestions if you get a currency suggestion passed? Exclusive badge? An event? Whatever ideas you have, or even if you think it's not all that important of a problem, feel free to leave your input. I'm still interested in seeing what public opinion is regarding things.

[Speedwagon]

I completely agree. The mods/admins generally handle price suggestions with a "wait and see" kind of approach and it can get very annoying considering keys and ref will always be volatile. Although I've never priced ref/keys before, every time I see a suggestion I kinda feel bad for the person that suggested it because I know they'll be put through hell just for trying to update the price. This also affects the price of ref because the value of refined is measured by the cash sales of keys divided by whatever the current (and now outdated) price of ref->key is, which just seems off to me, but the last time people asked admins about it their reasoning was that it would be too much work for the site's devs to reprogram everything. I remember when marketplace had ref going for .05 a while back and a suggestion was reposted to include the change, then that suggestion was left up for a week before an admin looked at it and said that we can't take marketplace ref sales into account, and it had to be reposted a third time just at .04.  I think both your ideas are good, but I also think that if they couldn't change the system they should at least check ref/key suggestions more often to follow the fluctuations in price, or even make a suggestion themselves if they can in situations like now where nobody has made a real one in almost a month (I'd bet because nobody wants to go through the effort just to get it ignored and/or closed because the price moved again).

[LaughingLollipop]

sloth in pricing is endemic to the existence of this site tbh

 

the suggestion model just doesn't work well for items that are dynamic and commonly traded (see rant about currencies) nor does it work well for items that are rarely traded (see the sheer number of 'reasons' that one and only sales are rejected as evidence, the fact that high tier suggs are likely to sit at least 2 months now and the fact that a vast majority of high tiers are >1 year outdate)

 

the sweet spot is somewhere in shit tier unusuals, price suggestions have a half dozen sales to pull from sit around a month and then get accepted

 

I don't have a solution, just a half decade of observations

 

 

[Teeny Tiny Cat]

I think the main issue is them not being handled quickly enough, which isn't down to mods taking an intentional approach as much as it is that most of them don't feel comfortable handling currency. I am frequently busy, but I will try to keep an eye on currency and handle those suggestions more quickly. 

[Pigeon]

Price's change without suggestions, but that is obvious. With keys you literally look at the buy/sell orders and if you have no brain you can figure out what the price they are trading at. This is because they are a frequently traded item. Ref is only needed to value everyone's backpacks. If this is in accurate it does not matter. When you cash out you will go for keys instead of ref anyway. 

 

They could just automate the price of ref by using average marketplace.tf sales per day. 

[Zeus_Junior]

2 hours ago, Pigeon said:

Price's change without suggestions, but that is obvious. With keys you literally look at the buy/sell orders and if you have no brain you can figure out what the price they are trading at. This is because they are a frequently traded item. Ref is only needed to value everyone's backpacks. If this is in accurate it does not matter. When you cash out you will go for keys instead of ref anyway. 

 

They could just automate the price of ref by using average marketplace.tf sales per day. 

The idea of automatic price changes is something we should stay far away from.

That would only create idea that backpack.tf is the all knowing price bible. Instead of it being suggested prices.

[FP jh34ghu43gu]

10 hours ago, Zeus_Junior said:

The idea of automatic price changes is something we should stay far away from.

That would only create idea that backpack.tf is the all knowing price bible. Instead of it being suggested prices.

Seeing how keys and ref are displayed on the front page and those values are expected to be the current trading prices, they might need to be an exception if they (bp) cannot accept suggestions fast enough (probably not ref imo but keys ya). Right now keys are underpriced by 12% which is ridiculous. My understanding of backpack is it is a reporting service, it reports what items sell at, what items exist, etc... and keys constantly sell for different prices than what is reported because these reports seem to average the prices out for the previous few weeks; if they just reported the average buy price for the top 25 buy orders at the start of each day then accuracy for what keys are currently selling for would vastly improve. Then again bots probably use this price to create higher buy orders which would create a nice price inflation loop for a bit, wouldn't that be fun to see /s.

 

My biggest argument for the auto pricing is that there is such a high volume of keys traded every day that auto prices would make sense for them compared to every other item in tf2. Suggestions for other items work because they are traded on the scale of a few a day to once every 6+ months which means suggested prices are indeed suggestions - this item has sold for this much in the past so it is probably worth this much. Keys are not probably worth whatever that value is on the front page unless that suggestion is <2 days old or auto suggestions for them are implemented. As an example - if we had an auto price of the current top 25 buy orders averaged out (like mentioned earlier) we get a price of 49.88-50 ref. This price is indeed probably what a key is worth if you saw this price on the front page and is indeed about how much ref it would cost for you to buy or sell a key right now.

 

Numbers edit to reinforce just how much more volume keys have: There are 3929 keys sold on scm in the last 24 hours, the next item with most activity is tour of duty tickets at 584.  That's 670% more key sales than another popular item. Lets go to page 6 and click on the catcher's companion - 69nice  sales in the last 24 hours, keys sell almost 5,700% more than that. Obviously these are scm sales so the comparison might not be super direct to the ref/key trade market but it is safe to assume the % differences are probably similar.

[Zeus_Junior]

1 hour ago, FP jh34ghu43gu said:

-text-

Using the reason “accepting as close enough” more often on currency suggestions for the fast flucuating key prices could be a solution. Still not entirely accurate but better than, to quote you, being underpriced by 12%

 

Its a fact keys sell far more often than any other item on TF2 SCM. But i dont think that is a valid argument in this situation. Mosty because keys are currency and tickets (while often considered pure in trades) and the Catchers Companion are not.