Back in 2012, I started writing this site with the interest of covering topics that I always felt were under-represented. My latest passion project, the Hero Builds project, started back in late February of 2013, were about collecting an agreed version of standard play for each Dota 2 hero. As opposed to this opinionated site, I was coming from a place without confidence in my opinions and with much to learn. In many ways, I’ve grasped more than just playing better and also much faster than if I had just kept this project to myself. I am much farther in grasping the ability to not only discern how heroes are to be played but also just how flexible their roles and item/skill builds can be. This is what makes this project continuous, endless and, for the most part, an enjoyment for me to pursue.
This topic will be an overview about this project; the motivation, smaller nuances and statistical achievement. As stated, it’s been about a year and a half since I started this ambitious idea but it’s also been a full year (since October 30th 2013) that I’ve collected statistical data for each and every created guide and its growth. We’ll examine how much we’ve grown, how it still grows and the varying issues from the enlarging time-consumption to the dying competition.
For those new to some of Valve’s Dota 2 features, back in February of 2013, Valve announced a feature called Hero Builds: guides that can be accessed directly in-game and during a match, you are suggested what items to buy and what skills to level. For each selected hero, you were given explanatory prompts that made learning a hero or the game incredibly easy, accessible directly during matches without pausing or needing another window open. Overall, it was a part of Valve’s approach towards approaching Dota 2 easier alongside their other features, such as coaching and their tutorial.
Today we celebrate reaching over 40 million subscriptions across 140 guides (44,774,270 as of November 15th, 2014) and a full year of statistics to inspect and analyze.
From 3 million to 40 million – A year of Statistics
A few things to note:
1. A user cannot subscribe to more than one guide per hero. There are a total of 108 heroes released, thus a total of 108 guides can be subscribed per user.
2. The subscribers per guide are unique, but users can subscribed to several guides, creating overlap in total subscriptions.
3. Users subscribed are still using that guide in-game with all its prompts and suggestions which is different from a subscription to a youtube channel.
4. Data is readily available for those willing to cooperate in retrieving further relevant statistics (please contact me privately)
The short answer to all this is, we’re consistently growing. As Dota 2’s popularity expands, so does the project’s impact. Originally starting at 3 million, after only 8 months since our start (106 total guides), we’ve managed to attain 40 million subscriptions and multiply our average subscription numbers from 30,000 per guide to about 300,000 per guide (106 guides to 140+ total).
The blue indicates the total subscription numbers of that month while the orange indicates the difference in totals between the month prior and the current month it sits on top. To simplify, The orange is how much more from the previous month’s total to today’s total are new subscribers (which is 1.7 to 2 million every 15 days)
Without including any new guides to be created, we’re expecting 115 million (1,064,815 minimum unique) total subscriptions in the coming years (12 months to 18 months), presumably more as new heroes are released and new patches open up new ways to play old heroes.
Typically, the highest growth comes from weeks just after a new patch releases (1.55 to 2.04 million) and soon after The International 4 concluded (2.248 million subscription jump). This is typically due to both the influx of new users after a major event (The International) and how patches either remake some heroes, rendering previous outdated guides even more blatantly outdated or promotes our own guides through their consistency in updates and title indication (all guides in this project are titled with the current patch it is updated on).
In terms of the project’s impact, we typically are within the top 3 to 5 with an average of 300,000 subscriptions per guide. So although we have high subscription number for all guides, we are not leading in as many guides as we’d like, just consistently popular overall. We estimate a minimum of 12% of the total Dota 2 player base are subscribed to our guides (~1.12 million out of 9.7 million) to up to 20%/optimistically 34%). The data set below is our attempt at grabbing the majority population of subscribed users per hero and then seeing what our take is (~21%) and then projecting it onto 108 guides (maximum number per user).
Not 100% accurate as per the issues stated prior to this section. This is a ballpark estimation with 12% being the minimum unique subscribers.
The current issues with the guides’ scene is both the decline in competition of other guide creators as well as the disparity between total subscriptions and lack of ratings. We found that on average, only 30% of the total subscribers actually bother rating the guide they’re using which is quite low. This is a significant issue for newly-released guides as those who remain on top, despite not being updated since 2013, continue to collect a small portion of ratings and newer guides are not rotated in for exposure and ratings. At the moment, 16 guides out of 140 of this project still lack enough ratings to be included in-game.
Another issue is the decline of other guide makers, offering little variance or consistent quality work across all heroes. Against the current top 4 guide creators in the world, only two are currently active: myself at 42.3 million (now 44.7 million/140 guides) and greyshark at 13.7 million and 102 guides. Both Purge and eXplosion maintain high ratings, but no longer actively update their guides or create new ones (and GreyShark rarely updates his guides, making nearly the majority of them outdated).
I’d be lying if I said I was doing this solely for the good of the community. In many ways, the silent dependency of current subscribers, the pages and pages of appreciative comments and the dying scene of in-game Hero Build creators motivates me to keep going. It’s an inflated view, but nonetheless I thrive on this feeling to persist. I like to joke and sternl state that I do use this entire project to also improve myself as it gives me a utilitarian reason to poke my better friends into giving me tips and advice without them thinking less of me (though I’m not sure if that’s any more possible!).
My initial motivation when creating guides was my frustration with the random players I get matched with (we all like to judge others) and you’ll find that common frustrated motivation in many passionate people today. In turn, I also removed any frustration on myself and the project has pushed me into getting more involved in the competitive scene, the strategies and evolving metagame as well as watching many, many public matches from some of the highest rated players around the world. Even when I’m the least interested in Dota 2, either due to a losing streak, straying curiosity of other competitive genres, I stay in touch through these guides, with the scene and with the community. So as long as there is a dependency and interest in my work, I will stay involved.
The most challenging part of this project is two-fold: 1. Keeping up with the changes both in-game (balance-wise) as well as the metagame and 2. Maintaining the direction of the project; being restrictive in what items to include and not include (as you know, in Dota 2, there are so many situations for each individual hero that calls for specific items) and continuing to serve an audience that expands rapidly (roughly 2 million new subscriptions every 15 days).
1. Keeping Up – Guides can be as grueling as a typical desk job or as time-consuming as summer school.
This project has its ups and down, at the moment, with how the current patch is; there is constant task of keeping up with builds, emerging popular items and the coming of new heroes (Oracle should be released momentarily, two new guides to rapidly draft, publish and improve). A key thing to note about newly-released heroes is that the first guides to be released and receive a sufficient amount of ratings will also be the ones to establish the most subscribers and, subsequently, keep those subscribers no matter how wrong or misrepresenting they can be (from this contributor’s perspective).
A little more challenging and time-consuming are pre- and post-patch releases which can take up to 22 hours of elapsed work time in a very short period of days (3 days typically). As an example, the most recent 6.80 patch required an initial 20-hour work schedule across three days and then an additional 4 hours to not only receive and judge newcoming feedback, but to also follow gameplay changes via Dotabuff, professional matches, cross-matching with the top 10 public players per hero (thanks Dotabuff.com) as well as downloading the replays, watching live matches and talking to some friends who are at a higher skill level than myself to get their opinion. This doesn’t include the amount of time it takes to also create new guides for emerging new ways to play a specific hero that was previously not possible or remotely viable.
For patch 6.80, we only had to apply 370 changes and create 8 new guides. For a complete overhaul, which is an inspection of all hero guides, general descriptions, item descriptions and builds in correlation to the highest competition that’s taking place, The International, it can take several days to complete and over 2,500 rewrites, changes and note-taking.
In short, keeping up with the guides is a continuous affair that has its ups and downs, sometimes it can take up all of my time equal to a full-time job and at others, between those blips of updates, new heroes and a transitioning meta: it slows a lot down; consuming perhaps 4 hours a day reading feedback, watching matches/checking Dotabuff, and testing the builds myself in-game.
2. Key Framing – knowing your audience and trusting their prior knowledge is key to making accurate and to-the-point instructional builds.
Setting up how you want your guides to be perceived, how much and in what ways you trust your subscribed player base is key to achieving guides that are very fixed in expectations, but flexible in their range of information and abilities. Here’s a basic list of things I keep in mind both in detail to each guide and specific hero as well as overlying points about the audience accumulated audience.
+ Trust your audience and assume the appreciation of player’s choice:
We offer the idea of “Dual-Core” builds. Two core ways to play a hero that are different in approach and/or role (example: late-game carry Tiny and ganking Tiny). One may be the more standard and expected way to play but we don’t dismiss other also viable ways to play, especially if your allied team needs that role specifically.
+ Restricting the amount of information in a guide:
To keep things simple, I restrict the number of items per section of a guide (situational, extension, etc.), we also categorize them by price and typically priority and/or frequency of that situation arising that calls for an item.
+ Adding additional information that sets yourself apart from others:
Most subscriptions come from in-game, so they cannot see when a guide was last updated or other guides I have made, so we provide that information as a separate tab in the Suggested Items prompt in-game for users to be aware and interested.
+ Incorporating popularity within your target’s range of play while also leading them into better suggestions:
Another balancing act is both incorporating what’s currently popular for public players while also leading them into the right suggestions. For the image below, you can see Bane does not have the most typical items usually found in public games: Aghanim’s Scepter and Soul Ring. They are among the top items bought in the game according to Dotabuff. We instead opted for Urn of Shadows because at higher levels, it is generally much better for how Bane is played overall (through all phases of a match) than the advantages of, say, Soul Ring, early-game.
The image above shows Tiny’s item build split into two core roles to play the hero and Bane’s guiding direction in avoiding two commonly bought items in public matches.
Why does your title include which lane/area the hero to go rather than its role?
The initial reason was because I felt that was more relevant and useful for newcomers than stating how the hero should be played and the other is for the sake of conciseness. If I were to include a guide for every single role a hero could be played or is played by standard, I’d have about 200 guides rather than the current 140-150. With how time-consuming it is becoming to update guides, especially with each new patch, I would not be able to do it with the consistency and pace I can now.
As our audience expands, so becomes the flexibility and range of expectancies. What I set today as the standard and goal of these guides is not necessarily what it may be a year from now.
The overall gain from this work mostly stems from pride and self-fulfillment from having a historical dedication to a project for a long-time. I get a lot of satisfaction in being a part of something and consistently striving to keep its quality up to a high standard. There is no monetary gain and usually it isn’t talked about when I place it on my CV but nonetheless, it is simply an enjoyment for me, even during those times of chaos and time-crunched moments.
I’ve been asked on occasion about my interest to funding the continuation of this project through Patreon or Kickstarter or to simply allow donations. Unfortunately, it’s not something I am comfortable with doing. Despite how time-consuming and draining this is, it does not consume the entirety of my week. It’s still something I can do on the side while working full-time and I am very much qualified both in gaming and in eSports. When the project becomes too overwhelming on a daily or weekly basis, I may seek assistance (not necessarily financial), but at this present time, we’re doing al’right and the service continues to help new players get comfortable with Dota 2. Valve’s Hero Builds system is free and available for anyone to use to either create their own guides or to take my format and modify it to their own interests. Since the system is free and I’m more than happy doing it for free, I don’t feel relying on Patreon or Kickstarter fits into my moral compass nor current line of motivation.
To add, a lot of recurring community members and friends give a lot of their time to talk with me and give input/feedback and even explain some misconceptions I have. For me to take financial advantage of the situation would be to discredit their involvement and to place my own above all. It’s not something that’s fair when it’s not only knowledge/experience going into the guides, just the legwork (and obsession haha). Simply put, in the future, it something we may have to consider if it consumes all of my free time consistently but for now, you can support me by either rating the guides, providing feedback or spreading awareness to your friends you want to convert to Dota 2!
Lastly, however, if I could pitch ideas to Valve Software and collaborate during my free time to improving their newcomer systems to both Dota 2 (improving the guide system, workshop, new tutorials, etc.) and Counter-Strike: Global Offensive (I have a few ideas too as someone who’s at Silver Elite), that would be a real honor and something I am passionate about.
On a side-note, if you’re interested in my other pieces of writing or in myself, you can follow me on Twitter (@TorteDeLini), read my other articles and give feedback (ArmchairAthleticism.com)! Otherwise, continuously providing feedback and rating my guides (whether in approval or disapproval) is very much appreciated.
In short, thank you to those who trust me and even skeptics who constructively told me I’m improving (but still suck)! Information regarding changes made and the subsequent discussion following can be found on LiquidDota.com
- In-Game Standard Guide Project – Main Hub Topic
- To give feedback on any guides for improvement is best posted in that topic or via Twitter or the bi-weekly Reddit posts I make to show current growth of the guides.
Lastly, there have been a variety of users who have been supportive of this project and consistently provide insight, good or bad, that helped shaped these guides.
TheYango, Doomblaze, Sn0_Man, maru~, a slow decay, ChrisXIV, Synapse, Chaosquo, Comeh, Firebolt145, Whole, Pokebunny, Cecek, idonthinksobro, Tobberoth, LonelyCat, Coil1, LuckoftheIrish, Laserist, SpiritoftheTunA, Alurr, BluemoonSC, tehh4ck3r, Buckyman, Belisarius, SKC, CosmicSpiral, Laserist, Evilfatsh1t, Nevuk