A day in the life of our Trade-analysis software

I thought I would share some values that our trade software output from today.  These trades are from midnight until now, around 6:00 PM EST.  This is a typical number of trades we see from a mid-week day for this 18 hour period.  Weekends often have much more volume.

The fourth column is refined metal traded, the fifth is key(s) traded.  Next to each trade I wrote in notes that help you see action taken, if any.  When I say the price matches sheet it either exactly matches or, most often, is in a range around the number on the sheet.  That is why we have a "+/-" or 2 numbers listed because of this implied range.  Again, there are no exact prices.

Actual trade prices always trump voting on prices.  Prices are what people pay, not what self-interested owners think their item is worth.

We will post more of these as we get closer and closer to the site relaunch.

Spreadsheet accuser's evidence examined

Ok everyone, I thought that the steamrep thread for this issue had been closed but someone forwarded me a new link.  I thought that before reddit labeled my account as "scammer" they would take a closer look at this person's manufactured evidence, however, I went through it carefully and have addressed it in detail.

Take a look and read through it.  Examine this accuser's evidence carefully.  I think you'll find like I did he has distorted, exaggerated and flat-out lied.  If you like our website, our prices, giveaways, (even if you don't) make sure you sound-off on the reddit thread, (the steamrep thread was locked because an admin "Datastorm" apparently didn't approve of pro-spreadsheet comments or is just having a bad day).  Perhaps he should use his power for a better purpose - examining why his forum was and is still being used to smear this site with false statements.

(comments here are ok, although I'd prefer you visit the reddit thread linked above and speak there.  When you post here, no external links/sites and keep it clean.)

Thanks for all the comments and support through this issue.  As stated above, you can support us the best by reading through the accuser's "timeline" and our detailed response.  Post your comments on the reddit thread.  When someone you know repeats this accuser's false story, correct them.  Additionally - link to us, use our guide daily, tell us if a price is wrong, and participate in our giveaways.  Big upgrade is on its way, ETA: end of December.

Trade Scanning Software question cont'd...

Another trader had a follow-up question about our Trade Scanning software that monitors actual trades to give us pricing information...(along with our team of expert traders of course, can't forget them).

"How exactly does it work? I mean does it connect to the Steam API and how do you know a price is real and not just a wild trade or something?  Wouldn't a voting system be better?

- Jimmy Fallon's Gangsta Persona"

A quick re-cap of our response to a previous question:

• First off, our software runs 24/7/365. It is still a work in progress and for our site revamp we are going to have it integrated into the site so that we are real-time. It shows prices of trades that occur and the value in keys and/or metal that is exchanged. For now, I go through the output which is a mysql database and examine each trade.
• Many of the trades from the software (80-90%) confirm prices already on the spreadsheet. So if you don't see a price change, it often means that is just where a price has 'settled'. If a trade shows a price different from the spreadsheet I try to confirm this with feedback from my pricers as well as publicly listed prices for items (outpost, tp, trade servers). About half the time a trade is not for the price listed, there is no confirmation in the marketplace. In that case, the price stays the same. What causes this? Typically a quicksell, someone just trying to liquidate. The other half of the time, the price has indeed shifted and after the same research as above, we adjust the price on the sheet. And if there are multiple trades in the database showing a real change in price then we look for confirmation but ultimately require less proof since it is already there. Most often in this case the price is adjusted.

In response to Jimmy's (great name btw) question:

While we can't divulge exactly what happens with the software, it does track trades using the Steam API.  We see the relevant information: the item, the quality (genuine, vintage, dirty, etc), the date and then a value that was exchanged for it.  It can be in keys or metal or both.  When we have it tweaked to be real-time we will have a 'ticker'.  More on that later.

With trades that are wildly out of the market-price range, based on standard deviations, they are automatically ignored.  Anything that sneaks through, say, for a new item that has no average price yet, as I said we inspect each trade so I will just see the discrepancy then.

Voting on item values is not a good alternative to actual trade prices.  I get emails every day from people who think their item is worth a HOUWAR because it has the same rarity.  There are teams of people out there whose goal is to bump the prices of their own items.  Remember the B.M.O.C. back in the early Spring?  Just like the stock market, prices are what people pay (not what owners think their loot is worth).

Integrity of the spreadsheet

Unfortunately we've had to address these types of accusations before.  Running a site like this, it comes with the territory.  However, this time things have taken a pointed tone.  It seems that a particular reddit user has been monitoring my accounts trying to catch us doing misdeeds via this little spreadsheet.  He has done his digging for seemingly a long time and posted his attack on our credibility.  He found that we've bumped our own threads on Steam user forums via alternate accounts, said nice things about our own site, and not been so nice to a previously competing site.

However, the proof he offers up that we are the devil-incarnate and are only out for our own gain (ignoring that we've given away over 70 Keys and Tour of Duty tickets combined in the past 3 months alone) is that we purchased an item before the sheet was updated and profited 2 keys the next day.

Here's what occurred: we were gathering pricing information, particularly the botkiller weapons.  There were no prices updated at that point on the sheet.  A particular user appeared to be quickselling his botkiller medi gun at a low price of 8 keys, as opposed to the 10 to 14 keys on average (up to a buds) everyone else was asking.  We bought this item for 8 keys.  Updated all prices to the ranges we found (10-14 keys).  The next day, we re-updated the prices to their current (9-10 keys) number and sold it for 10 keys.

Look.  Only the reddit poster truly knows his motivations.  We update this site daily, spend money to keep it running and upgrade it, hold contests where we give away 90% of our items.  I trade not for profit but to keep current on what's going on, I check our contributors prices against multiple sources.  We view his accusations as illogical but wanted to let you know about them since we don't want this to be just a reddit witch-hunt.  Questions and comments welcome, positive or negative but keep it PG.

Update: We exchanged some calls and emails about this.  Going forward no one associated with this site will trade any newly released items before there is a price in the guide.  Bad form on my part, no malice intended but I understand why it was not the right thing to do.  Rare or glitch items, low craft #'s, unusuals, are exempt.

Update to the update: I will no longer be trading.  We believe it is a better policy.  My contributors will continue as discussed above but we agreed with our Steam API trade analysis software running it isn't necessary for me to trade anymore.  Lurking however, is still fair game.

Hindsight is 20/20:  This accuser stacked what he presented as "evidence" without the bulk of asking prices for the item which were much higher, cherry-picked one trade we did out of thousands he witnessed, mis-stated the timing of events, and never answered questions about his own reason for being disgruntled nor his  affiliations.  Reddit labeled my accounts as "scammer", likely retribution for pointing out the inexperience of their own redditor's defunct pricing site.  Meanwhile, I've stopped trading and we've continued giving away many dozen keys, tickets, hats, and more.  We use Steam API trade-scanning software, as well as our contributors, to provide the most accurate pricing available.

Sheet slowness issues (update: fixed)

As you are probably aware, the spreadsheet is very slow at times or not loading at all.  We are working on this and hope to have a fix done sometime today, Sunday.  Until then it is possible that the sheet will take up to a minute to load when you hit the page.  Let the page load, if you see the refresh button available in your browser then it didn't load and just refresh and repeat until it does.  When it does load up, keep it open in a window and don't refresh the page.  We are aware of the frustration and are getting a resolution in place.

Thanks for your patience.

*update: the sheet should be much faster now.  We set up several mirror images of the data.  While the site itself is on many of Google's servers all over the world, the sheet comes from one location.  That bottleneck looks to be the culprit.  We ran through most everything from removing rows to eliminating unnecessary formulas to getting rid of the Unusuals tab (which should have been done long ago).  But the fact that everyone is hitting the same sheet seems to be the issue.  From now on we'll have to turn off the sheet for a few minutes when we update everything but this approach should be much better on speed.

Spreadsheet loading issues (updated)

We are aware that since around 1pm EST yesterday that the spreadsheet has been loading slowly or not at all.  There is a heavy volume of users and there appears to be a Google Docs slowdown.    Until this is fixed the sheet may not appear immediately or at all when you load the site.  In that case refresh your browser until you see "waiting for spreadsheets.google.com" in your status bar.  It is slower than normal and we apologize for the inconvenience.

* Update: we are working with Google to get the availability issues fixed.  In the meantime, we've embedded the raw sheet in a framed window below (was above) the usual spreadsheet as a temporary measure.  If it is blank, click on the "key" tab and then switch back to the "Master Sheet" tab and it should render.  Use Control-F to search.

Spreadsheet upgrade on its way

A lot has been happening loyal traders!  We've been working with our developer to bring a more useful UI experience to the spreadsheet and like what we see so far.  Multi-column sorting, custom search results, and more!  We are excited to get the prototype live but have to wait a wee bit longer until some of the finishing touches have been added.  Stay tuned.

How to interpret prices

We see that some misinterpret the sheet and want to make sure it is very clear what the prices mean.  Prices are in refined metal, unless noted.  When an item trades in keys or buds typically we will reflect that on the sheet.  Most importantly: prices are averages and never absolute.  If a price says "3 +/-" that means it could be 2.66 refined or 3.33 refined is the range we see as an average price.  Could that mean it trades for 2 ref?  Not likely and you are likely to do a transaction closer to the range stated...although that could involve time.  If the price says "2 - 3" that means on average prices are between 2 and 3 refined.  Could that item trade for 1.88 refined?  Certainly.  Would it trade for 1.33 refined?  Not likely a good deal for a seller and you could take more time to find a transaction closer to the range stated.

Someone who wants an item will likely pay a price in the area of what is on the sheet.  Traders who are only looking for profit will offer lower and/or try to sell higher.  Prices are not infallible and the sheet is briefly behind the market as changes in prices inevitably happen.  When trading and someone tells you something that is different than what you believe or is on the sheet, get several opinions.  If you think something is off, tell us.

Quick tidbit on integrity

'Allo fellow TF2'ers.  Every once in a while we like to address things going on in the TF2 community.  Today is a busy one, as you can already read about the Genuine Purity Fist & Clean 2011 Skull post here.  However, it is also important to talk about some hearsay about our little spreadsheet and "price fixing".

While we have already discussed our methodology, we never thought our actual integrity was being questioned until some other competing websites started popping up looking curiously similar to ours.  Their creators were claiming that TF2 Spreadsheet (that's us) is corrupt!  With no evidence of course they go on to accuse us of price-fixing for our own gain, therefore their new price website is the white knight in shining armor coming to save the day!

Let us just say this: we would not have lasted as long nor be as often used a resource if people thought our prices were fixed.  All of our editors are active traders who submit their price reports and I, your humble Admin, compile them to one price for each item.  I personally do not trade for profit anymore.  My gains were made trading unusuals (we don't even offer prices on them) and regular items way back in the early months of 2011.  However, I am on trade servers and forums for several hours a day.  Yes I am probably watching you right now.

So when someone says something insane like we're fixing prices, and when they can't offer proof tell them to go and play Naughty Bear instead.

Thank you TF2 Community! Over 440,000 hits in 1 month!

We would just like to thank everyone for visiting our little spreadsheet and putting your faith in us to bring you up-to-date and accurate information on the TF2 trading world.  You have been hitting on us like Ron Jeremy at a Swedish Bikini Team meeting!

Update: the prize has been chosen and acquired.  You will want to win this prize.  That is all.  Be well.

New item percentages

In updating the TF2items hat and item percentages, you may notice that most of the items percentages have dropped by approximately half.  For example, Bills hat which was in around 20% of the backpacks is now at around 11%.  What gives, you say?  Well, as we've said before TF2items (and thus, we all are) is dependent on Valve for whatever data they give out.  The percentage drop is likely due to the influx of new players, and it would seem that the number of current players has nearly doubled since all percentages of items has been effectively cut in half.  This is reinforced by the Ghastlier Gibus percentages - cut in half also signifying that more than half of the active player base was not playing Halloween of 2010.

Of note, you will recall we ran a poll asking which stats you would prefer: percentage of items or estimated # of items.  We have gone through our channels and Valve will not give us the actual # of active TF2 players.  Sooooo...we are "stuck" with the better statistic which is percentage of items.  I know many of you really liked estimated #, however it is a severely flawed number and even moreso now that TF2items data from Valve keeps getting more limited.  If the number of players is ever released we will re-calculate an estimated # of items.

That said, if you think there are any other metrics that effect value and price that we should show in our little spreadsheet: let us know.  That is all.  Be well.

New poll

Speaks for itself.  We all hate converting Bill/Buds to refined and noone trades them that way anyways (even though they can trade more than 8 items rather easily with the new Steam trading beta).  And since we started putting Max/HOUWAR/Medals in terms of buds why not switch to keys?  Your opinion counts.  Let us know.  Be well.

Poll closed, you've voted that you are undecided!

So we could see where things were going with our poll regarding % of items vs. the old # of items.  From the start it was a horse race and didn't let up until the end.  The "# of items" choice ultimately won out by a mere 19 votes.  Because we appear to have a split we will keep the % of items and re-add the # of items column while doing all the math for you.  The only caveat is that we want to use accurate data on how many actual TF2 players there are currently - which we will attempt to get in good faith and with haste.  When we get that mystical # we will re-add the # of items back.  Fear not.

I like % ownership, it's more accurate	140 (39%)
Go back to # of items owned, its not perfect but I like a number	159 (44%)
I don't care.	55 (15%)

Votes total: 354

TF2 Google Spreadsheet has moved!

Yes.  The spreadsheet has a new home.  While editing the latest item count percentages which replaced the item counts, I realized that there is never enough room for commentary on the items or what is happening in the TF2 marketplace.  This will allow the spreadsheet to remain the most important item but also to make commentary on the TF2 marketplace possible.  It will also allow more information on statistics to see where people are coming from to see the spreadsheet.