Your web-browser is very outdated, and as such, this website may not display properly. Please consider upgrading to a modern, faster and more secure browser. Click here to do so.

one crate per week

i, buddhapest, will burn one key and crate per week until i either run out of money, die of stress or malnutrition or old age, or get an unusual. unboxings will typically be done on Sundays at 20:00GMT
Oct 13 '13


greetings and salutations 

welcome to the last entry in the one-crate-per-week blog. thank you for coming along for this ride, your fascinating feedback and encouragement, and most of all, thank you for giving me a small piece of the most precious thing you own: your time.


i want to leave a proper tombstone at this link, and that is a Table of Contents of what i consider the more interesting articles from my 33 week journey (incredibly not all weeks were particularly interesting :D)

after the ToC i’ll give you some interesting numbers, some closing thoughts, and most importantly a warning.

one-crate-per-week Table of Contents:

the myths and maths of unboxing:

musings on philosophy, behavior, and their parallels in team fortress:

the tf2 economy:

it’ll be fun to revisit some of these articles in the future. i’m sure my naiveté will border on the embarrassing.

* * *

now i’d like to re-introduce you to the unusual i unboxed last week:


the name is an homage to this blog, the probability of the unusual unboxing event. the description is the opening line of O Fortuna, a medieval Latin poem which bemoans the nature of Fate and is part of a massive 11th-13th century poetry collection named Carmina Burana; best known because of Carl Orffs 1936 musical masterpiece.

the description means:

O Fortune, you are ever shifting like the Moon, always waxing, always waning

the unusual itself is nothing special but as i mentioned last week, it’s an amazingly apropos hat to unbox for this blog. while i probably could have rationalized a bubbling larkin somehow, the fact that i unboxed the ordinary version of this very hat in week 13 and proceeded to make fun of it, and the fact that i’ve been writing about team fortress for the last 8 and a half months is an interesting coincidence.

and the coincidences did not end there.

i wanted to establish a baseline value for the unusual so that i would be able to update the final yield. because did not report any logo titanium pillars and because and tf2 price check did not have a value yet, i went looking to see if such a hat existed on outpost.

it did. i decided to add the owner. to my surprise he said he was just about to add me. however, he was about to add me not because we owned the same hat, but rather because he has been investigating one of the myths i wrote about in week 15 (new effects and old crates). he knew that i had unboxed a tf logo (an old effect) and that it came from crate #24. (fwiw, a survey of some sort is something i am still noodling on)

but it was only after i pointed it out to him that he realized that we had the exact same hat!

and if you think the coincidences end there? nope. turns out that the original unboxing spreadsheet that long ago got me thinking about the value of crates was his creation as well!

if only there were more weeks to write about Coincidence, there’s so much to say on the topic. i’ll give you a tl;dr version here:

we experience events non stop, 24/7 all day baby, all day. what we call coincidence are simply events which we remember because they were significant.

but that doesn’t mean we don’t stop and go woah!


* * *

assigning value to my unusual has been difficult. experienced unusual traders have said everything between 1-4 buds which is a useless range. given what other titanium pillars sell for and what a logo non robo pillar sells for i think the value is probably around 3 buds. it really doesn’t matter as i’m not likely to ever sell this hat.

it is, in all its ugliness, iconically themed given its origin and the perfect memento of the journey.

* * *


i thought long and hard about whether to continue this blog or not. i finally decided: “this is the end”.

i didn’t want the blog to become a chore to write (and read).  and i didn’t want to keep unboxing.  i decided to walk away a “winner” in this particular gambling scenario.

but before i sign off, i wanted to give some parting thoughts on the state of the TF2 economy and what it means to have an alter ego in a virtual reality; things i planned to write about over coming months if there were more crates to unbox.

* * *


wealth is relative. although it is sometimes difficult not to covet incredible items in someone’s backpack, just keep in mind: wealth - is - relative. i was once in a meeting with some unbelievably rich men in the silicon valley. where you and i might talk about the best price for 4 lbs of chicken, they were discussing the upside of upgrading their Gulfstream GIV-SPs to the GV. wealth is relative. don’t get sucked in by greed. greed is not good (and fuck Gordon Gecko). greed destroys relationships, economies, ecologies, and, eventually, your soul. enjoy the game of team fortress itself and don’t waste too much time in the trading world possibly stepping on people along the way just to make a small virtual profit.

* * *


pricing guides are guides; prices are not, and should not be set in stone. if you make the mistake of trusting a pricing guide, you will eventually fall prey to price manipulation. traders hoard goods and manipulate pricing all the time, a lot more than you know. and even when it is completely innocent, our most commonly used pricing guide is often misguided, delayed, and incorrectly used by too many traders.

* * *


anonymity on the internet doesn’t give you the right to be an asshole. in fact, take a step back and think hard about this: the availability of this anonymity shows who you truly are on the inside because your social inhibitions can be (and do get) put aside. how do you treat your fellow anonymous virtual denizens? do you make racist and misogynistic statements without remorse? are you a raging shithead? perhaps that’s who you really are. are you happy with yourself for being such a person? ‘cause it’s not too late to rehabilitate yourself. better yourself.

* * *


and finally… please be aware of what you’re getting into when you are gambling.  remember:

i was remarkably lucky!

my chances of unboxing this familiar were not even 1 in 5 for 33 crates. don’t use my luck as a motivation to go to the mannco store and blow $82.17 on keys. i have actually unboxed more like 60 crates, the 33 are only since i started this blog. this is my first unusual. i unboxed in slow motion allowing me to trade for keys and not having to spend real world money. most importantly, statistics are always going to win once the sample size is large enough. what does that mean? a 0.66% chance of unboxing means you will get an unusual approximately every 151 crates ($375.99 in keys). the average unusual is worth $179.76.

this gentleman has unboxed 4300 crates and yielded 25 unusuals (a rate of 1 unusual every 172 crates or ~0.58%). even buying keys on paypal at $1.75 that’s $7525 ($10707 at the mannco store rate) for average return of $4494.

so please don’t use my experience as inspiration to gamble. if you want to gamble understand the risks, mitigate some of that risk by unboxing intelligently, and i hope that our journey together has helped you realize that there’s countless ways of looking at value, quality, chance, and more.

thank you for your company, your time, eyes, and thoughts.


Oct 6 '13

week 33

probability of unboxing an unusual after 33 attempts?  ~19.81% (probability today?  0.66%)

fall is here, halloween approaches, and of course that means yet more event crates. hurray Valve, thanks for the October 2nd update and the “mysterious” acorn and gourd crates.


there was definitely a collective groan felt around the TF2 world - but Curiosity as well. what’s in them fall crates this year?

the magic eight balls says:


but hang on, there’s two of these crates. in 2012, there was one fall crate and it contained hats (and miscs, including the still highly popular robro 3000).

the last time we had 2 event crates, the naughty and nice crates of 2012 and 2011, we got one crate filled with hats/miscs and the other with weapons, some strange. of course those were “festives” and historically those come out at christmas and besides, what would a festive fall weapon be?  a shotgun that shoots acorns?  a strange kritzkrieg adorned with fall foliage?

perhaps the contents of these crates will be different altogether. perhaps we’ll see more strangifiers, or more chemistry sets… but most likely, community items, of which there is no shortage, some of them absolutely fantastic:


i honestly don’t want to prognosticate since i prefer numbers and stats.

but i did want to talk more about this subtle disgruntlement some, occasionally me, feel toward Valve because so little has changed in the game - everything revolves around hats.

and how is it that the tf2 and the economy around it survives?

you sort of have to take a step back too see a clearer picture. although i’m definitely not enough of a gamer to know about all the different virtual economies out there, it’s pretty clear to me that tf2 may be one of the best games where personal appearance is so well integrated into a game that doesn’t suck.

lets consider for a second the other super popular Valve games today, Dota 2 and Counter Strike.

we’ll start with Counter Strike, which in CS:GO recently received its own economy: items, crates, keys, etc. but the problem with customization (which is vital for a thriving economy) in CS is that it’s a

Very Serious Game

care must be taken not to give players advantages or disadvantages based on how they look (or what they can afford in real life). so far, the economy is limited really to just skinning and statistic tracking (strange) weapons and while it’s possible items like tattoos or maybe even payday style masks will eventually make it to the game, the variety of items will never match that of TF2 because the interest in customization will always have to be very carefully balanced with game play.

this isn’t going to happen:


* * *

at first glance, Dota 2 seems an obvious candidate for a capacious economy. it’s based in fantasy, with magic and legend allowing a virtually endless variety limited only by the imagination of the designers.

and in fact a very healthy economy does exist, including some very cool looking equivalents of the tf2 unusual hat, the unusual courier (some of them commanding astronomical prices).

but the problem with Dota 2 is the presentation. you’re playing the game practically top down and, to have the most information, zoomed out. it’s not the most interesting (or flattering) way to portray your custom loadout.


without a doubt dota 2’s economy is robust tho it is difficult to say if this is simply because of the much larger number of players. as of this writing, for example, valve’s own stats site shows nearly an order of magnitude (that’s general geek speak for tenfold) more players in dota than in tf2 (and tf2 trouncing cs:go almost 2:1)

* * *

really the whole idea of cosmetic items is silly in a game. but tf2 is a silly game! it’s the perfect platform for such a weird mix of levity and competitive entertainment.

1. it’s incredibly well balanced which is surely one of the reasons why Valve doesn’t dare mess with the actually gameplay too much.

2. it takes real skill, not just luck or a fat wallet, to be a great player.

3. it is full of laughs which is why it’s okay for a great (and generous) player to also look like this:


assuring its longevity is the very casual nature, be it dropping into a game for 30 minutes between classes or playing for 16 hours straight while out on furlough…

and the trading community is still ebullient despite its age and the limited attention span of most gamers. there have been many cries that the end of tf2 is near but it seems to me that interest is still far from waning.

* * *

now onto the business at hand.

for this week, i’m opening crate #24 because i once again failed to find a good trade for a brown cooler key and i haven’t bought keys in the tf2 store since… well, since the first and last 8 long before the one-crate started.  and #24 is again 2nd best according to my spreadsheet.

(oh, and speaking of keys, a big shoutout to Snap who added me a few days ago to donate a few to the project! much obliged for your gift sir!)

and now for my weekly crack at 0.66%…

[creaking sound]

















and then


and so it was that on october 6, 2013 at 18:19 Greenwich Mean Time while playing on skial’s 32 player dedicated turbine server in the company of my great buddy Rew (who made the image above :D) i turned my demo three times counterclockwise and unboxed the one-crate-per-week unusual.

* * *

funny fact that Rew pointed out: i made fun of the ordinary version of this hat when i unboxed the one and only robocrate in week 13. the effect is certainly nothing incredible either, but the cool thing is that as of right now, i’m only aware of one other unusual like this in existence. and it’s a great effect for someone who has been writing about team fortress for the last 33 weeks so really the perfect hat!

it’s silly and ironic and iconic (literally)…

i’ll write more about this hat and my plans for it and the tumblr in next week’s entry (so there’ll be at least one more :D) but for now…

the community price hasn’t been set on this unusual yet, but my initial and prolly naïve estimate is that this hat is worth somewhere in the neighborhood of 3-4 3 buds and therefore:

tally after 33 weeks:

  • spend: 189.55 refined
  • return: 487.66 refined 430.66 refined
  • return/spend ratio: 2.573 2.272


Sep 29 '13

week 32

probability of unboxing an unusual after 32 attempts?  ~19.27% (probability today?  0.66%)

there are several ways of looking at the results of an unboxing.

other than:


for one, there’s the Quality of the item unboxed, and then there’s the value, and sometimes they are not tied together. you might unbox something you don’t care about at all, but someone else might and so it still has value.

that is the main reason why i built the spreadsheet - to give various ratings or metrics to each crate so even if you don’t get your unusual, even if you don’t like the item unboxed, you can still make out ok and not lose the full value of your key.

i ran across an interesting thread on reddit posted soon after the crate spreadsheet went live. in this thread the author claims that by opening just the brown cooler (series #68) one can actually turn a profit.

this isn’t true, because the author messed up on their understanding of probability and overvalued the average unusual, but it also made me realize that while opening 1000 crates would in theory provide a good return, in practice you’d have to be unbelievably crafty and lucky to actually somehow manage to sell all 1000 items to reach your average yield.

clearly, no one is going to open 1000 brown coolers, but it does make for an interesting thought experiment.

as an aside, i’ve updated the spreadsheet to include the specific key pricing for each non-standard key - my experience has been that it’s nearly impossible to buy brown cooler keys for the price of one standard key, but neither nor reflect this yet i imagine due to the extremely low volume of trade (i see 4 trades on outpost and they’re primarily trade one color for another or uncraft for craft, don’t ask me why) - most would be buying these in the valve tf2 store. i think i may have to do my own adjustment in the future because in the marketplace the keys sell for more than they do in the store (can anyone explain that to me???) since early september.


anyway, this reminded me of several conversations i had with poncho before and after the Chance of Sadness revelations in week 23.

as much as crate #42 has been a thorn in my side because of all those bottles and rakes, it’s still actually a damn good crate.

the most important thing to realize about #42 is that you don’t even have to resell the bottle or rake or even pistol and you can still make a good average yield.

note: as always, the following analyses is only valid around the date of this posting

so lets compare crate #42 to crate #28 again and this time do it in terms of a mass unboxing - if i were to open 100 #42s and you 100 #28s statistically we’d end up with:

yield from my #42s:
~20 rakes (0.33 ref each) 
~20 bottles (0.22 ref each)
~20 pistols (2 ref each)
~10 strange part: headshot kills (2.6 keys each)
~10 strange part: buildings destroyed (1.6 keys each)
~10 strange part: projectiles reflected (1.2 keys each)
~10 craft hats (1.33 ref each)

yield from your #28s:
~22 shotguns (1.2 keys each)
~22 jags (2.77 ref each)
~22 frontier justices (2 ref each)
~22 strange wrenches (1.88 ref each)
~5 safe and sounds (1.66 ref each)
~5 mining lights (1.33 ref each)

~ means approximately, as there is surely deviation from a relative small sample such as 100 but just bare with me anyway

now check this out:

if i sold just the strange parts from crate #42 (~30 items), i would pull in somewhere around 55-60 keys as my yield.

you would have to sell ~66 crate #28 items to reach around a 49 key yield and ~88 items to hit 60.

making the naive assumption that it takes about the same effort to sell each item, you would have to expend almost 3x my effort to realize your yield. but even if one accounts for strange parts being more difficult to sell, it’s hard to argue that selling 30 desirable (otherwise they would not be priced so high) strange parts is preferable to selling 88 strange weapons.

in other words, even though the #28 is a lovely crate, in a mass unboxing it might prove to be a pain in the ass to realize your average yield.


(this awesome photograph is by boris reinhard - used without permission, please don’t sue me boris!!!)

i think in the near future i may need to look into making yet another metric that captures this phenomenon but right now consider it a placeholder and a reminder:

just how much work are you willing to do to realize your average yield after you do a bulk unboxing???

and by the way, in the case of the guy unboxing 1000 brown coolers, by the time he would be done selling even a portion of those items, the prices on them would have fallen, and most likely in part because of him flooding the market with ~125 instances of each of the 8 items!

* * *

so is crate #42 off the shitlist again? well, no. for casual unboxing it is still poison but for bulk (or gambling) it’s quite an attractive crate.

but the appeal of crate #42 is finite.  just how much longer is it going to have the yield that it currently commands?


(please don’t sue me banksy!  and yes, i’ve used this image before - and i will use it again - also, it’s my tf2 spray.  so sue me.  no wait, don’t)

consider the following:

strange parts were introduced in crate #41, a normal crate that dropped between 2012.03.22 - 2012.05.17 (56 days). it contained (among other things):

  • Strange Part: Airborne Enemies Killed (briefly worth as much as 3 keys, now worth 2.33 refined)
  • Strange Part: Heavies Killed (worth as much as 1.5 keys, now worth 1.33 refined)
  • Strange Part: Gib Kills (once worth as much as 2 keys, now worth 2.22 refined)

why did these strange parts fall so dramatically in price?

on 2012.12.20 crate #55 was introduced. along with 2 new parts (Pyros Killed and Defender Kills) it re-released the Airborne Enemies Killed part. immediately, the value of that part plunged.

on 2013.06.19 crate #59 was introduced and with it the Heavies Killed part which also plunged.

and on 2013.08.27 crate #71 re-released the Gib Kill part (as someone who resells a lot of strange weapons with parts i can tell you that one really really hurt)

but also a pattern emerges. since december of 2012 and the release of crate #55, whenever a new normal (non-rare, non-event) crate got added, Valve introduced 2 new strange parts and re-released one previously released one.

so far only 3 strange parts have been re-released and all of them from crate #41.

logically it follows that the next normal crate will contain two new strange parts (perhaps Strange Part: Assists and Strange Part: Times You Wish You Had The Perfect Comeback To The Guy Who Just Trolled You) and one strange part from crate #42.

whatever part is released, but especially if it is headshot kills, the average yield of crate #42 will plunge. and the two normal crates that follow further down the road will nail crate #42s coffin, forever making it irrelevant (not to mention making your strange huntsman with buildings destroyed and headshot kills lose 2-3 keys of value).

this is an interesting dichotomy for someone who, like me, sells weapons with strange parts but also relishes the metrics they provide.

for me, the Buildings Destroyed part is one of the most meaningful of all strange parts. along with Allies Extinguished and Sappers Removed mounted on weapons like the Homewrecker it shows that you are a team player and are not just death matching while your teammates get wrecked.

anyway, i think crate #42s days as a good unboxing crate are numbered. new normal crates seem to be introduced around every 2-3 months or so. my guess is we’ll see our first #42 neutering crate appear around november, and by march or april of 2014, it’ll fade into obscurity. until then, enjoy the high yield of crate #42 and prepare to mourn the passing of one of the most interesting of all normal crates.


(please don’t sue me mr. copola!)

all this talk about crate #42 probably makes you think i’m gonna open one.

nope. too many bad memories. remember, it’s still a terrible crate for casual unboxing. if it’s not on the casual top 10 i’m not opening it.

i could not find anyone to trade me a brown cooler key for a normal key 1:1 so i’m opening the 2nd best casual, crate #24 (which is #42 backwards, haha)

[creaking sound]


okay, that’s a good return (3 ref today). i have to be careful how i feel about these unboxings because in almost all cases i raffle these items off anyway so really i’m losing a key plus the cost of a crate. but since someone out there is gaining a 3 ref item as a result, i will continue to…

tally after 32 weeks:

  • spend: 183.22 refined
  • return: 88.66 refined
  • return/spend ratio: 0.484
Sep 23 '13

week 31

probability of unboxing an unusual after 31 attempts   ~18.73% (probability today?  0.66%)

gotta make today’s entry very short. it’s been a busy week and an even busier weekend.

first, i have a new banner for the tumblr which was made for me by one of my bestest buds: Rew.


it’s pretty much perfect i think as i too have sat in my room of crates at home contemplating the many bottles and rakes (and where to stick them…)

* * *

also, i’m super stoked to announce that last week i took all my various bits and pieces of number crunching scripts and put together a Crate Analysis web page.  i’ve been working on a way to present this data to y’all for many weeks but i’m not much of a UI designer (although there are worse).

the new tool is now hosted at the excellent


all the various ramblings about sadness, disappointment, and yield are now in one easy to use (i hope) tool.  so enjoy your new unboxing knowledge (and please leave feedback)

to celebrate the launch of the new tool, i am opening the brown summer cooler which is, at this writing, at the top of both the top 10 bulk and top 10 casual lists.

[creaking noise]


this crate just plain rocks!  (and as i have 2 of these already, one a low craft, someone at tf2r just got lucky as well)

tally after 31 weeks:

  • spend: 177.05 refined
  • return: 85.66 refined
  • return/spend ratio: 0.484
Sep 16 '13

week 30

probability of unboxing an unusual after 30 attempts   ~18.18% (probability today?  0.66%)

DE-Climax, an excellent steam buddy of mine who also happens to co-host the awesome tf2 weapons 101 tumblr that everyone needs to follow (if you’re not already), asked me a really interesting question the other day:

do you think valve made the whole split crate/key system back when this all started instead of just letting people buy crates for 2.50 that you could open without keys? do you think it was just a temptation thing, like they knew some people wouldn’t be able to resist if they saw the crate in their inventory?

i don’t know how the whole crate and key thing came about at Valve but tf2 certainly isn’t the first game to use a virtual economy, treasure chests, a sort of gambling system, and there are mainstream and more obscure games from a decade ago that took advantage of this sort of scheme.

rather than dive into the history of these economies lets talk about why a crate and key scheme works.

although true f2p didn’t come to tf2 until june of 2011, the game was, in many ways, already very affordable, lets call it “cheap to play” (c2p), mainly because it was released as part of the Orange Box.

after 3 years valve needed new ways to monetize what was now an extremely popular online FPS all while keeping entry into the game as economical as possible.

on september 30th, 2010 Valve introduced the Mann-Conomy update and with it came the short lived (just 19 days of drops) crate #1, the Mann-Co store, and the ability to buy a key to open this crate.

as i mentioned in week 17, unusuals were also introduced, but were extremely misunderstood.

in designing the crate/key system for tf2, Valve played on some very powerful emotions and drivers: competition, peer pressure, vanity, a false sense of fairness that i’ll address in at a later time, and curiosity.

so this week, lets talk about Curiosity a bit.

researchers delving into Curiosity classify it as an emotion. the word emotion itself has a very interesting etymology and history.


ways to describe our feelings existed even before language, but the actual word didn’t appear until the 16th century and rooted in the latin movere (move) and the much latter french émouvoir (excite).

modern psychology teaches that emotions are an adaptation that allowed animals to survive, multiply, and thrive much more effectively.

the 18th century philosopher (and politician) Edmund Burke said:

the first and simplest emotion which we discover in the human mind, is curiosity.

Curiosity has been linked by psychologists to motivational mechanisms in the brain and is considered to be the most vital component of scientific inquiry, or any inquiry for that matter.

we don’t actually understand the exact neurological processes that occur in our grey matter but some evidence points to release of dopamine when our Curiosity about something is satisfied. (dopamine is called that because it makes you feel dope!) the satisfaction of Curiosity has therefore an internal reward system, primitive but extremely effective.

as a driving force behind all inquiry Curiosity could be said to be responsible for everything from science


to religion


to killing cats.


* * *

the subtlety that DE-Climax mentions at the top of this entry is paramount. one could argue that paying USD 2.50 to buy an item that then immediately opens to reveal a randomly picked item (including the small chance of an “extremely rare object”) would already play to satisfy our Curiosity.

but that isn’t actually so. you’re much less likely to see a movie without a trailer, a poster, a teaser of some sort. in fact the teaser, unlike a product ad, builds suspense and suspense is really just an aspect of Curiosity.

the crate is the teaser. it shows up in your backpack and there it sits poking at you “open me, open me”.  a constant reminder, a constant source of Curiosity that has only one path to being satisfied.

and most of all there’s that blue text:

or an Exceedingly Rare Special Item!

i’m curious to see what the random number generator has for me this week. even tho i’ve done extensive number crunching on these crates and have opened them week after week, i still get excited when this moment arrives. so lets pop open this crate #28 and…

[creaking sound]


nice. 2.66-3 refined and the 2nd best item in the crate.

tally after 30 weeks:

  • spend: 171.00 refined
  • return: 80.77 refined
  • return/spend ratio: 0.472
Sep 9 '13

week 29

probability of unboxing an unusual after 29 attempts   ~17.63% (probability today?  0.66%)

this week as promised, a closer analysis of the new crates, #60 and #71, strangifiers and chemistry sets, and then a quick word on what has Valve done for us lately.

first the crates.  crate #71 is the standard drop rate crate with the following contents and prices (as of this writing and based on

  •         Battalion’s Backup (19.80%): 1.11r
  •         Hitman’s Heatmaker (19.80%): 3.33r
  •         Bonk! Atomic Punch (19.80%): 2.11r
  •         Strange Part: Gib Kills (9.90%): 3.44r
  •         Strange Part: Kills During Victory Time (9.90%): 1.56r
  •         Strange Part: Robot Spies Destroyed (9.90%): 0.17r
  •         Buccaneer’s Bicorne (4.95%): 1.33r
  •         Soviet Gentleman (4.95%): 1.33r

as of this writing, the prices of all the items (except the hats which were already craft hats) have dropped dramatically already.  the gib kills part was originally introduced in crate #41 and commanded a very nice price until 12 days ago.  and it’ll continue to drop.

i know traders who were able to shark people out of 2 keys for the heatmaker, but they are the rare minority.  as the number of crate #71s grows, so will the supply of strange heatmakers and the price will continue to fall.

already crate #71 has an average yield of just 1.96 refined with no chance of profit and a 49.5% chance of getting something rage worthy.

so my current evaluation: it’s not the worst crate to unbox and i doubt it ever will be, but it’s definitely not in the top ten.  it’s not great for bulk or casual unboxing.

* * *

crate #60 is an entirely different beastie…


the contents and prices as of this writing:

  •         Disciplinary Action (19.80%): 7 keys
  •         Loose Cannon (19.80%): 11 keys
  •         Fan O’War (19.80%): 4 keys
  •         Pomson 6000 Strangifier (19.80%): 9.2 keys
  •         Strange Part: Kills with a Taunt Attack (19.80%): 5.2 keys

the loose cannon is a fantastic, long misunderstood, and recently modified weapon and it’s great to see it as a strange.  the pomson 6000 strangifier has an added bonus - it can be applied to existing pomsons which means those which were crafted.  my friend poncho (he of the crate #28 fame) got himself a pomson #57 and strangified it last night.  brilliant.  think about this for a second.  there will never be more than 100 of these low craft, strange pomsons!  (apologies for the screenshot.  my fonts got all messed up when we were playing)


back to the crate #60 analysis.  the crate can be purchased for 6.4 keys and has an average yield of 7.2 (but don’t forget to add the price of the opening key).

there’s therefore a 39.6% chance of making a profit, and the maximum possible profit is 3.6 keys.  if you get the fan, you lose 3.4 keys and the taunt kills part will lose you 2.2 keys so the chance of sadness is also 39.6%.

so if you’re the gambling type, this really is quite a nice crate right now.

and while in time the strange fan o’war will be the strange splendid screen of this crate, the other items are all pretty cool and should hold value in line with the cost of the crate itself for quite a while (but do keep in mind this is really early days still and most prices will fall as more #60s drop)

as an aside, i guess i shouldn’t be completely surprised, but the value of crate #50 items is already pushing upward rapidly, strange pans are back to 8 keys and the strange watch is also pushing upward - much faster than crate #40 stranges when that crate stopped dropping.  lets face it, the stranges in #50 are some of the most interesting (PDA, invisibility watch) and highly sought (pan/bacon) in the game.

* * *

i’m currently in the middle of a love/hate relationship with the chemistry sets/strangifiers.  on the one hand, i really like the idea of strange miscs and the chemistry sets are very clever.  as discussed last week, we like to keep score, and i love numbers and stats and there’s something very satisfying about knowing that the number reflects all your efforts, not just kills.

on the other hand, there are currently very few items which can be strangified which means that if you want a strange misc item, then your cosmetic loadout variations have shrunk by 25%.  now surely Valve will release more chemistry sets and more and more items will be strangifiable in the future but for now, after a barrage of items this summer that allowed everyone to customize their alter-egos like never before, the strange vice has just taken away a misc slot…   so don’t be surprised if many demomen become a variant of this… (this is my demo as imaged in SFM by Essoplerck):


in addition, trading sites are having a very difficult time dealing with strangifiers and chemsets because the standard steam API doesn’t support revealing the formula or the final result.  chemistry sets formulas have a pattern but are generated randomly.  the rare variety uses hats and the more common uses a bunch of normal weapons and one strange one, chosen at random. this actually has caused quite a bit of confusion because the strange portion of the formula being random means a set requiring a strange bottle is just as likely as one requiring a strange kritzkrieg…


the confusion turned into outright anguish for players who rushed to complete the formulas and not realizing that they were random.  i feel terrible for the guy who turned his strange festive bat into a strange all-father but…


(clearly the chemistry sets with the cheap stranges as ingredients are more valuable than those with the expensive ones.  and be weary when trading these because the bait/switch is very strong in these.  always check the formula after both sides have agreed to the trade and before you actually commit.)

but the biggest problem i have with all this is:

Valve has given us very little this year in terms of game play

sure, we got some weapon fixes, some map fixes, and 2 “new” maps (which have simply been made official as they’ve been around, including in competitive tf2, for quite a while)

everything else we’ve gotten this year has been cosmetic - to satisfy our vanity, not to refresh the game itself.  from the robotic boogaloo to the summer coolers to rome vision in mvm, everything has been eye and ego candy.


we’ll have to see what happens at halloween, but my guess is more unusual effects and other cosmetics.  some noise makers, perhaps.  some weird form of metal to craft a purple weapon that’s just a skin for an existing item?  and at christmas time?  more festive weapons, with a small percentage of them being strange.

and definitely more crates, definitely more ways to open them.  kching…

now of course i realize this is partly because the game is so well balanced (shutup mini-sentry haters) and still so fun that there’s not much Valve could do.

however a new tour for mvm would be more than welcome as would entirely new weapons and maybe a new game mode.

one can dream.

oh ya, i suppose i should open a crate.  i really would rather not open the damn #71 (and i don’t have or want to buy a #60).  i like the idea of having even a small chance of making a profit from time to time.  so it’s #28 again this week and that means…

[creaking sound]


alright, it’s not great.  2 items from #71 could have beaten this price wise, but i don’t mind.  it ain’t a rake (or robot spies killed) and someone at tf2r is gonna like winning it…

tally after 29 weeks:

  • spend: 164.77 refined
  • return: 77.77 refined
  • return/spend ratio: 0.472
Sep 2 '13

week 28

probability of unboxing an unusual by now (as it was in week 1)?  ~17.08% (probability today?  0.66%)

so last week i asked “valve, where’s my hat?” ‘cause the TF2 birthday (august 24) came and went. well the birthday bomb dropped on the 27th instead.

and i got my hat


and with the birthday valve defecated the next great big tf2 update which really can be summarized as…


honestly, when the chemistry set (and strangifier) were first discussed after on their appearance in the item schema butbefore any of them started to drop for actual players, the idea of strange hats/miscs made me laugh.

but i was wrong.  ladies and gentlemen, we’ve got strange hats!


before we talk about strange hats tho, lets very quickly review the other aspects of the update.  salvage #50 and crate #56 were removed from the drop system.  their lifespan: 168 days.  crate #71 replaced #56, and the new special reserve (#60) replaced the #50.

and finally all the items from the summer coolers became craftable causing a frenzy of crafting.  8 coolers, each with 8 items, and 100 so called “low craft #s” means 6400 low craft numbers getting created in the first few hours after the update.

i took 3 refined and ran the fabricate headgear recipe just for fun and was amazed (much more than i should have been given the 6400 low crate #s available) when i crafted the #88 brawling buccaneer.


in retrospect, i should have probably tried more crafts.  i certainly had the metal and there really was never a time when so many low craft opportunities entered the economy at the same time.

but i did the wise adult thing and walked away.

i’ll probably revisit the whole low craft # thing someday when i talk about collectors because it’s worth talking about more in depth.

i’m also not going to analyze crate #71 or crate #60 this week because it’s still too early, prices are nowhere near defined enough to make a valuable analyses tho i have to say, a strange loose cannon is #1 on my strange wish list now and the strange heatmaker (with robots killed) will be very popular with MVM players.

also strange part: taunt kills.  (video and taunts by my mvm and scrim buddy Sensei Dave)

* * *

instead, lets talk about this whole strange thing.  first an extremely apropos quote from a steam conversation i had long ago (and saved because of its wisdom) with a steam friend named Veeblebrox:

I find that when you provide a metric, I play to that.
It’s a crippling vice.

stranges were introduced during the summer of 2011 and the first three crates containing stranges were the #19, #20, and #21.

the idea is very simple: these weapons keep track of the number of kills (later other things, such as assists) the player gets while using them.  they “level up” - a concept familiar to all gamers although leveling up simply gives them a funny name, no special powers.

in march of 2012 an expansion of strange weapons was introduced: strange parts.  strange parts arrived in crates #41 and #42 and could be attached to a strange weapon and count additional stats such as buildings destroyed, number of kills that resulted from a headshot, number of players who were exploded into little chunks of flesh that make mothers shudder and scream about how violent video games are these days, etc.



* * *

personally, i’m addicted to strange weapons.

like Veeblebrox says it’s a vice of mine, these numbers.  i understand where the value of strange weapons comes from and it’s important to realize that the attraction of strange weapons is really just another manifestation of vanity.

be honest to yourself.  you like strange weapons and strange parts and now strange hats because they let you shout to the world : i’m no one to be trifled with!  or something.

and we love to keep score.

even if you hate math or statistics you have to admit, you love to keep score. whether it’s sport, video games, academic tests, there are few humans who are humble enough to truly not care.  even the word itself “score” has made into our language as a slang term used to describe events or goals such as making a deal, taking revenge, or getting laid.

we are at our most primitive levels competitive creatures.  in fact, we would not exist without competition because competition drove the creation of complex life.


aside from the incredible “on the origin of species”, charles darwin was also the first to write about the natural world of plants and animals as that of constant and violent struggle.  in fact he was very deeply disturbed the deeper he delved into his ideas and formulated his theory of evolution.  the victorian view of animals was almost religious, an ordered ladder of consumer and food with the human at the top.  the actual violence and suffering of animals eating and being eaten was conveniently avoided.

but even if you don’t subscribe to the theory of evolution (and so live in kansas) you’d have to be completely blind to not see that competition is actually part of *everything* we do in our lives.

we compete for sexual partners, business deals, and the best seat on the bus. we compete on football fields, racetracks, and balance beams. we compete in academics, board games, and to get the attention of the bartender to bring us another drink. we compete to be heard, seen, recognized. we compete against other humans and sometimes just against ourselves. even when there is no prize, no real winner, we compete.

remember vanity?  the peacock’s feathers actually make it easier to catch and eat the damn bird, but this is more than made up for by the fact that the bird with the best tail gets laid allowing its genes to propagate…

* * *

speaking of birds, i got myself a chemistry set and crafted a strangifier for my birdman of abardeen and now the bird is always with me.


because i’m one vain motherfucker and valve has masterfully combined the peacock’s feathers and the scoreboard into one type of item.

* * *

i know i should be opening a crate #71 this week, but honestly my data shows me that it’s not even in the top 10 normal crates after just 5 days.   to summarize very quickly, the Chance of Sadness is already up to 49.5%, there’s no chance of profit, and the best item inside, the gib kills part, was only worth a lot because it was previously available only in crate #41 which stopped dropping in may of 2012. the market price of gib kills has been sinking by the hour since #71 started dropping.  it’ll probably fall to somewhere in the 2 ref range at best.

meanwhile crate #28 continues to astound.  this week it’s just a 9% chance of sadness and the shotgun continues to command a great price.

it’s also week 28 :P

[creaking sound]


hmm, well it’s sort of like a shotgun…  and i actually love this weapon. coupled with everyone’s beloved valve-nerf-this-now-weapon, the mini-sentry, it’s a fantastic way to … “settle the score”

oh and speaking of score…

tally after 28 weeks:

  • spend: 158.55 refined
  • return: 75.77 refined
  • return/spend ratio: 0.478
Aug 26 '13

week 27

probability of unboxing an unusual by now (as it was in week 1)?  ~16.52% (probability today?  0.66%)

this week i want to expand on week 26’s entry and discuss some interesting psychological phenomenon and their relevance to tf2.


ever say a word over and over and over until it loses its meaning?

this phenomenon is called Semantic Satiation and is the subject of some debate and interesting psychological studies.  several theories exist as to why our brains do this, but the most likely is that it has to do with how we are able to separate signal from noise.

one way to experience a similar phenomenon is to be in a room with a consistent sound such as an electric fan.  after some time passes, you are no longer aware of the fan noise until someone suddenly, without warning, turns the sound off.  suddenly you notice that *lack of noise*.

your brain has noticed a change in your environment which, when we were still monkeys, promoted an evolutionary strategy for caution…


this doesn’t just relate to sound or language either, it can be visual as well. the brain is incredibly well suited to filtering out the background noise from any of our senses to allow us to keep attention and focus in complex environments.

a recent Dartmouth/Davis study looked at this phenomenon at the synaptic level by monitoring neuron activity while subjects were exposed to various levels of visual stimuli.  their findings expose low level mechanisms core to the way our brains work to filter out unnecessary information in order to keep focus.

how the fuck does this relate to TF2?

i first noticed this after trading for my first unusual, an orbiting fire sober stuntman.


i played a ton of MVM at the time and on the bigrock missions realized how easy it was to spot incoming spies disguised as me.  i took this knowledge to a dustbowl server and again this proved remarkably useful.  the fire stood out from across the map and, because spies very often pick demo as the disguise (mostly because it’s the laziest way to disguise, hitting 4 and 4 on the keyboard, but also because it’s the one class not constantly doing something that a disguised spy wouldn’t do while maintaining a good rate of movement) i was able to pick that “signal” out over the “noise” and warn my team early.

if you play a class likely to be used as a disguise (pyro, demo) consider wearing some sort of obvious piece of kit, maybe paint your hair something that stands out well.  your brain will pick that out easily over the otherwise drab blues, reds, and grays if a spy choses to disguise as you.

* * *

avoiding satiation can also an important factor in trading.  if you are selling a common item, such as a craft hat, or an item that suddenly no one wants, it is difficult to get your trade noticed.  you might try a low price but then you’re either taking a loss or at best not making a profit.

the other way is to make the trade interesting to make it stand out.  include the hat as part of set.  add the hat to a trade that has other, more unique items even if it’s not part of a set.

in online advertising, one of the most interesting and frustrating (to the advertiser) phenomenon is the idea of Banner Blindness.  even if you don’t use browser tools to filter out ads, your brain will begin to do it for you.

advertisers have developed many methods to combat this, such as unusual locations or sizes of ads, or frequency capping the ads so that a single user does not see it over and over causing it to disappear into the background noise.

this absolutely applies to trading in tf2 as well.

so deal with items that always stand out.  buy and sell painted cosmetics, rare items, level sets, class sets, seasonal stuff (start early), and my current personal favourite, stranges with interesting parts.

also don’t post too many trades at the same time, be patient, mix things up (i like to throw a auction into the lot, but never more than one at a time)

avoid buyer burnout through variety.

* * *

so now that it’s time to pop a crate open lets mix things up again.  i looked at my data and noticed a sleeper crate, #24.  you can pick one up for just a scrap and yet its average yield is ~3 refined and it has a very low Chance of Sadness.  in fact right now it’s the 2nd lowest (brown cooler is *still* the best). at just 27% CoS and even without any chance of profit, it looks like an apropos one to open.

[creaking sound]


fantastic - the 2nd best item in the crate.

tally after 27 weeks:

  • spend: 152.33 refined
  • return: 73.77 refined
  • return/spend ratio: 0.484

congrats to Epsilon eSports on winning an incredibly entertaining grand final at i49.

oh and happy birthday TF2 and what happened to my party hat and the noisemaker, Valve???  :(

Aug 19 '13

week 26

probability of unboxing an unusual by now (as it was in week 1)?  ~15.96% (probability today?  0.66%)

i have been traveling a ton this summer (hence why this post is late yet again) but the travels are almost over and soon i will settle back into the mediocrity of my “regular” life.  my summers go this way pretty much every year and at the end there is a mix of relief and a sort of sadness.  life can get pretty boring.

we humans come in many varieties and it is interesting that what is exciting to one person may seem silly to another and that even the very pursuit of adventure is a foreign idea to so many.


with the advent of computers and true interactive entertainment, so many of us who might otherwise seek out real world adventures have chosen to live out fantasies in a virtual world.

* * *

i spoke with an absolutely intriguing gentleman this weekend who has been representing world class athletes for almost 4 decades.  he said the most notable thing with young athletes today is that they come from a world of video games (he spoke broken english and made a lot of gestures to get that point across, it was very entertaining).


i asked if it gave them a sense of invulnerability but he said no, that has always been the same in those who practice sport and face physical challenges at their highest levels.  he said what is different is that they no longer return to “normality” - that when they are done competing, when they are done training, they don’t shut down anymore, they continue their pursuit of thrill in the virtual world.

he believes it gives them an edge - they are sharper, smarter, and with unbelievable reflexes.

but i think for the rest of us our lives consist of long periods of mediocrity punctuated by short moments of excitement, much of it virtual.  in america, especially, we are very much shackled to lives with short periods of “time off” (be it from work, school, etc.) and we measure our successes in areas that require a lot of sit down time and the height of excitement is closing that deal or water cooler gossip.


so it is no surprise at all that the video gaming industry has found such a tremendous audience in the modern world.  we have created and inhabit worlds where we can be something we could never be in real life, where even the laws of physics don’t have to apply.  we can have our alter-egos.


but even inside our virtual realities, the mediocrity eventually creeps in. games come and go, what was once considered ground breaking becomes common place, and most importantly, the hats start to get old.

and this is where Valve’s genius lies.  they are certainly not the first to introduce a virtual economy to a video game but they have always been excellent at involving the community.  hell, their two most successful franchises, Counter Strike and Team Fortress, came from the community.  and even if the Robotic Boogaloo was itself a failure, there’s now a proven system to quickly bring community created items into the game.  the summer coolers are filled with all sorts of great stuff and, along with much needed tweaks to the game itself, have definitely brought back the interest and excitement.  and this is just the beginning.

yes, there’s definitely a downside to the firehose of items.  prices are flattening out on many items - the bill’s hat, for example, is starting to tank for the first time in years.

this may suck for pure traders, but for the rest, it’s all a good thing.  your alter ego never had this much variety and you don’t have to be rich virtually (or in real life) to pimp yourself out.

so this week fuck the stats, i’m gonna open a crate i haven’t opened before - for variety’s sake.  lets make it the green cooler:

[creeking sound]


see now this is perfect.  something i never even knew existed!  also, i immediately gave it away to a friend for their birthday (sorry rafflers!)

tally after 26 weeks:

  • spend: 146.44 refined
  • return: 70.44 refined
  • return/spend ratio: 0.481
Aug 12 '13

week 25

probability of unboxing an unusual by now (as it was in week 1)?  ~15.40% (probability today?  0.66%)

when i first announced the beginning of this project, i was surprised how many people, friends and strangers alike, were almost annoyed with my determination to open just one crate per week.

"go big or go home" 

said one

"how about one crate per day"

said another

even those who understood my motives would suggest stuff like

"maybe once a month you should open 5 instead of 1".

(in addition, almost invariably each entry has at least one comment that says “tl;dr” or “i just scroll down to the bottom to see what you got”.)

there are several reasons why i chose to open one crate per week.

  • a week gives me plenty of opportunities to trade for a profit of 1 key - i do not wish to buy keys for RL$ to keep going and so far i haven’t had to
  • it takes a long time to prepare each article.  besides writing a bunch of text i have conversations with friends, do research, write software to crunch numbers or test theories, etc.
  • this is a discipline for me and an exercise in Patience

Patience is a form of endurance, sometimes in difficult circumstances.

despite the play on words in my steam nickname, i’m not a Buddhist (i’m an ardent atheist if you must know), but i do find Buddhism a fascinating religion.

also Buddha is one of the best manga comic series i’ve ever read.


in Buddhism, Patience, or क्षान्ति (kSanti), extends to forgiveness and is one of the 10 practices of perfection.  it is also a critical component of meditation.

meanwhile in western scientific study, Patience is a form of self control and is considered a major component of decision making.

and in TF2, as my friend Strider demonstrated during the first week of the summer cooler madness, impatient traders make excellent targets if you’re looking to make a profit…


the reason i bring all of this up is because the past two weeks i’ve been talking about disappointment that can come from unboxing a single crate.  as i’ve continued to work on my analyzer scripts i realized that there are two very distinct ways to approach unboxing and that mine, one at a time, has somewhat different parameters than for someone who unboxes in bulk.

for people who open just a few crates at a time, it’s important to avoid crates that can disappoint.  but for those that unbox in bulk, the most important thing is to come out with the least profit loss or the best average yield.

the trick is figuring out what exactly is a bulk unboxing.  25 crates?  50?  100?

want to be nearly assured of getting an unusual in one sitting?  i suggest opening 689 crates (probability of 99%)

but there’s actually a more scientific approach to finding out when an unboxing becomes “bulk”.

given the current price and percentage for a given item contained within, one can calculate (as i did way back in week 2) the average likely yield.

looking at crate #42 again (using today’s prices via we find the following:

  • 19.8% chance of Strange Bottle worth 0.27r
  • 19.8% chance of Strange Backscratcher worth 0.17r
  • 19.8% chance of Strange Pistol worth 1.44r
  • 9.9% chance of Strange Part: Buildings Destroyed worth 9.24r
  • 9.9% chance of Strange Part: Projectiles Deflected worth 6.93r
  • 9.9% chance of Strange Part: Headshot Kills worth 15.02r
  • 4.95% chance of Outdoorsman worth 1.22r
  • 4.95% chance of Ol’ Geezer worth 1.22r

take 19.8% * 0.27 + 19.8% * 0.17r * 19.8% * 1.44r + etc.

and you end up with an average yield of 3.62 refined

this is actually not so bad, in fact, i ran a script comparing all active crates in TF2 as of this writing, and crate #42 has the 7th best average yield.  (for those curious, the top 10 are #68 (Brown Cooler), #65 (Green Cooler), #66 (Aqua Cooler), #64 (Yellow Cooler), #28, #67 (Blue Cooler), #42, #20, #24, and #33)

but as discussed 2 weeks ago, crate #42 has a nasty 69.3% chance of getting a crappy item.

so the challenge is, how many unboxings would it take before crate #42’s yield would start averaging out to that 3.62 promise.

there are mathematical ways to calculate this but they are very advanced and math is hard so i chose to write a simulator program instead.

i took two approaches, in the first, i ran my simulator 100 times.  each iteration consisted of 1 more unboxing, i.e. on the first run i simulate unboxing one crate #42.  on the 2nd 2 crate #42s.  on the 69th, 69 crate #42s all the way to 100.

for each iteration i calculated the overall average yield and then plotted these values on a scatter plot.  here’s what that looks like:

it is clear that the overall average yield starts to converge toward a reliable value the more crates you unbox.

however the number of unboxings at which point this becomes a reliable value is difficult to see.

so in a second approach, i decided to run the same simulator but this time measure the variance of each iteration from the expected 3.62 refined.  once this value stabilized i stopped the simulator and returned the number of unboxings it took.  because of the randomness involved i ran the simulator itself 100 times and averaged the results.

for crate #42 i found that average yield becomes reliable at approximately 34 unboxings.

each crate series has a different value.  some crates, such as the infamous robocrate, converge almost immediately (because all the items within suck almost equally) while others, such as #4 take 60 unboxings to stabilize (that one goes all the way from a normal weapon to white paint)

but if you take all of these values together, one for each crate series, and average them out, you end up around…

25 unboxings.

so in theory, if you open less than 25 crates at a sitting, you should be looking at the Chance of Sadness and Degree of Rage. but if you open more, concentrate more on crates that have the best Average Yield.

* * *

funny enough tho, one could say that i have now, this being week 25, reached a point of bulk unboxing…


but as conan the barbarian was fond of saying.  ”enough talk”.  this week, i open the #28 ‘cause i’m still only opening one crate folks and i really would like to avoid being disappointed.

[creaking sound]


sweet.  that’s the best item in the crate and an actual profit.

tally after 25 weeks:

  • spend: 140.55 refined
  • return: 67.11 refined
  • return/spend ratio: 0.477