trade.tf's pricesheet

[Jymotion]

Please make a setting were you can choose to only search Outpost, backpack etc.

 

I'd suggest using the site's "suggest a feature" link.  I don't know how often he checks this thread.

[F CHARLIE]

The problem with this, is that it only helps lowering the prices. We already have a problem with that. It's easier to post a suggestion to lower prices, but raising them is harder. 

 

However, it's still helpful for suggestions. I'll welcome your spreadsheet as evidence on suggestions. Only if backed up AND if you add the date an amount (if you didn't add it already) as Brad said.

 

First, why is lowering prices a problem?

Second, the spreadsheet works just fine to show price increase.

http://trade.tf/price/143/6

http://trade.tf/price/1037/6

 

Can any of you explain this:\/

http://trade.tf/price/8903/6

http://backpack.tf/stats/8903/6/0

[Jymotion]

First, why is lowering prices a problem?

 

Not Chief D, but I'll give my input.  The issue isn't that it helps lower prices.  The issue is that it really only helps with lowering prices.  The Deals section shows us a bunch of people selling below bakpack.tf prices.  This might lead us to believe the price should drop.  However, what we don't see on the site is the number of trades occurring above the current price.  It's just a bias in the research.  If an item should clearly drop, then it is a very helpful tool.  I used it to lower a couple genuines.

 

Second, the spreadsheet works just fine to show price increase.

 

This means nothing if you don't have faith in trade.tf's system.  The point of backpack.tf's suggestions is that you can look at it and see the trades that caused the change.  If you simply link to trade.tf's graph, you aren't really showing any trades.  You are just trusting that trade.tf's system monitored enough trades and analyzed them accurately.  You are free to use it for your own personal use in this way, but it doesn't serve the purpose of suggestions.

 

Can any of you explain this:\/

http://trade.tf/price/8903/6

http://backpack.tf/stats/8903/6/0

 

No, but I can guess.  Doesn't this item disappear when you boot tf2?  In which case, wouldn't that require a lot of exchanges between alt accounts?  That might explain it, but then again, trade.tf measures medians, not means.  So, no, I don't have a good answer myself.

[F CHARLIE]

Not Chief D, but I'll give my input.  The issue isn't that it helps lower prices.  The issue is that it really only helps with lowering prices.  The Deals section shows us a bunch of people selling below bakpack.tf prices.  This might lead us to believe the price should drop.  However, what we don't see on the site is the number of trades occurring above the current price.  It's just a bias in the research.  If an item should clearly drop, then it is a very helpful tool.  I used it to lower a couple genuines.

 

Well it may be that items are priced too high.  There is an inherent problem with getting proof from the majority of trades on Outpost.  Logically, open trades are ones looking for the highest price.  This would also be true of "pro" traders as opposed to player/trader sales who sell for less but not quicksale price. 

 

Look at supply and demand as a pyramid with low price transaction volume on the bottom and high priced transactions on the top.  We who trade for fun are looking for prices at the top 10 - 20% of the range with quick/fast sales happening in the bottom 30%.  This scenario does not represent the majority of trades which occur between 30 - 80% price range.  I stress this: There are no proofs for trades in that range but they do happen and since we don't accept quickbuy as proof we are only left with high end ranges.  The top 3rd or so.

 

This means nothing if you don't have faith in trade.tf's system.  The point of backpack.tf's suggestions is that you can look at it and see the trades that caused the change.  If you simply link to trade.tf's graph, you aren't really showing any trades.  You are just trusting that trade.tf's system monitored enough trades and analyzed them accurately.  You are free to use it for your own personal use in this way, but it doesn't serve the purpose of suggestions.

 

As noted many times, we can't search for closed trades, we don't actually know what was exchanged and we don't monitor trade servers or convenient trades that happen while a posted trade happens (I will have, every so often, a buyer ask if I have an different item also.  One that wasn't posted at that moment and they will buy that item as well) .  That alone makes Backpack.tf horribly inaccurate.  Trade.tf can at least claim that it scanned transactions as they occurred.

 

No, but I can guess.  Doesn't this item disappear when you boot tf2?  In which case, wouldn't that require a lot of exchanges between alt accounts?  That might explain it, but then again, trade.tf measures medians, not means.  So, no, I don't have a good answer myself.

 

Of-course, your right, the seller would never keep them in his primary account so it is often exchanged for nothing unless they are selling from the alt which most don't.

[Jymotion]

Well it may be that items are priced too high.  There is an inherent problem with getting proof from the majority of trades on Outpost.  Logically, open trades are ones looking for the highest price.  This would also be true of "pro" traders as opposed to player/trader sales who sell for less but not quicksale price. 

 

Look at supply and demand as a pyramid with low price transaction volume on the bottom and high priced transactions on the top.  We who trade for fun are looking for prices at the top 10 - 20% of the range with quick/fast sales happening in the bottom 30%.  This scenario does not represent the majority of trades which occur between 30 - 80% price range.  I stress this: There are no proofs for trades in that range but they do happen and since we don't accept quickbuy as proof we are only left with high end ranges.  The top 3rd or so.

 

I see what you are saying, but I don't fully see how it applies to the bias we were discussing.  The main point of my answer was that trade.tf helps you find people selling below the current price.  It does not help you find people successfully selling above the current price.  It does help you find buyers above the current price, but they are more difficult to come by.

 

That alone makes Backpack.tf horribly inaccurate.  Trade.tf can at least claim that it scanned transactions as they occurred.

 

Horribly inaccurate is an overstatement, in my opinion.  It could definitely be said there are issues with the integrity of the system, given the difficulty of finding 100% valid proof, but I think it does as good a job as a manual system can do.

The key word in the second sentence is "claim".  I trust trade.tf and I use it as a resource, but almost all of its computations are behind closed doors.  Assuming it behaves exactly as we are told, you are correct, but it still requires faith in the system.  For people who have doubts, your argument is meaningless.

[Swag Inc]

I trust trade.tf and I use it as a resource, but almost all of its computations are behind closed doors.  Assuming it behaves exactly as we are told, you are correct, but it still requires faith in the system.  For people who have doubts, your argument is meaningless.

 

Actually both trade.tf and backpack.tf are based on trust.

 

The transparency of backpack.tf is an illusion since the decision is mostly based on a few outpost ads, which nobody can tell how representative they really are of overall trades (e.g. friend-to-friend trades, trade servers, or even actual trade since outpost is just trade advertisement nothing proves the trade really executed at the asking price). On top of that, the trades ads that remain open on outpost are the most expensive, the sold ones are closed and they were almost never used in proof before trade.tf deal function came into play.

 

So if you make the argument that trade.tf is not transparent since it's based on maths and automation, then I can ask what proves that backpack.tf is representative of reality (e.g. of all trades, not just outpost remaining open ads) ? Please let's not make it one-sided.

 

Let us just accept that no system is perfect, and everyone can make informed decisions in using both lists knowing their respective limitations.

[Jymotion]

So if you make the argument that trade.tf is not transparent since it's based on maths and automation, then I can ask what proves that backpack.tf is representative of reality (e.g. of all trades, not just outpost remaining open ads) ? Please let's not make it one-sided.

 

Let us just accept that no system is perfect, and everyone can make informed decisions in using both lists knowing their respective limitations.

 

I tried to cover the issues of both sides when I said, "[backpack.tf being h]orribly inaccurate is an overstatement, in my opinion.  It could definitely be said there are issues with the integrity of the system, given the difficulty of finding 100% valid proof, but I think it does as good a job as a manual system can do."  I agree with your closing sentiment.

[F CHARLIE]

On top of that, the trades ads that remain open on outpost are the most expensive, the sold ones are closed and they were almost never used in proof before trade.tf deal function came into play.

 

This is exactly my point with the pyramid analogy.  The majority of proofs that would generally be found are the top 3rd or so of the price point on the supply/demand curve.  What this means is that the price we show here is always high especially with high priced items with the exception going to items that have increasing demand as a majority of sellers and a majority of buyers will always be meeting at the higher price point.  Bp.tf proof system, and philosophy, was built around how the mann-conomy used to be when the amount of new users, rate of player growth, was high.

 

In other words, when in growth, the accurate price is the leading price near the high end.  When there is decline, the accurate price is the leading price near the low end.

[F CHARLIE]

Here is the problem I have been pointing out.  Lets say I want to resell the Strange Fest Grenade Launcher.  I like to flip items fast (low patience) so I usually offer a decent price to get an item and sell for a little above that.  So if I go to bp.tf price guide I see this:

 

http://backpack.tf/stats/1007/11/0

 

The logical buying offer for me is a little over nine keys.  This way if I don't get a seller at 11 then I still profit at 10.  Additionally, if I go lower than that then it is less likely that I will find a supplier because they will want to get as close to 10 keys as they can.

 

But wait:

 

http://trade.tf/price/1007/11

 

I would have made a big mistake by purchasing at currant trade price "according to trade.tf) and trying to sell it above that.  I recently did this with the S pda and will now take a loss.  Fortunately the robo-hats are turning out well for me.  Also, same for the S Invis watch.  Players are more hesitant to sell at such low offers because bp.tf shows "X" price and I am hesitant to offer more because I am afraid I can't flip it.

 

To me, there is a huge issue here that must be addressed by bp.tf.

[Swag Inc]

F Charlie,

 

I don't know if you saw that, but in

http://backpack.tf/vote/id/519aaafaba2536b606000001

 

Baloo posted an annotated trade.tf graph:

http://i.imgur.com/A9dXbSb.png

 

So technically that drop you see could be the influence of backpack.tf's vote on the market, however it may recover later like it did before. With the graph under your eyes you'll see what I mean. So anticipating recovery could also be a way to make profit from panic sellers. Just a thought.

[F CHARLIE]

F Charlie,

 

I don't know if you saw that, but in

http://backpack.tf/vote/id/519aaafaba2536b606000001

 

Baloo posted an annotated trade.tf graph:

http://i.imgur.com/A9dXbSb.png

 

So technically that drop you see could be the influence of backpack.tf's vote on the market, however it may recover later like it did before. With the graph under your eyes you'll see what I mean. So anticipating recovery could also be a way to make profit from panic sellers. Just a thought.

 

I think the assesment of it being a recovery in price is incorrect.  I look at the trend the preceeded it.  Votes do, however, influence short term behavior, look at the Bills again :

 

http://trade.tf/price/126/6

 

There were rally sales around the last series of votes but after that the Bills continued its decline as before.

 

Also, there is a key in price difference.  That is usually what people hope to make in profit.

[F CHARLIE]

The site now has, along side each strange part and paint, how much that item effects the item it is attached to.

[Hugh Jazz]

That's a really good feature.

[Slocumruls]

Bar graphs showing how many of an item has been sold at X price over 24 hours or 1 week/month would be significantly more useful than a statistical median or min/max without knowing how many trades have actually been recorded to represent any particular period in time.

[F CHARLIE]

Bar graphs showing how many of an item has been sold at X price over 24 hours or 1 week/month would be significantly more useful than a statistical median or min/max without knowing how many trades have actually been recorded to represent any particular period in time.

Swag is working on that. Other concerns are a higher priority for his site.

[F CHARLIE]

I asked Swag earlier about volumes sampled.

 

http://steamcommunity.com/groups/trade_tf/discussions/0/846954921892025268/