TF2: Nobody's Home

00:59:31
https://www.youtube.com/watch?v=2stmQfv93oQ

概要

TLDRThe video explores the state of Team Fortress 2 (TF2), emphasizing its continued popularity despite being 16 years old. It reveals that the reported player counts are misleading due to the presence of bots, which inflate the numbers. The video categorizes bots into three types: cheaters, traders, and idlers, explaining how they contribute to the player count without genuine human activity. Data analysis shows discrepancies between Steam's reported player counts and actual human activity, suggesting that a significant portion of the player base consists of bots. The video concludes with a call to action for the community to address the issues surrounding botting and the neglect from Valve, urging players to hold the company accountable for the game's state.

収穫

  • 🎮 TF2 is 16 years old but still popular.
  • 📊 Reported player counts are inflated by bots.
  • 🤖 Three types of bots: cheaters, traders, idlers.
  • 📉 Actual human player count is around 25,000-30,000.
  • 🔍 Data analysis shows discrepancies in player counts.
  • 🚨 Valve has neglected the bot issue for years.
  • 💬 Community needs to speak out against bots.
  • 💰 Players should stop spending money until changes are made.
  • 📈 Teamwork.tf provides more accurate player data.
  • ⚠️ The game's state is worse than it appears.

タイムライン

  • 00:00:00 - 00:05:00

    Team Fortress 2 (TF2) has reached 16 years of age and remains one of the oldest supported video games, consistently ranking among the top 10 most played games on Steam. Despite its age, it appears to have a thriving player base, but there are underlying issues that suggest otherwise.

  • 00:05:00 - 00:10:00

    The apparent player count reported by Steam is misleading. There is a significant discrepancy between the number of players in-game and the reported figures, with many servers being filled by cheating bots rather than actual players, indicating a neglect of the game.

  • 00:10:00 - 00:15:00

    Bots in TF2 come in three main types: cheaters, traders, and idlers. Cheater bots disrupt gameplay, while trading bots facilitate item exchanges. Idler bots collect timed item drops, contributing to the inflated player count reported by Steam.

  • 00:15:00 - 00:20:00

    Idler bots are particularly problematic, as they exploit the item drop system, leading to tens of thousands of bots running at any given time. This artificial inflation of player numbers misrepresents the actual player base of TF2.

  • 00:20:00 - 00:25:00

    To analyze player activity accurately, high-resolution player data is needed. Steam's data is often averaged and lacks detail, making it difficult to assess true player engagement. A bot was set up to collect TF2's player counts directly from Steam's API for analysis.

  • 00:25:00 - 00:30:00

    Observations of player data reveal that TF2's player counts do not exhibit typical human behavior, such as daily fluctuations. Instead, the data shows flat cycles with random spikes and drops, indicating a high volume of bot activity.

  • 00:30:00 - 00:35:00

    The lack of normal player behavior suggests that the player counts are heavily influenced by bots. When server outages occur, many accounts go offline and do not return, further supporting the idea of bot networks operating continuously.

  • 00:35:00 - 00:40:00

    Evidence suggests that bot networks are likely responsible for the inflated player counts, as they can be turned on and off en masse. This behavior aligns with the patterns observed in other heavily botted games.

  • 00:40:00 - 00:45:00

    The analysis of player data from teamwork.tf, which tracks players inside game servers, shows significantly lower player counts than Steam, reinforcing the idea that a large portion of reported players are bots rather than humans.

  • 00:45:00 - 00:50:00

    The true player count for TF2 is estimated to be around 25,000 to 30,000 humans on any given day, while Steam's reported numbers are inflated by bots, with estimates suggesting that 70-80% of the player count may be non-human.

  • 00:50:00 - 00:59:31

    The video concludes that TF2 is facing severe neglect from Valve, leading to a bot-infested environment that detracts from the player experience. The community is encouraged to voice their concerns and demand better support for the game.

もっと見る

マインドマップ

ビデオQ&A

  • What is the main issue with Team Fortress 2's player count?

    The player count is inflated by bots, making it seem like there are more players than there actually are.

  • What types of bots are present in TF2?

    There are three main types of bots: cheaters, traders, and idlers.

  • How does bot activity affect the gameplay experience?

    Bots can disrupt gameplay, especially cheaters, while idlers and traders inflate player counts without contributing to genuine play.

  • What evidence is presented to support the claim of inflated player counts?

    Data analysis shows discrepancies between Steam's reported player counts and actual human activity, indicating a high volume of bots.

  • What should the community do about the bot issue?

    The community should speak out against the bot problem and hold Valve accountable for addressing it.

  • How many real human players are estimated to be playing TF2?

    The estimated number of real human players is around 25,000 to 30,000 on any given day.

  • What has Valve's response been to the bot problem?

    Valve has shown little interest in addressing the bot issue, despite the game's ongoing profitability.

  • What is the significance of the data collected from teamwork.tf?

    Teamwork.tf provides a more accurate representation of player counts by excluding idle and trade bots.

  • How does the community perceive the state of TF2?

    There is a disconnect between the community's positive discourse and the actual state of the game, which is plagued by bots.

  • What is the call to action for TF2 players?

    Players are encouraged to stop spending money on the game until Valve addresses the bot problem.

ビデオをもっと見る

AIを活用したYouTubeの無料動画要約に即アクセス！
字幕
en
オートスクロール:
  • 00:00:01
    [Music]
  • 00:00:06
    recently reaching 16 years of age Team
  • 00:00:09
    Fortress 2 is one of the oldest actively
  • 00:00:12
    supported video games on the market
  • 00:00:14
    despite being old enough to legally
  • 00:00:15
    drive players still come out in droves
  • 00:00:17
    to play TF2 every day frequently placing
  • 00:00:20
    it in the top 10 most played games on
  • 00:00:23
    Steam fans often site this placement as
  • 00:00:25
    evidence of the game's continued growth
  • 00:00:27
    and popularity which is an exceptional
  • 00:00:29
    feed in face of its age thanks to this
  • 00:00:32
    especially to someone unfamiliar with
  • 00:00:34
    TF2 everything seems fine on the surface
  • 00:00:38
    this relic of a game still receives
  • 00:00:40
    updates has active developer support and
  • 00:00:43
    has tens of thousands of players every
  • 00:00:45
    single day really seems like this game
  • 00:00:48
    is destined to live
  • 00:00:52
    [Music]
  • 00:00:53
    forever yeah I don't buy that nope look
  • 00:00:56
    at the number of players reported by
  • 00:00:58
    Steam intuitively if you tally up every
  • 00:01:01
    player inside of Game servers that
  • 00:01:03
    figure should be pretty close to what's
  • 00:01:05
    reported by
  • 00:01:07
    steam uh well all right that's a massive
  • 00:01:10
    discrepancy you know what ignore that Q
  • 00:01:13
    for a game you'll often find that the
  • 00:01:15
    game coordinator struggles to fill up an
  • 00:01:17
    entire server even on popular Maps
  • 00:01:19
    during peak hours despite tens of
  • 00:01:22
    thousands of people apparently playing
  • 00:01:24
    the game and more often than not these
  • 00:01:26
    servers are filled by cheating Bots this
  • 00:01:29
    doesn't add up where are all the players
  • 00:01:32
    for years players myself included have
  • 00:01:34
    taken tf2's player count at face value
  • 00:01:38
    as true but this number is not real Team
  • 00:01:43
    Fortress 2 faces abhorent neglect which
  • 00:01:47
    has left it wide open for abuse and
  • 00:01:49
    exploitation the magnitude of which very
  • 00:01:52
    few people understand
  • 00:02:03
    until now
  • 00:02:09
    [Music]
  • 00:02:31
    [Music]
  • 00:02:46
    Bots or non-human activity have been a
  • 00:02:49
    part of Team Fortress 2's ecosystem for
  • 00:02:51
    many years Bots come in various forms
  • 00:02:54
    but most of them fall into three basic
  • 00:02:56
    categories we have cheaters Traders and
  • 00:02:59
    Idol ERS the type of Bot that I'm sure
  • 00:03:01
    most of you are unfortunately familiar
  • 00:03:04
    with are cheaters these are non-human
  • 00:03:06
    accounts that are set up to well cheat
  • 00:03:09
    they've plagued valve's official game
  • 00:03:11
    servers for many years now making the
  • 00:03:13
    game unplayable more often than not yet
  • 00:03:16
    despite their annoying presence there's
  • 00:03:18
    not that many of them Bots that actually
  • 00:03:20
    play the game require a lot of resources
  • 00:03:23
    like computer memory Argo a single
  • 00:03:25
    person can't host that many cheater Bots
  • 00:03:27
    without some serious investment most bot
  • 00:03:30
    hosters operate less than 100 at an
  • 00:03:32
    individual level typically on their own
  • 00:03:34
    Hardware or by using free trials of
  • 00:03:37
    cloud services like Amazon for example
  • 00:03:39
    if these self-reported figures are to be
  • 00:03:41
    trusted the total number of cheater bots
  • 00:03:44
    in TF2 likely numbers around a thousand
  • 00:03:47
    at any given time likely slightly less
  • 00:03:50
    than that of course this figure does
  • 00:03:52
    vary throughout the year because bot
  • 00:03:54
    hosters tend to fire up more Bots
  • 00:03:56
    specifically to disrupt new
  • 00:03:58
    updates the second type of Bot that I'm
  • 00:04:00
    sure many of you are also familiar with
  • 00:04:03
    are Traders these are automated accounts
  • 00:04:05
    that handle trade requests for
  • 00:04:06
    thirdparty sites or buy sell and trade
  • 00:04:09
    items for individual players that lease
  • 00:04:11
    them I'll go ahead and lump gambling
  • 00:04:13
    site Bots into this group as well but
  • 00:04:15
    they're definitely a minority trading
  • 00:04:17
    Bots aren't required to be in game to
  • 00:04:19
    trade but quite a few of them idle in
  • 00:04:21
    game for various reasons aesthetic
  • 00:04:23
    purposes for example perhaps this
  • 00:04:25
    entices someone to trade with a bot more
  • 00:04:28
    often I'm not entirely sure
  • 00:04:30
    unsurprisingly the total number of
  • 00:04:32
    trading Bots outnumbers cheater Bots
  • 00:04:35
    reaching well into the thousands where
  • 00:04:37
    there's money to be made people will
  • 00:04:39
    come trading Bots don't play the game
  • 00:04:42
    they simply move items around requiring
  • 00:04:44
    far fewer resources than cheater Bots to
  • 00:04:46
    operate so we should expect more of
  • 00:04:49
    these out in the wild however only a
  • 00:04:51
    fraction of trading Bots have the game
  • 00:04:53
    running likely around 3 to 500 or so
  • 00:04:56
    Bots from larger trading websites don't
  • 00:04:58
    idle typ TYC Al only those operated by
  • 00:05:01
    individual players or rental
  • 00:05:04
    services and that brings us to our final
  • 00:05:07
    type of Bot the one the least of you are
  • 00:05:09
    familiar with idlers these are accounts
  • 00:05:12
    that are set up to collect tf2's timed
  • 00:05:14
    item drops which can then be sold for a
  • 00:05:16
    profit in the past valve didn't take
  • 00:05:19
    kindly to this and took action against
  • 00:05:21
    idlers they limited how many items are
  • 00:05:23
    dropped per week and made it so players
  • 00:05:25
    had to be inside of Game servers in
  • 00:05:27
    order to receive them these were decent
  • 00:05:29
    stop gaps but not the best deterrence in
  • 00:05:31
    the long run because idlers still
  • 00:05:33
    operate to this day given the fact that
  • 00:05:36
    TF2 and steam are both free it's very
  • 00:05:40
    easy to exploit the item drop system in
  • 00:05:42
    total the number of idle Bots reaches
  • 00:05:45
    into the tens of thousands Yes you heard
  • 00:05:48
    me right tens of thousands and if you
  • 00:05:52
    lump all types of bots together a vast
  • 00:05:55
    majority of them have the game running
  • 00:05:57
    this means that they contribute toward
  • 00:05:59
    towards the total player count as
  • 00:06:01
    reported by steam this artificially
  • 00:06:03
    inflates the player numbers making Team
  • 00:06:05
    Fortress 2 look far more populated than
  • 00:06:08
    it really is Team Fortress 2 doesn't
  • 00:06:11
    really have 60 to 100,000 players
  • 00:06:13
    throughout the year in fact it's far far
  • 00:06:17
    less and I'll prove it to you
  • 00:06:25
    [Music]
  • 00:06:32
    starting at square one if you want to
  • 00:06:34
    see Bing activity you need high
  • 00:06:37
    resolution player data the unfortunate
  • 00:06:39
    thing about Steam and therefore websites
  • 00:06:42
    like steam DB is that they don't
  • 00:06:44
    preserve high resolution historic player
  • 00:06:46
    counts player counts are sampled every
  • 00:06:48
    10 minutes which is perfect but over
  • 00:06:50
    time steam gets rid of data in order to
  • 00:06:52
    save space the data get gradually
  • 00:06:55
    averaged to hourly and then daily and so
  • 00:06:58
    on and so forth which removes a lot of
  • 00:07:00
    information so I had to make a record
  • 00:07:03
    myself with a lot of help of course Very
  • 00:07:06
    graciously Mega scatter bomb volunteered
  • 00:07:08
    his time and set up a bot for me that
  • 00:07:11
    collects tf2's player counts directly
  • 00:07:13
    from steam's API we've been sampling
  • 00:07:15
    tf2's player count every 10 minutes
  • 00:07:18
    since mid December of last year and
  • 00:07:20
    that's the data that I'll be using for
  • 00:07:22
    my analysis I should note now that steam
  • 00:07:25
    reports an instant snapshot of players
  • 00:07:28
    in game not some weird average value
  • 00:07:30
    over time which is a common
  • 00:07:32
    misconception the number you see is the
  • 00:07:35
    actual number of players in game well or
  • 00:07:37
    so I've long since thought so to kick
  • 00:07:40
    things off I can sail this ship most of
  • 00:07:42
    the way home with some very basic
  • 00:07:45
    observations it doesn't take a rocket
  • 00:07:47
    science to prove that TF2 is heavily
  • 00:07:49
    boded just a pair of eyeballs and some
  • 00:07:51
    common sense right now everyone watching
  • 00:07:54
    this video what I want you to do go to
  • 00:07:57
    steam DB and pick any game on Steam and
  • 00:08:00
    what you're going to see are player
  • 00:08:02
    counts that look like this here's some
  • 00:08:04
    player data from rust Apex Legends and
  • 00:08:08
    Counterstrike what you're going to
  • 00:08:10
    notice are very clear cyclic player
  • 00:08:13
    Peaks and lows these Cycles happen on a
  • 00:08:16
    daily basis separated by about 24 hours
  • 00:08:19
    intuitively player counts fluctuate like
  • 00:08:22
    this because of a natural human need
  • 00:08:25
    sleep most humans sleep most people turn
  • 00:08:28
    off their games and their computers
  • 00:08:30
    before they go to bed of course there
  • 00:08:32
    exists a nonzero number of humans that
  • 00:08:35
    leave their games running all the time
  • 00:08:36
    or at odd hours but they definitely
  • 00:08:39
    represent a negligible minority this is
  • 00:08:42
    the dominant expected Behavior inside of
  • 00:08:45
    player data one should expect normal
  • 00:08:48
    human activity and Bots they don't need
  • 00:08:51
    sleep they run all the time or with
  • 00:08:53
    behavior that doesn't reflect human
  • 00:08:55
    activity so if a game's player data
  • 00:08:57
    doesn't fluctuate like this something is
  • 00:09:00
    a miss or the game is truly Dead one
  • 00:09:03
    could even consider these daily Cycles
  • 00:09:05
    like a video game's heartbeat its pulse
  • 00:09:08
    the stronger the rise and fall the
  • 00:09:10
    stronger the heart therefore daily
  • 00:09:12
    player fluctuations are a far better
  • 00:09:14
    indicator of real human activity in a
  • 00:09:17
    video game rather than the peak player
  • 00:09:19
    count on any given day now with that
  • 00:09:22
    basic but very important observation
  • 00:09:24
    established let's look at tf2's player
  • 00:09:27
    data do we see human behavor avor here
  • 00:09:30
    the short answer is no not really I'll
  • 00:09:33
    start by showing the entire data record
  • 00:09:36
    since mid December this is pretty hard
  • 00:09:38
    to read and a lot to digest so let's
  • 00:09:41
    zoom in to pretty much any given week
  • 00:09:43
    like the other games from a qualitative
  • 00:09:45
    standpoint it looks like tf2 has very
  • 00:09:48
    little human activity its daily Cycles
  • 00:09:51
    are very flat compared to other video
  • 00:09:53
    games while player counts remain
  • 00:09:55
    consistently High the data are also very
  • 00:09:58
    noisy it's riddle with random spikes and
  • 00:10:00
    drops ranging from a few thousand to
  • 00:10:03
    often times tens of thousands of players
  • 00:10:06
    they're very rapid as well occurring
  • 00:10:08
    over the course of a couple of hours or
  • 00:10:10
    less compared with other video games TF2
  • 00:10:13
    has a comparatively weak and AR rhythmic
  • 00:10:15
    heartbeat the entire Player record looks
  • 00:10:18
    like this zooming into any other week's
  • 00:10:20
    worth of data shows the exact same
  • 00:10:22
    behavior what's going on here well you
  • 00:10:24
    know given the focus of this video your
  • 00:10:26
    mind may go to Bots and it is Bots but
  • 00:10:29
    mind my job in this video is to play
  • 00:10:31
    Devil's Advocate so to play the role of
  • 00:10:33
    Devil's Advocate I have a little friend
  • 00:10:35
    here say hello to my she devil well that
  • 00:10:37
    wasn't very nice anyways let's address
  • 00:10:40
    the first observation what would cause
  • 00:10:43
    flattened player Cycles well I'm sure
  • 00:10:45
    this came into some people's minds
  • 00:10:47
    wouldn't flat player Cycles be caused by
  • 00:10:50
    a bunch of accounts that never go
  • 00:10:51
    offline well yes you're correct for
  • 00:10:54
    example let's say you have a game where
  • 00:10:56
    10,000 humans log on and off every
  • 00:10:58
    single day here's what that player data
  • 00:11:00
    would look like then let's add 50,000
  • 00:11:03
    accounts that never drop Offline that
  • 00:11:05
    raises the minimum Baseline of online
  • 00:11:07
    accounts so player Cycles get squished
  • 00:11:10
    now of course I could adjust the scaling
  • 00:11:11
    of my charts to make the cycles look
  • 00:11:13
    normal again but note that steam always
  • 00:11:16
    plots from zero to a little bit above
  • 00:11:18
    the maximum player count so it's not
  • 00:11:20
    chart scaling messing with our eyes here
  • 00:11:22
    and that's where things get suspicious
  • 00:11:25
    my devil's advocate says this is due to
  • 00:11:28
    a high volume of play at any given time
  • 00:11:30
    thanks to tf2's Global popularity there
  • 00:11:33
    is indeed this common belief amongst TF2
  • 00:11:36
    fans that the game is globally popular
  • 00:11:38
    which isn't entirely incorrect this
  • 00:11:40
    should translate into more people
  • 00:11:42
    playing at any given time because you
  • 00:11:44
    know time zones exist this would indeed
  • 00:11:47
    smear out player activity across 24
  • 00:11:50
    hours while keeping player counts
  • 00:11:51
    consistently high but this argument is
  • 00:11:54
    debunked immediately given that about
  • 00:11:57
    80% of player activity is in the Western
  • 00:12:00
    Hemisphere and in Europe given this we
  • 00:12:03
    should expect for the vast majority of
  • 00:12:05
    player activity to occur within a much
  • 00:12:07
    shorter time frame in other words taller
  • 00:12:10
    Peaks and lows not shorter ones this
  • 00:12:12
    isn't me saying that tf2's player count
  • 00:12:14
    should reach zero every day just that
  • 00:12:17
    the fluctuations should be more
  • 00:12:19
    pronounced for example a globally
  • 00:12:21
    popular title would be something like
  • 00:12:22
    Warframe we have much shorter player
  • 00:12:25
    Cycles but consistently High player
  • 00:12:27
    counts but a regionally popular game
  • 00:12:29
    would be something like nakaka blade
  • 00:12:31
    points assuming I pronounced that right
  • 00:12:33
    this game is only popular in a very
  • 00:12:35
    narrow swath of Asia therefore its
  • 00:12:37
    player Cycles are much larger as nearly
  • 00:12:40
    all players log on and off at around the
  • 00:12:42
    same time TF2 should be somewhere in the
  • 00:12:45
    middle of these games but it's not well
  • 00:12:49
    my little she devil says hey well you
  • 00:12:51
    know what a bunch of humans simply leave
  • 00:12:53
    their games running all the time this is
  • 00:12:55
    an argument that I've seen posed many
  • 00:12:57
    times before however there's no
  • 00:13:00
    incentive for the common player to do
  • 00:13:02
    this other than idling which hasn't been
  • 00:13:04
    profitable at the individual level since
  • 00:13:06
    2009 there are other Niche reasons for
  • 00:13:09
    leaving TF2 running continuously but the
  • 00:13:12
    individual humans that do this almost
  • 00:13:14
    certainly represent a tiny minority of
  • 00:13:16
    the player base so the argument of a
  • 00:13:20
    large volume of humans being present all
  • 00:13:22
    the time isn't plausible now for that
  • 00:13:26
    second weird behavior all of the random
  • 00:13:29
    spacks and dips in the data some of
  • 00:13:31
    these are human yes a few sharp dips
  • 00:13:33
    coincide with server outages as players
  • 00:13:36
    get knocked offline but a key feature
  • 00:13:38
    with outages is that we see a complete
  • 00:13:40
    recovery back to normal player activity
  • 00:13:43
    soon after servers come back online
  • 00:13:45
    here's a few games as an example normal
  • 00:13:47
    player counts resume within a short time
  • 00:13:49
    frame a few hours if not faster and most
  • 00:13:52
    importantly these outages are correlated
  • 00:13:55
    and reflected in other games data at the
  • 00:13:58
    same time of reported issues that's the
  • 00:14:00
    sure fired way of knowing that these
  • 00:14:02
    dips are outage related and not
  • 00:14:04
    something else but what's very strange
  • 00:14:07
    about outage related dips in tf2s data
  • 00:14:10
    is that we often times see an incomplete
  • 00:14:13
    recovery huge chunks of accounts get
  • 00:14:15
    knocked offline and then never return
  • 00:14:18
    January 30th 2024 an outage knocks
  • 00:14:21
    players offline 12,000 accounts that
  • 00:14:24
    were present before the outages never
  • 00:14:27
    return little she devil says you know
  • 00:14:29
    what these could just be discouraged
  • 00:14:31
    players that will come back later but
  • 00:14:32
    even many hours later there is no
  • 00:14:35
    recovery this isn't an uncommon
  • 00:14:37
    occurrence either this happens almost
  • 00:14:39
    every time there's a server outage and
  • 00:14:42
    sometimes these massive dips happen
  • 00:14:44
    without server outages thousands or tens
  • 00:14:46
    of thousands of accounts suddenly vanish
  • 00:14:49
    also outages only address sudden dips in
  • 00:14:53
    players not spikes like this one where
  • 00:14:56
    tens of thousands of accounts suddenly
  • 00:14:58
    come online we we also have very
  • 00:15:00
    frequent smaller spikes like these where
  • 00:15:02
    a few thousand accounts suddenly appear
  • 00:15:04
    and then go offline about 30 minutes
  • 00:15:06
    later or vice versa and yeah sure game
  • 00:15:09
    updates account for some large spikes in
  • 00:15:11
    players but meaningful updates only
  • 00:15:14
    happen three times a year
  • 00:15:16
    inconsequential patches about once a
  • 00:15:18
    month if even that not every week not
  • 00:15:21
    every day so aside from the few
  • 00:15:23
    infrequent server outages and updates
  • 00:15:26
    what causes all of these spikes and dips
  • 00:15:29
    in terms of random spikes and drops
  • 00:15:32
    rapid player fluctuations are totally
  • 00:15:34
    normal and yes they are for all video
  • 00:15:36
    games but the difference is that they're
  • 00:15:38
    not random if I take the hour by hour
  • 00:15:41
    difference in player data that shows me
  • 00:15:44
    the gain or loss in players over the
  • 00:15:46
    past hour let's plot that for
  • 00:15:48
    Counterstrike Apex Legends and DOTA for
  • 00:15:51
    a few examples and as you can see large
  • 00:15:54
    volumes of players log on or off at
  • 00:15:56
    around the same time every day
  • 00:15:59
    additionally they're consistently sized
  • 00:16:01
    volumes this behavior is true for all
  • 00:16:04
    video games and what you're seeing are
  • 00:16:06
    those daily Cycles again and this makes
  • 00:16:08
    sense people get off of work or come
  • 00:16:10
    home from school and eventually go to
  • 00:16:12
    bed at around the same time every day
  • 00:16:14
    but for Team Fortress 2 hour by- hour
  • 00:16:16
    player fluctuations are entirely random
  • 00:16:19
    we see huge volumes of players logging
  • 00:16:22
    on or off at random times and absolutely
  • 00:16:25
    zero consistency in terms of the volumes
  • 00:16:27
    of players coming and going these larger
  • 00:16:30
    spikes can reach into the thousands or
  • 00:16:32
    tens of thousands of accounts reflecting
  • 00:16:34
    5 10 20% or more of the player count at
  • 00:16:38
    the time well all right my little she
  • 00:16:40
    devil now says TF2 simply experiences
  • 00:16:44
    huge chunks of players logging on or off
  • 00:16:46
    at random times it records player
  • 00:16:48
    numbers in some weird way end of story I
  • 00:16:51
    didn't straw man this by the way I've
  • 00:16:53
    seen this argument posed before and I
  • 00:16:55
    have no way of directly debunking this
  • 00:16:58
    but as I've shown this is not normal in
  • 00:17:01
    any video game how on Earth can TF2 be
  • 00:17:05
    the only exception are flattened player
  • 00:17:07
    Cycles human are these dips and spikes
  • 00:17:11
    human no way not a chance in hell this
  • 00:17:14
    is assuredly bot activity but do I have
  • 00:17:17
    proof of this the short answer yeah I do
  • 00:17:35
    one of the more convincing pieces of
  • 00:17:37
    evidence I have is when servers go
  • 00:17:39
    offline let's go back to that example on
  • 00:17:41
    January 30th after nearly every server
  • 00:17:44
    outage we see that incomplete recovery
  • 00:17:47
    where thousands of accounts don't return
  • 00:17:49
    but oh hey look usually about a day or
  • 00:17:52
    two later Boop all of these missing
  • 00:17:54
    accounts suddenly return all at once
  • 00:17:56
    over the course of about an hour or so
  • 00:17:58
    we we have thousands of accounts getting
  • 00:18:00
    turned off and then thousands of
  • 00:18:02
    accounts getting turned back on or
  • 00:18:03
    sometimes vice versa here's an example
  • 00:18:06
    of when there wasn't an outage January
  • 00:18:08
    27th 2024
  • 00:18:11
    22,000 accounts suddenly go offline and
  • 00:18:14
    then 6 hours later they all come back
  • 00:18:16
    online another example January 20th 2024
  • 00:18:21
    25,000 accounts suddenly come online 26
  • 00:18:25
    hours later they all drop offline then
  • 00:18:28
    about a day later they all come online
  • 00:18:30
    again to then all turn off again about 2
  • 00:18:32
    days later what the [ __ ] man and here's
  • 00:18:35
    another
  • 00:18:36
    example and
  • 00:18:39
    another and
  • 00:18:41
    another and
  • 00:18:43
    another this weird Behavior was the
  • 00:18:46
    first thing I noticed when I started
  • 00:18:47
    looking at tf2's player data and when I
  • 00:18:49
    saw this just think about that right
  • 00:18:51
    tens of thousands of bots running all
  • 00:18:53
    the time that get knocked offline when
  • 00:18:55
    servers die and then their owners turn
  • 00:18:57
    them back on when they wake up the
  • 00:18:58
    following day that's what this looks
  • 00:19:00
    like to me but I couldn't just present
  • 00:19:02
    that as fact and then call it a day so I
  • 00:19:05
    got a second opinion I spoke with mega
  • 00:19:07
    scatter bomb someone far more
  • 00:19:09
    well-versed in bing activity than I am
  • 00:19:11
    and after some conversations and showing
  • 00:19:13
    him the data it turns out that these are
  • 00:19:16
    more than likely what are known as bot
  • 00:19:18
    Nets huge networks of bots that seem to
  • 00:19:21
    be running all the time most of these
  • 00:19:23
    dips happen on a Tuesday when Steam
  • 00:19:26
    servers undergo routine maintenance this
  • 00:19:28
    log pretty much everyone's account out
  • 00:19:30
    of steam humans with their clients open
  • 00:19:32
    at an individual level quickly log back
  • 00:19:34
    on but botn Nets they remain offline
  • 00:19:37
    each of the individual bot accounts has
  • 00:19:39
    to log back in so it's a lot faster to
  • 00:19:42
    Simply restart the entire network at a
  • 00:19:44
    set time it seems like bot holters have
  • 00:19:46
    this set up to automatically happen
  • 00:19:48
    about 24 hours after every Tuesday ends
  • 00:19:51
    this is a consistent observation seen
  • 00:19:53
    across all outages nearly every single
  • 00:19:57
    week but why would someone run thousands
  • 00:20:01
    or tens of thousands of bots all at once
  • 00:20:03
    is this even possible the only type of
  • 00:20:06
    Bot that makes sense to operate at this
  • 00:20:08
    scale are those outside of Game servers
  • 00:20:10
    since it becomes way too resource
  • 00:20:12
    intensive to host that many Bots playing
  • 00:20:14
    the game like I mentioned earlier so
  • 00:20:16
    it's not cheaters and trading sites
  • 00:20:19
    don't need 10,000 Bots that leaves
  • 00:20:21
    idlers they're far less resource
  • 00:20:23
    intensive and using cracked versions of
  • 00:20:26
    the game running in text mode it's very
  • 00:20:28
    easy to run a ton of bots on minimal
  • 00:20:30
    Hardware so it's very much possible and
  • 00:20:33
    evidence suggests that there is a
  • 00:20:35
    shitload of bots now listen I've known
  • 00:20:38
    for years that Bots used to be set up to
  • 00:20:40
    idle for item drops but apparently this
  • 00:20:43
    is still happening despite it having
  • 00:20:45
    been over a decade since valves
  • 00:20:47
    supposedly made that harder to do so I
  • 00:20:50
    did some digging I reached out to some
  • 00:20:52
    Traders and eventually came across a guy
  • 00:20:54
    who specifically deals in hundreds of
  • 00:20:57
    thousands of it item cases yes item
  • 00:21:00
    cases and he typically acquired his
  • 00:21:03
    cases in bulk from one seller he has a
  • 00:21:06
    connection that receives up to 2 million
  • 00:21:09
    cases in one year now assuming this to
  • 00:21:12
    be true either this guy's connection
  • 00:21:14
    manages to buy or trade for 225 cases
  • 00:21:17
    every hour on the hour 5,480 cases a day
  • 00:21:21
    for a year straight which isn't humanly
  • 00:21:23
    possible on a single steam account or he
  • 00:21:26
    has a network of accounts doing this for
  • 00:21:28
    him given that figure of 2 million cases
  • 00:21:31
    from one Trader and assuming about two
  • 00:21:33
    case drops per week which is roughly
  • 00:21:35
    average per Steam account that would
  • 00:21:37
    require
  • 00:21:39
    19,23 accounts running year round which
  • 00:21:43
    is similar to the magnitude of the
  • 00:21:45
    spikes and dips we're seeing in the data
  • 00:21:48
    oh God and you know what that's a lower
  • 00:21:51
    bound if you want to speed things up
  • 00:21:53
    you're going to need more accounts or
  • 00:21:55
    separate bot networks and this is just
  • 00:21:57
    one guy there's definitely other people
  • 00:22:00
    operating bot farms at various scales
  • 00:22:03
    now I'm sure some of y'all are thinking
  • 00:22:05
    to yourself why item cases why not idle
  • 00:22:08
    for regular items won't those be worth
  • 00:22:10
    more in the long run because you can
  • 00:22:12
    craft them into metal and yeah your
  • 00:22:14
    typical item case is cheaper than a
  • 00:22:16
    piece of refined metal but it's a lot
  • 00:22:19
    faster to get one case drop than the 18
  • 00:22:22
    weapons required for a single piece of
  • 00:22:24
    refined metal cases are a lot easier to
  • 00:22:26
    acquire because you don't need to log
  • 00:22:28
    into gamees servers unlike regular items
  • 00:22:30
    which drop on a separate timer cases are
  • 00:22:32
    dispensed in Mass several times a week
  • 00:22:34
    if you simply have the game running for
  • 00:22:37
    free valve really wants people to get
  • 00:22:39
    into the gacha and if they're reminded a
  • 00:22:41
    few times a week to do so well you know
  • 00:22:43
    gambling addicts we also have botn Nets
  • 00:22:46
    that fire up specifically to harvest
  • 00:22:48
    gift Ault drops during the Christmas
  • 00:22:50
    season but more on that later and as was
  • 00:22:53
    confirmed by others I spoke with
  • 00:22:55
    traditional item Farms do still exist
  • 00:22:58
    but but likely use far smaller botn Nets
  • 00:23:01
    these periodically log into Game servers
  • 00:23:03
    because they're farming normal item
  • 00:23:04
    drops with the aforementioned
  • 00:23:06
    restrictions you remember those smaller
  • 00:23:08
    spikes in the data these are likely
  • 00:23:09
    smaller botn Nets a few thousand strong
  • 00:23:11
    firing up a few times a week some come
  • 00:23:14
    online for about a half an hour get
  • 00:23:15
    their drops and then all turn off
  • 00:23:17
    because it's way too resource intensive
  • 00:23:19
    to run these Bots all the time just a
  • 00:23:22
    disclaimer this is a very very general
  • 00:23:26
    overview of how idle Farms operate it's
  • 00:23:28
    definitely not 100% accurate I'm lacking
  • 00:23:31
    a lot of the Nuance details and I wanted
  • 00:23:33
    to know more but extracting detailed
  • 00:23:35
    information about bot Farms proves
  • 00:23:38
    nearly impossible bot hosters are tight
  • 00:23:40
    lipped they don't want these details out
  • 00:23:42
    in the open because there is inherent
  • 00:23:44
    competition because guess what money is
  • 00:23:46
    involved something else the case Trader
  • 00:23:49
    told me is that he knew people that used
  • 00:23:51
    to host item Farms himself included but
  • 00:23:54
    they don't do so anymore he also told
  • 00:23:56
    Mega scatter bomb that idle Farm Farms
  • 00:23:58
    tend to get banned right before seasonal
  • 00:24:01
    updates arrive or at least they used to
  • 00:24:03
    be in fact these band waves and the
  • 00:24:06
    sheer size of the bot Farms are
  • 00:24:08
    corroborated by comments from a valve
  • 00:24:11
    employee who told Uncle DNE back in 2017
  • 00:24:14
    not sure why people weren't more alarmed
  • 00:24:16
    by this that they manually banned
  • 00:24:19
    hundreds of thousands of bots set up to
  • 00:24:22
    farm items and craft metal this was back
  • 00:24:24
    in 2017 mind you years after restricted
  • 00:24:29
    item drops in particular he mentioned
  • 00:24:31
    the reoccurring problem of bots that
  • 00:24:33
    people already use to slowly obtain free
  • 00:24:35
    metal and told us about a specific
  • 00:24:36
    instance very recently where he had to
  • 00:24:38
    personally ban over 200,000 bot accounts
  • 00:24:41
    we were all very impressed by this
  • 00:24:42
    before he told us that the troublemakers
  • 00:24:44
    simply just made 200,000 more accounts a
  • 00:24:47
    few days later so it kind of puts into
  • 00:24:48
    perspective how problematic the people
  • 00:24:50
    who were completely willing to take
  • 00:24:52
    advantage of the free economy that Team
  • 00:24:53
    Fortress 2 offers and how difficult it
  • 00:24:55
    might be to implement a metal sync
  • 00:24:57
    system that would prevent that kind of
  • 00:24:58
    thing from happening this confirms that
  • 00:25:01
    item farming still happens at a massive
  • 00:25:04
    scale even back when valve actively took
  • 00:25:06
    measures against it and Lord knows they
  • 00:25:08
    don't do that now Nothing is Stopping
  • 00:25:10
    them meaning the problem has not only
  • 00:25:12
    been happening for years now but it's
  • 00:25:14
    likely the worst it's ever
  • 00:25:18
    been and Now for Something Completely
  • 00:25:21
    Different remember a little while ago
  • 00:25:23
    when I said almost all video games have
  • 00:25:26
    player counts that exhibit human
  • 00:25:28
    behavior avior keyword almost here's
  • 00:25:30
    unturned another popular game sometimes
  • 00:25:33
    on par with TF2 in terms of player
  • 00:25:35
    counts here's unts player numbers right
  • 00:25:38
    away we see similarities with tf2's
  • 00:25:40
    player data very flat player Cycles with
  • 00:25:43
    simultaneously High player counts a
  • 00:25:46
    jagged player curve riddled with random
  • 00:25:49
    sometimes massive spikes and drops all
  • 00:25:51
    of this is strikingly familiar isn't it
  • 00:25:54
    lo and behold it's a well-known fact
  • 00:25:57
    that un turned as heavily boded players
  • 00:26:00
    caught wind of this a couple years ago
  • 00:26:01
    when almost all reported players
  • 00:26:03
    suddenly went offline and then didn't
  • 00:26:05
    come back for a while no outages no
  • 00:26:08
    cause just gone they also note a massive
  • 00:26:12
    discrepancy between the number of
  • 00:26:13
    players in Game servers and the numbers
  • 00:26:15
    reported by steam we'll get back to that
  • 00:26:17
    later this is just like tf2 except the
  • 00:26:20
    developers of unturned have no idea why
  • 00:26:23
    botting of this level is still happening
  • 00:26:25
    years after the game's timed item drops
  • 00:26:27
    were disabled some suspect that people
  • 00:26:30
    artificially inflate the player count to
  • 00:26:32
    make the game seem more popular than it
  • 00:26:34
    actually is but in reality it's just
  • 00:26:36
    people exploiting the free nature of the
  • 00:26:38
    game and generally speaking if a game is
  • 00:26:41
    live service is free to play and has
  • 00:26:43
    some incentive to rank up accounts or
  • 00:26:45
    offer some type of free currency it's
  • 00:26:47
    almost guaranteed to be botted for
  • 00:26:50
    profit here's another example lost Arc
  • 00:26:53
    it's been well known for years that this
  • 00:26:55
    game is massively boted hundreds of
  • 00:26:57
    thousand thousands of bots are set up to
  • 00:26:59
    collect gold the in-game currency and
  • 00:27:02
    sure enough we see similar behavior in
  • 00:27:04
    this player data flattened player Cycles
  • 00:27:07
    tons of random spikes and drops just
  • 00:27:09
    like tf2 here's another example this
  • 00:27:12
    random game popped up in steam's top 100
  • 00:27:14
    about a month into writing the script
  • 00:27:16
    Governor of Poker 3 yeah for years this
  • 00:27:20
    shitty poker game was dead no one
  • 00:27:22
    touched it and then all of a sudden
  • 00:27:25
    whoop thousands of players but perfect
  • 00:27:28
    flat player curves plus huge spikes and
  • 00:27:31
    drops I'm seeing some sad souls in these
  • 00:27:34
    reviews but then oh look sure enough we
  • 00:27:36
    have another heavily botted game for the
  • 00:27:39
    sake of turning a profit so heavily in
  • 00:27:41
    fact that you can't even see the daily
  • 00:27:43
    player fluctuations this shit's flatter
  • 00:27:45
    than a week old soda in general this is
  • 00:27:48
    what botting activity looks like
  • 00:27:50
    flattened player Cycles massive random
  • 00:27:53
    spikes and dips in the data so alongside
  • 00:27:56
    this thus far I've shown what I believe
  • 00:27:59
    to be pretty solid evidence for high
  • 00:28:02
    volumes of continuous botting activity
  • 00:28:04
    tens of thousands of accounts that
  • 00:28:06
    remain online all the time artificially
  • 00:28:09
    inflating the player numbers a good
  • 00:28:11
    chunk of tf2's player count as reported
  • 00:28:14
    by steam is just Bots farming items for
  • 00:28:17
    profit this was all deduced by simply
  • 00:28:20
    looking at the raw data from Steam and
  • 00:28:22
    then comparing it with other games and
  • 00:28:24
    now for the million-dollar question how
  • 00:28:27
    many Bots are are there how many humans
  • 00:28:30
    are playing TF2 at any given time and
  • 00:28:33
    for that answer we need some
  • 00:28:48
    math here is a wonderful website
  • 00:28:51
    teamwork. TF this website also tracks
  • 00:28:54
    player numbers but it processes the data
  • 00:28:56
    from steam's API a little bit
  • 00:28:58
    differently instead of showing all
  • 00:29:00
    online players teamwork. TF reports a
  • 00:29:03
    snapshot of the number of players inside
  • 00:29:06
    of Game servers this is greatly
  • 00:29:09
    beneficial because this opens up an
  • 00:29:11
    entirely new type of analysis few other
  • 00:29:14
    games have but before that something I
  • 00:29:16
    should note because I know for a fact
  • 00:29:18
    people are going to [ __ ] at me for
  • 00:29:20
    referencing these data there's nothing
  • 00:29:22
    Shady about teamwork. tf's numbers as
  • 00:29:24
    I've seen many others insinuate before
  • 00:29:27
    they simp add up all of the accounts
  • 00:29:29
    inside of Game servers data that's
  • 00:29:31
    directly accessible through steam's API
  • 00:29:33
    and the game itself these data are
  • 00:29:36
    trustworthy and are reasonably accurate
  • 00:29:38
    that aside what's useful about Teamworks
  • 00:29:41
    player counts is that it excludes all of
  • 00:29:44
    the accounts on the main menu or in Q
  • 00:29:46
    for a game Argo it filters out nearly
  • 00:29:49
    all idle and trade Bots at the cost of
  • 00:29:52
    omitting some humans as well however
  • 00:29:55
    these data also include cheater Bots and
  • 00:29:58
    idle Bots inside of Game servers since
  • 00:30:01
    these cannot be parsed from humans
  • 00:30:04
    teamwork. tf's data is not showing 100%
  • 00:30:08
    human activity but it's definitely
  • 00:30:10
    mostly human and that's evident by
  • 00:30:13
    simply looking at the data itself we can
  • 00:30:15
    see very strong daily player Cycles no
  • 00:30:19
    weird spikes or dips either at first
  • 00:30:22
    glance this looks like data from any
  • 00:30:24
    other healthy video game and thankfully
  • 00:30:26
    unlike Steam we can look at historic
  • 00:30:29
    data from
  • 00:30:31
    teamwork.com hourly but only from the
  • 00:30:34
    last couple of weeks thankfully however
  • 00:30:37
    some unsung hero has been saving daily
  • 00:30:40
    snapshots of this web page on the
  • 00:30:42
    Wayback machine something you cannot do
  • 00:30:45
    with steam or steam DB so spent a couple
  • 00:30:48
    of days and I scraped all the data as
  • 00:30:50
    much of a [ __ ] it was and check this
  • 00:30:52
    [ __ ] out what we have here is an over
  • 00:30:55
    threeyear long highresolution hisory
  • 00:30:57
    historic record of the number of players
  • 00:31:00
    inside of TF2 Game servers from late
  • 00:31:03
    2020 to present day this is the first
  • 00:31:06
    time in tf2's history that we've seen
  • 00:31:09
    such a complete record of player numbers
  • 00:31:11
    so let's dig in let's compare teamwork.
  • 00:31:14
    tf's data to the data that I've been
  • 00:31:16
    collecting from Steam to keep things
  • 00:31:17
    simple let's just take about a week's
  • 00:31:19
    worth of data to start I'll plot them on
  • 00:31:21
    the same chart and the first thing
  • 00:31:23
    you'll notice is a massive discrepancy
  • 00:31:26
    teamwork. TF has far lower player counts
  • 00:31:29
    than steam it's anywhere from 25% of
  • 00:31:32
    steam's values to often times way less
  • 00:31:34
    and only half of that 25% resides in
  • 00:31:37
    valve servers which is a very
  • 00:31:40
    interesting observation one that
  • 00:31:42
    confirmed some suspicions of my own such
  • 00:31:45
    a low volume of human players explains
  • 00:31:47
    the long Q times during anything other
  • 00:31:49
    than Peak player hours if we really had
  • 00:31:52
    70,000 people playing most of the time Q
  • 00:31:55
    times even during off hours would be
  • 00:31:58
    atrocious for anything other than the
  • 00:32:00
    five most popular Maps additionally such
  • 00:32:03
    a low volume of players inside of valve
  • 00:32:05
    servers explains why we feel such a low
  • 00:32:08
    volume of cheater Bots if we really had
  • 00:32:10
    70,000 humans playing TF2 1,000 cheater
  • 00:32:14
    Bots wouldn't amount to much but with
  • 00:32:16
    only 5 to 10,000 humans inside of valve
  • 00:32:19
    servers that explains why so few cheater
  • 00:32:22
    Bots are way more tangible because that
  • 00:32:24
    means on average inside of valve servers
  • 00:32:27
    one out of every eight players is a
  • 00:32:30
    [ __ ] bot but are Bots alone what
  • 00:32:32
    causes this massive discrepancy Devil's
  • 00:32:35
    Advocate says hey this could be mostly
  • 00:32:37
    humans excluded by teamwork. TF and yes
  • 00:32:40
    teamwork. TF excludes players not inside
  • 00:32:44
    of Game servers so people in queue for a
  • 00:32:46
    game or messing around in their backpack
  • 00:32:48
    or opening cases gawking at fempire or
  • 00:32:51
    what have you they don't show up but we
  • 00:32:53
    should not expect for the number of
  • 00:32:55
    people not playing the game to be that
  • 00:32:58
    much greater let alone three times
  • 00:33:00
    greater than those playing the game if
  • 00:33:03
    that were true something is wrong the
  • 00:33:06
    matchmaking system is broken the servers
  • 00:33:08
    are down something not to mention about
  • 00:33:11
    40 to 60% of server activity takes place
  • 00:33:14
    inside of community servers where
  • 00:33:16
    players tend to stay put for longer
  • 00:33:17
    periods of time this would not translate
  • 00:33:20
    to three out of four players waiting
  • 00:33:22
    outside of Game servers additionally
  • 00:33:25
    even if these were all humans outside of
  • 00:33:27
    game servers they should also go to bed
  • 00:33:30
    just the same as any other human so why
  • 00:33:32
    does it seem like they never log off as
  • 00:33:34
    a very basic observation steam's player
  • 00:33:37
    numbers look almost exactly like
  • 00:33:39
    teamwork numbers just noisier with all
  • 00:33:42
    of those spikes and shifted upwards by a
  • 00:33:44
    good amount again this would happen due
  • 00:33:47
    to a bunch of accounts that never go
  • 00:33:48
    offline raising the minimum Baseline of
  • 00:33:51
    online accounts so is this discrepancy
  • 00:33:55
    human no mostly not this discrepancy
  • 00:33:58
    occurs because again teamwork. TF
  • 00:34:01
    excludes Some Humans but also all of the
  • 00:34:04
    idling and trade Bots that stay online
  • 00:34:07
    all the time and now to address the
  • 00:34:09
    million-dollar question how many Bots
  • 00:34:12
    are there what is Team Fortress 2's real
  • 00:34:15
    human player count now bear with me I'll
  • 00:34:18
    take y'all through this nice and slow
  • 00:34:21
    math can be scary but I'll be your guide
  • 00:34:26
    [Music]
  • 00:34:34
    I'm sure y'all's first instinct is to
  • 00:34:36
    Simply take the difference between the
  • 00:34:37
    peak player counts and that difference
  • 00:34:39
    would be all the Bots you ship the
  • 00:34:41
    player curves back down and then you're
  • 00:34:43
    done this wouldn't be accurate on
  • 00:34:45
    average Teamworks player Cycles are
  • 00:34:47
    about 8,000 players strong whereas steam
  • 00:34:50
    Cycles are about 18,000 players strong
  • 00:34:54
    we already know that player Cycles are
  • 00:34:56
    mostly if not in entirely human because
  • 00:34:59
    remember people got to sleep so given
  • 00:35:01
    that difference it seems like 10,000
  • 00:35:03
    humans aren't accounted for in teamwork.
  • 00:35:06
    tf's Daily player fluctuations Argo a
  • 00:35:08
    direct shift of values would not be
  • 00:35:11
    appropriate nor accurate however what we
  • 00:35:13
    can do is compare percentages for
  • 00:35:17
    example let's take a look at DOTA pubg
  • 00:35:21
    rust and Apex Legends if I take the
  • 00:35:24
    player data and then split it up by day
  • 00:35:26
    if I then take the differ between the
  • 00:35:28
    maximum player count and the minimum
  • 00:35:30
    player count on each day we get the
  • 00:35:32
    average size of that game's daily player
  • 00:35:35
    flux but then if I divide the size of
  • 00:35:37
    that player flux by the maximum daily
  • 00:35:40
    player counts this Returns the
  • 00:35:42
    percentage of the total player count
  • 00:35:44
    that is represented in that day's
  • 00:35:46
    fluctuation for instance Apex Legends
  • 00:35:49
    the size of its player Cycles represents
  • 00:35:52
    about 75% of its maximum player counts
  • 00:35:55
    pubg about 82% DOTA 57% rust 50% and
  • 00:36:02
    what's fascinating is that this
  • 00:36:04
    percentage is robustly consistent
  • 00:36:06
    varying very slightly regardless of the
  • 00:36:09
    time of year regardless of game updates
  • 00:36:11
    Etc this holds true even after major
  • 00:36:14
    game updates where the volume of players
  • 00:36:16
    changes drastically for example Apex
  • 00:36:19
    Legends saw a massive update back in
  • 00:36:22
    February doubling the player count but
  • 00:36:24
    the size of its player Cycles still
  • 00:36:26
    represents a about 75% of the peak
  • 00:36:29
    player counts and this makes sense you
  • 00:36:32
    should expect for the proportion of
  • 00:36:34
    daily activity to scale with the volume
  • 00:36:36
    of humans engaging with your game adding
  • 00:36:39
    or removing players would not change the
  • 00:36:41
    proportion of people that fall into
  • 00:36:43
    General sleeping patterns and we are
  • 00:36:46
    extremely fortunate to have teamwork.
  • 00:36:48
    tf's player numbers again those player
  • 00:36:51
    counts are mostly human so we can use
  • 00:36:54
    teamwork. tf's player Cycles as a tool
  • 00:36:57
    to estimate tf2's true player count
  • 00:37:00
    let's take a look at a more recent
  • 00:37:01
    week's worth of data teamwork. tf's
  • 00:37:04
    player fluctuations represent about 50%
  • 00:37:07
    of its daily Peak player counts we
  • 00:37:09
    should expect for this value to be the
  • 00:37:11
    same if we look at steam's player Cycles
  • 00:37:14
    again the size of player Cycles should
  • 00:37:16
    scale up with an increasing volume of
  • 00:37:19
    humans so if we lump all of the humans
  • 00:37:21
    on the main menu or trading what have
  • 00:37:23
    you with the humans inside of Game
  • 00:37:26
    servers that percentage AG should remain
  • 00:37:28
    at around 50% but it doesn't steam's
  • 00:37:32
    fluctuations represent a mere 16% of its
  • 00:37:36
    total player counts this reinforces that
  • 00:37:38
    steam's numbers are massively inflated
  • 00:37:41
    by a bunch of accounts that never go
  • 00:37:42
    offline again raising that minimum
  • 00:37:45
    Baseline of online accounts shifting
  • 00:37:47
    that player curve upwards so how far do
  • 00:37:50
    I need to shift steam's numbers down to
  • 00:37:53
    reach that figure of 50% about 48,000
  • 00:37:57
    accounts need to go away so that tells
  • 00:38:00
    me right now about 70% of steam's
  • 00:38:04
    reported player account is
  • 00:38:08
    Bots this is a conservative estimate
  • 00:38:10
    given that I can't reliably account for
  • 00:38:12
    Bots inside of Game servers and the
  • 00:38:14
    trickiest part about this process is
  • 00:38:16
    that you can't get an exact number of
  • 00:38:18
    bots at any time that's impossible and
  • 00:38:21
    educated but ballpark estimate is the
  • 00:38:23
    best we can do masterc the source code
  • 00:38:25
    wizard that I'm sure some of y'all are
  • 00:38:27
    famili with wrote a tool that apparently
  • 00:38:29
    detects botting activity they've only
  • 00:38:31
    ever given one public estimate of about
  • 00:38:33
    20 to 30,000 Bots from this tool from a
  • 00:38:36
    couple years ago but this appears to
  • 00:38:37
    only capture known bot and cheater
  • 00:38:40
    accounts so it's an incomplete estimate
  • 00:38:43
    Additionally the ratio of bots to humans
  • 00:38:46
    changes throughout the year from dayto
  • 00:38:48
    day even as bot Nets of various sizes
  • 00:38:50
    come and go for example if I go back to
  • 00:38:52
    say the first week of February teamwork.
  • 00:38:55
    tf's player Cycles are still pretty much
  • 00:38:57
    much the same still representing 50% of
  • 00:39:00
    the daily Peaks but steam's data is very
  • 00:39:03
    different here steam's player counts are
  • 00:39:05
    much higher with the daily player Cycles
  • 00:39:08
    now representing only 12% of the daily
  • 00:39:11
    peaks in this case I need to knock off
  • 00:39:14
    about 70,000 accounts in order for it to
  • 00:39:17
    reach that 50% during this week about
  • 00:39:20
    80% of steam's player count was bots in
  • 00:39:25
    general as a very generous IM in favor
  • 00:39:28
    of players tf2's true player count
  • 00:39:30
    likely hovers around 25 to 30,000 humans
  • 00:39:34
    on any given day this is still pretty
  • 00:39:36
    good for a now almost 17-year-old game
  • 00:39:39
    but that player count reported by steam
  • 00:39:41
    is mostly Bots and if this doesn't
  • 00:39:43
    concern you as a tf2 fan if I have yet
  • 00:39:46
    to convince you that TF2 is heavily
  • 00:39:49
    boded I'm sorry and I really am saying
  • 00:39:51
    this with a heavy heart because it makes
  • 00:39:53
    me sad as well but it's about to get
  • 00:39:55
    worse a lot worse
  • 00:40:02
    [Music]
  • 00:40:07
    December 21st
  • 00:40:09
    2023 tf2's player count was around
  • 00:40:12
    69,000 players less than 36 hours later
  • 00:40:16
    the player count triples to nearly
  • 00:40:20
    216,000 2 weeks after the Smith SMI
  • 00:40:24
    update launched there was a patch on
  • 00:40:26
    this day but I seriously doubt
  • 00:40:29
    increasing the size of the frog on
  • 00:40:31
    Carrier brought out 147,000 new players
  • 00:40:35
    Devil's Advocate says this could be a
  • 00:40:37
    bunch of people being let off for the
  • 00:40:38
    holidays but no other games see a
  • 00:40:41
    proportional spiking players at this
  • 00:40:43
    time which is what you would expect if
  • 00:40:44
    this was the case well the devil now
  • 00:40:46
    says well you just got a bunch of people
  • 00:40:48
    coming online to unbox the new case
  • 00:40:50
    after the update came out that would be
  • 00:40:52
    believable if it wasn't 2 weeks after
  • 00:40:55
    the update launched and then January 7th
  • 00:40:59
    2024 50,000 accounts go offline in mere
  • 00:41:04
    minutes the same day when the smithi
  • 00:41:07
    update ended the same hour the update
  • 00:41:10
    ended you honestly can't expect me to
  • 00:41:13
    believe that more than onethird of all
  • 00:41:16
    reported players at the time suddenly
  • 00:41:18
    decided to log off all at once no steam
  • 00:41:20
    outages no server issues no vac waves
  • 00:41:24
    just gone what you're looking at are
  • 00:41:27
    massive botn Nets firing up during the
  • 00:41:30
    smithi update specifically to harvest
  • 00:41:32
    the new item cases and as I mentioned
  • 00:41:35
    earlier gift Ault drops during their
  • 00:41:37
    most popular time window of use huge
  • 00:41:39
    chunks of bots sporadically disappear
  • 00:41:41
    about 40,000 Bots right here for example
  • 00:41:44
    and then finally when the smithi update
  • 00:41:46
    ends the remaining 50,000 all go offline
  • 00:41:50
    there's no sense in running these Bots
  • 00:41:51
    anymore if players aren't logging on
  • 00:41:53
    getting their stockings and then using
  • 00:41:55
    the gift Ault to send a gift I also
  • 00:41:57
    doubt this was someone at valve Banning
  • 00:41:59
    the accounts because they can't even
  • 00:42:01
    properly Implement a hat or a map these
  • 00:42:03
    days not to mention as discussed earlier
  • 00:42:05
    they tend to ban these bot Farms before
  • 00:42:08
    the updates come out so they can't
  • 00:42:09
    exploit the free items and you know what
  • 00:42:12
    it makes sense to run bot Farms of this
  • 00:42:14
    magnitude you collect and sell cases
  • 00:42:16
    when they're more popular and if your
  • 00:42:18
    Bots make up more than 2third of all
  • 00:42:20
    players you're getting more than 2third
  • 00:42:22
    of all gift Ault drops because they're
  • 00:42:24
    delivered to any random online account
  • 00:42:26
    you don't need to be inside of Game
  • 00:42:28
    servers oh what not good enough for you
  • 00:42:31
    all right let's compare steam's data to
  • 00:42:33
    teamwork. tf's data during the Smith
  • 00:42:36
    2023 update if that Spike of 216,000
  • 00:42:40
    players on December 21st was comprised
  • 00:42:42
    of at least some human activity we
  • 00:42:45
    should see some spiking players inside
  • 00:42:48
    of Game servers and enjoying the new
  • 00:42:50
    content right nope throughout the entire
  • 00:42:54
    month of December from the launch of the
  • 00:42:55
    mismas update to it final day player
  • 00:42:58
    behavior is stagnant no increase when
  • 00:43:02
    all those accounts come online on the
  • 00:43:03
    21st and no decrease when they all drop
  • 00:43:06
    offline on the 7th hell we don't even
  • 00:43:09
    see much of a bump in players when the
  • 00:43:10
    update launches it didn't even surpass
  • 00:43:13
    the normal player numbers that happen
  • 00:43:14
    over a weekend that's pretty [ __ ] sad
  • 00:43:17
    and to back this up masterc coms raised
  • 00:43:19
    an alarm about gift Ault Harvesters as
  • 00:43:21
    well 5 days after those 147,000 accounts
  • 00:43:26
    came online they saw the exact same
  • 00:43:29
    behavior no appreciable spacking players
  • 00:43:32
    only in boding activity and that's the
  • 00:43:34
    nail on the coffin Not only was the
  • 00:43:36
    smithi 2023 update massively botted but
  • 00:43:39
    that player peak of
  • 00:43:41
    216,000 players wasn't even close to
  • 00:43:44
    being real I'm sorry but how in the hell
  • 00:43:48
    am I supposed to believe any player
  • 00:43:49
    Peaks anymore I simply used to take
  • 00:43:51
    these at face value and think to myself
  • 00:43:54
    wow the game's doing really well but now
  • 00:43:56
    that I've seen Bing activity of this
  • 00:43:58
    magnitude during a lackluster smithm
  • 00:44:01
    update are any previous spikes in
  • 00:44:04
    players real was the all-time player
  • 00:44:06
    peak of
  • 00:44:08
    254,000 players even close to being real
  • 00:44:13
    no I don't think so and sadly I can't
  • 00:44:16
    get highresolution data from Steam
  • 00:44:18
    around the time of previous major
  • 00:44:19
    updates I've only been recording myself
  • 00:44:22
    since December but thankfully I have
  • 00:44:23
    other pieces of useful information the
  • 00:44:25
    one thing I could find from around the
  • 00:44:27
    summer update was this screenshot of a
  • 00:44:29
    game Journal article of that
  • 00:44:31
    highresolution data I can roughly
  • 00:44:34
    estimate the time during the update by
  • 00:44:35
    drawing a bar that represents a day and
  • 00:44:38
    then using that as a ruler this right
  • 00:44:40
    here is a daily cycle so I can use that
  • 00:44:42
    as my reference if I look at the initial
  • 00:44:44
    jump in players the player count almost
  • 00:44:47
    instantly doubles it looks like about
  • 00:44:49
    75,000 players log on over the course of
  • 00:44:52
    3 hours that's suspicious but could
  • 00:44:54
    easily be handwaved away as the initial
  • 00:44:56
    jump jumping players logging on for the
  • 00:44:58
    hype fine but then about 10 hours later
  • 00:45:02
    a patch releases seemingly knocking
  • 00:45:04
    those same 75,000 accounts offline oddly
  • 00:45:08
    familiar to maintenance day we have a
  • 00:45:10
    couple of hours of otherwise normal
  • 00:45:12
    activity so to speak and then suddenly a
  • 00:45:15
    100,000 accounts snap back online yeah
  • 00:45:19
    [ __ ] me if that doesn't look like
  • 00:45:21
    botting activity where botn Nets got
  • 00:45:23
    rapidly turned back on after that patch
  • 00:45:25
    came out but I can go even further I can
  • 00:45:28
    compare the player data from teamwork.
  • 00:45:30
    TF with the downsample data that we do
  • 00:45:33
    have from Steam sadly teamwork. TF has a
  • 00:45:36
    lot of missing data around this time but
  • 00:45:38
    with the data we do have we do see a
  • 00:45:40
    bump in players after the update comes
  • 00:45:42
    out but hilariously and somewhat
  • 00:45:44
    depressingly player counts aren't that
  • 00:45:47
    different than what we see during
  • 00:45:48
    previous year's updates of any kind not
  • 00:45:51
    to mention the number of players
  • 00:45:53
    reported inside of Game servers isn't
  • 00:45:55
    even remotely similar in magnitude to
  • 00:45:58
    what we see in steam's data no more than
  • 00:46:01
    10% of what's being reported during that
  • 00:46:03
    initial jump in players and just like
  • 00:46:06
    the smithm update when all of those
  • 00:46:08
    players drop offline from Steam we see
  • 00:46:11
    none of that reflected inside of server
  • 00:46:13
    activity player counts do remain
  • 00:46:15
    slightly elevated during this time but
  • 00:46:17
    they're entirely stagnant so that tells
  • 00:46:20
    me none of the activity reported by
  • 00:46:22
    steam was human otherwise it would have
  • 00:46:25
    been reflected even somewhat but
  • 00:46:27
    reflected inside of the human activity
  • 00:46:29
    we see inside of Game servers I think I
  • 00:46:31
    can firmly conclude that the summer
  • 00:46:33
    update was massively botted that player
  • 00:46:36
    peak of 254,000 players wasn't even
  • 00:46:41
    close to being real and what about
  • 00:46:43
    previous updates summer 2021 same story
  • 00:46:48
    huge spikes in players on Steam but a
  • 00:46:50
    very slight uptick in humans inside of
  • 00:46:52
    Game servers about the same amount we
  • 00:46:54
    saw in 2023 some 2022 even worse zero
  • 00:47:00
    uptick in human activity smithi 2021
  • 00:47:04
    another massive Spike on Steam but the
  • 00:47:06
    number of players in Game servers
  • 00:47:08
    doesn't even break what's normally seen
  • 00:47:10
    on a weekend smithi 2022 another massive
  • 00:47:14
    Spike on Steam but absolutely no change
  • 00:47:17
    in the number of players inside of Game
  • 00:47:18
    servers Scream Fortress well generally
  • 00:47:21
    we do see a very slight uptick in the
  • 00:47:23
    number of players in servers humans are
  • 00:47:25
    more incentivized to play at this time
  • 00:47:27
    because of contracts and rewards but
  • 00:47:29
    just like summer 2023 the spike in
  • 00:47:32
    players on Steam was insane peing at
  • 00:47:35
    just over 200,000 players but almost
  • 00:47:39
    zero noticeable Spike was seen in the
  • 00:47:42
    players inside of Game servers given
  • 00:47:45
    that almost zero human activity is
  • 00:47:47
    reflected inside of Game servers during
  • 00:47:49
    every single update I think I can at the
  • 00:47:52
    very least pose the possibility that
  • 00:47:55
    pretty much every major update for TF2
  • 00:47:58
    is botted to high hell and one final
  • 00:48:00
    thing to note if I look at the entire
  • 00:48:02
    record from teamwork. TF over the last 3
  • 00:48:06
    years player numbers have been stable no
  • 00:48:09
    tangible growth but no tangible decline
  • 00:48:12
    either whereas on Steam we see continued
  • 00:48:15
    steady growth since 2018 if this were
  • 00:48:19
    truly the case shouldn't we expect to
  • 00:48:21
    see that growth reflected in server
  • 00:48:23
    activity we don't so this suggests at
  • 00:48:26
    least for me that much of tf2's growth
  • 00:48:29
    after jungle Inferno has been driven by
  • 00:48:31
    increasing bot activity not new players
  • 00:48:34
    as valve gradually started giving less
  • 00:48:36
    of a [ __ ] as the years went by they
  • 00:48:38
    don't ban Idol Bots anymore let alone
  • 00:48:40
    any cheater Bots resulting in heavily
  • 00:48:43
    inflated player numbers seems like
  • 00:48:45
    that's been the case for many years now
  • 00:48:47
    heavily implying that roughly the same
  • 00:48:50
    amount of people have been playing TF2
  • 00:48:52
    for the greater part of the past decade
  • 00:48:54
    about 25 to 30,000 people
  • 00:49:01
    [Music]
  • 00:49:06
    while I may not have direct confirmation
  • 00:49:08
    from the mouths of Bot holters as to how
  • 00:49:10
    many Bots are running at one time I have
  • 00:49:12
    presented strong evidence that TF2 is
  • 00:49:15
    heavily boded we have evidence in tf2's
  • 00:49:18
    own player counts evidence from traders
  • 00:49:21
    that botn Nets exist and operate at huge
  • 00:49:23
    scales word from valve that they've
  • 00:49:25
    dealt with huge botn Nets in the past
  • 00:49:28
    Bots correlation of identical behavior
  • 00:49:31
    in other video games that are well known
  • 00:49:33
    to be botted evidence from that massive
  • 00:49:35
    discrepancy between Steam and the mostly
  • 00:49:38
    human player activity inside of Game
  • 00:49:40
    servers and the clear as day evidence
  • 00:49:43
    that pretty much every major update over
  • 00:49:46
    the last 3 years has been massively
  • 00:49:48
    boded I can firmly conclude that a large
  • 00:49:51
    majority of tf2's player count as
  • 00:49:53
    reported by steam at any time are Bots
  • 00:49:57
    mostly idlers we cannot trust any player
  • 00:50:01
    numbers let alone player Peaks being
  • 00:50:03
    reported by steam that's the main
  • 00:50:05
    takeaway from this video not an exact
  • 00:50:08
    number of bots in TF2 again that's
  • 00:50:10
    impossible but rather the observation
  • 00:50:13
    that about 70 to 80% of tf2's player
  • 00:50:16
    count as reported by steam is not human
  • 00:50:21
    thing is we only feel about 3% of those
  • 00:50:24
    Bots the cheaters idlers don't impact
  • 00:50:27
    the gameplay experience unless you're
  • 00:50:29
    involved in tf2's relatively small micro
  • 00:50:31
    economy but another major takeaway from
  • 00:50:34
    this video is that my work here shows
  • 00:50:36
    how that 3% of bots can [ __ ] the game
  • 00:50:39
    with so few real players inside of valve
  • 00:50:42
    servers it really doesn't take much to
  • 00:50:44
    cause a headache all right let's exhale
  • 00:50:47
    for a moment here this has been a long
  • 00:50:49
    video with a lot of information to
  • 00:50:50
    process that I guarantee isn't making
  • 00:50:53
    people feel very good you're not meant
  • 00:50:54
    to feel good after this video You're
  • 00:50:56
    meant to be informed and not really sure
  • 00:50:59
    how to close this out but it wouldn't be
  • 00:51:01
    a video on my channel if I didn't give
  • 00:51:03
    some thoughts of my own so I'll do that
  • 00:51:05
    now from my point of view a lack of
  • 00:51:08
    growth makes perfect sense for TF2 I've
  • 00:51:11
    been playing TF2 since launch I've seen
  • 00:51:13
    every single thing that this game has
  • 00:51:15
    been through why on Earth would anyone
  • 00:51:18
    have expected for TF2 to grow after 2018
  • 00:51:23
    in 2016 TF2 left its worst update behind
  • 00:51:26
    itself meet your match which alienated
  • 00:51:29
    countless players thanks to the horrific
  • 00:51:31
    matchmaking system and of course all the
  • 00:51:33
    other changes valve made with zero
  • 00:51:35
    thought didn't even ask us didn't even
  • 00:51:37
    test anything well they did the beta but
  • 00:51:38
    that was basically nothing and then
  • 00:51:40
    whoop There You Go they dumped that in
  • 00:51:41
    our laps and that was it in the two
  • 00:51:43
    years following players were begging
  • 00:51:46
    praying for some kind of a change a roll
  • 00:51:48
    back something to make that dumpster
  • 00:51:50
    fire of an update just a little bit more
  • 00:51:52
    bearable we thought that was going to
  • 00:51:54
    come in 2017 when jungle and Inferno
  • 00:51:56
    came around players came back in droves
  • 00:51:59
    for that update but once everyone
  • 00:52:01
    realized that the game was going to stay
  • 00:52:03
    as is players left in droves entering
  • 00:52:07
    2019 TF2 saw its lowest player counts
  • 00:52:11
    since the game went free to play and
  • 00:52:13
    then in the following years TF2 was
  • 00:52:16
    completely abandoned gradually flooded
  • 00:52:18
    by Bots and cheaters anti-consumer
  • 00:52:21
    practices punishing freeo place which
  • 00:52:23
    valve absolutely profits from mediocre
  • 00:52:25
    at best to terrible at worst content
  • 00:52:28
    assuming it functioned sure TF2 is free
  • 00:52:30
    TF2 is fun but when you have all of this
  • 00:52:33
    going against it and the fact that it's
  • 00:52:36
    unplayable for half the day more than
  • 00:52:38
    half the day why in the hell would we
  • 00:52:41
    expect for players let alone new players
  • 00:52:43
    to stick around and then make the Game
  • 00:52:45
    grow it makes no sense you see that to
  • 00:52:49
    me reinforces the validity behind tf2's
  • 00:52:52
    true player count being far far smaller
  • 00:52:55
    than what we see on Ste
  • 00:52:57
    and if I dare may be as so bold as to be
  • 00:52:59
    even more candid what's interesting at
  • 00:53:01
    least for me is that for anyone who
  • 00:53:04
    doesn't play TF2 the game looks totally
  • 00:53:07
    fine the vocal spheres surrounding TF2
  • 00:53:10
    are very misleading I'm not saying
  • 00:53:12
    they're doing this intentionally it's
  • 00:53:14
    simply the nature of people being able
  • 00:53:15
    to ignore or edit out all of the bad TF2
  • 00:53:18
    experiences so we have massive YouTube
  • 00:53:22
    channels making videos with millions of
  • 00:53:24
    views countless viral posts and art and
  • 00:53:27
    memes what have you that make it seem as
  • 00:53:29
    if though Team Fortress 2 is this
  • 00:53:31
    flourishing game where everything is
  • 00:53:34
    totally fine Team Fortress 2 the game
  • 00:53:37
    everyone loves talking about but as I've
  • 00:53:39
    shown not that many people play so as a
  • 00:53:42
    result we have this overly positive
  • 00:53:45
    often times toxically positive
  • 00:53:47
    misleading picture painted by the
  • 00:53:49
    discourse surrounding TF2 it's basically
  • 00:53:52
    taboo to say anything remotely negative
  • 00:53:55
    about this game you you can't even call
  • 00:53:57
    an objectively terrible weapon skin
  • 00:53:59
    stinky without a bigger figure strong
  • 00:54:02
    arming their way into the discourse and
  • 00:54:03
    trying to shut you down I don't know
  • 00:54:06
    what causes this perhaps itms peer
  • 00:54:08
    pressure to remain positive all the time
  • 00:54:10
    because it's good for your brand or
  • 00:54:12
    ignorance is bliss I don't know but no
  • 00:54:14
    one talks about the Bots no one makes
  • 00:54:17
    videos about the Bots if you only watch
  • 00:54:19
    TF2 YouTube videos or God forbid
  • 00:54:22
    consumed TF2 Twitter and Reddit and
  • 00:54:24
    didn't play the game you'd have no no
  • 00:54:26
    idea how bad things really are it's
  • 00:54:29
    almost an afterthought like the Bots
  • 00:54:30
    aren't even here and for some reason
  • 00:54:33
    it's frowned upon to talk about the Bots
  • 00:54:35
    if you address the problem people either
  • 00:54:38
    dismiss you outright or get mad at you
  • 00:54:40
    they tell you to stop being negative or
  • 00:54:42
    nothing will change so just shut up
  • 00:54:44
    about it if any other game were flooded
  • 00:54:46
    by Bots and cheaters rendering it
  • 00:54:48
    unplayable CS2 Apex Legends pick any
  • 00:54:51
    other popular game there would be
  • 00:54:53
    [ __ ] riots in fact that's happening
  • 00:54:55
    right now with CS2 valve is royally
  • 00:54:58
    [ __ ] their biggest Money Maker aside
  • 00:55:01
    from Steam and people are fed up and yet
  • 00:55:03
    for some reason TF2 fans are just
  • 00:55:07
    complacent they keep giving valve money
  • 00:55:09
    and praising this game despite their
  • 00:55:11
    game being unplayable for half the day
  • 00:55:14
    praising every little thing that comes
  • 00:55:15
    their way regardless of quality or care
  • 00:55:18
    there is no solution to this problem or
  • 00:55:20
    any other problem other than to speak up
  • 00:55:23
    about it because as consumers that's all
  • 00:55:25
    we can do that's what I've been doing on
  • 00:55:27
    my channel for years now in regards to
  • 00:55:30
    this game and that's why I made this
  • 00:55:32
    video I believe it's important for
  • 00:55:34
    everyone to understand the reality of
  • 00:55:36
    TF2 situation because no one says it so
  • 00:55:40
    I'll state it plainly the reality is
  • 00:55:42
    that TF2 the game is not thriving it's
  • 00:55:47
    up shits Creek and has been for many
  • 00:55:49
    years now TF2 faces abhorent neglect
  • 00:55:53
    resulting in a plague of cheaters and
  • 00:55:55
    Bots for which valve absolutely keeps
  • 00:55:58
    track of and has the resources to
  • 00:56:00
    address but thus far they've willingly
  • 00:56:03
    chosen not to I strongly doubt they
  • 00:56:06
    can't combat publicly available sheets
  • 00:56:09
    that you can download after a simple
  • 00:56:11
    Google search I seriously doubt they
  • 00:56:14
    can't take some kind of action the
  • 00:56:16
    multi-billion dollar company can't
  • 00:56:19
    handle a few guys in their basement oh
  • 00:56:21
    well sorry guys we're stretched too thin
  • 00:56:23
    right now so have another hat box that
  • 00:56:26
    that's just pathetic there is no excuse
  • 00:56:29
    for this and if anyone dares to utter
  • 00:56:31
    the phrase treadmill work [ __ ] off
  • 00:56:33
    because 90% of the jobs in the world are
  • 00:56:36
    treadmill work it's valve's
  • 00:56:38
    responsibility to maintain this game so
  • 00:56:41
    long as it's being offered is playable
  • 00:56:42
    with incentivized microtransactions yet
  • 00:56:45
    they consistently fail to meet their
  • 00:56:47
    responsibilities despite TF2 earning
  • 00:56:50
    them tens of millions of dollars a year
  • 00:56:53
    Millions earned by simply copying and
  • 00:56:55
    pasting often time times mediocre
  • 00:56:57
    content they don't even make content
  • 00:56:59
    they don't even ensure is functional
  • 00:57:01
    which in tandem with Bots results in
  • 00:57:03
    consistently buggy broken and unplayable
  • 00:57:06
    updates that's been the status quo for
  • 00:57:09
    seven no almost 8 years the onetime
  • 00:57:13
    concert event that amounted to a self-
  • 00:57:15
    congratulatory circle jerk praising how
  • 00:57:17
    much we love TF2 isn't going to do
  • 00:57:20
    anything safe TF2 was like trying to
  • 00:57:22
    tell the fire department that your house
  • 00:57:24
    is burning down by screaming how much
  • 00:57:26
    you love your house instead of telling
  • 00:57:28
    them hey my house is burning down and
  • 00:57:31
    even though I know for a fact valve will
  • 00:57:34
    continue to do nothing here's the right
  • 00:57:36
    way to handle this consistently and
  • 00:57:38
    directly address the problem tell valve
  • 00:57:41
    email them tweet at them make a video
  • 00:57:43
    something speak TF2 is full of bots we
  • 00:57:47
    cannot play it fix your [ __ ] game or
  • 00:57:51
    I'm not your customer anymore I know
  • 00:57:53
    that from this point forward I'm not
  • 00:57:54
    spending a dime on this game or any
  • 00:57:56
    other Valve game until something changes
  • 00:57:58
    and if anyone watching this seriously
  • 00:58:01
    wants valve to do something I suggest
  • 00:58:03
    you all do the same by spending money on
  • 00:58:05
    TF2 you're telling valve that this is
  • 00:58:09
    acceptable and financially incentivizing
  • 00:58:11
    them to keep things as they are speak
  • 00:58:13
    with your mouths and with your wallets I
  • 00:58:16
    can't reinforce that enough on this
  • 00:58:18
    channel I love this game and I want
  • 00:58:20
    nothing more than for it to thrive but
  • 00:58:22
    valve has shown zero interest in
  • 00:58:25
    addressing TF2 biggest problems but they
  • 00:58:27
    will happily keep taking our money and
  • 00:58:29
    that is some 100% grade A horseshit
  • 00:58:38
    [Music]
  • 00:58:54
    [Music]
  • 00:59:04
    [Music]
  • 00:59:29
    well this sucks
タグ
  • Team Fortress 2
  • TF2
  • bots
  • player count
  • Valve
  • gaming
  • cheaters
  • traders
  • idlers
  • community action