I’ll make one thing abundantly clear right at the start: captaining is, by a significant margin, the best way to improve your odds at having the enjoyable experience you envision in the moment you sign up — and this is chiefly because it allows you the opportunity of choosing exactly who you play with.
Anyone who disputes this does so because they either don’t care to influence that experience, and are instead in it to roll the dice — which is a perfectly acceptable decision — or because they suck at drafting, though they may claim otherwise (the silliest excuse being that captaining is too time-consuming, a take usually coming from people who spend that time shitposting on Discord anyway).
The aim of this writeup is to demystify the draft a bit by exploring how experienced captains approach it, and allow you to become level with them by noticing and employing the ideas they make use of.
Yes, I may harbor some ulterior motives here, given the struggles I’ve had with finding enough captains as a Clarity admin, and RD2L admin before that — but I’ve captained and drafted teams for much longer than I’ve admined these leagues, so you can trust me on that front.
The amateur Dota league auction draft opens up a whole new world to the drafter in comparison to the traditional MMR-based linear draft (oft referred to as an EU draft). While you can certainly develop strategies more complex than the usual in those formats, you’ll generally be fine going into an EU draft with the mentality of grabbing the best available option when it’s your turn, with the extent of the strategizing boiling down to avoiding having to pick in an unfavorable place in subsequent rounds (which is really just a nicer way to say, pick the 6k instead of the 7k, so that you can later pick the 4k instead of the 2k). You’re competing against yourself, and against the pool of options.
An auction draft, on the other hand, has you pitted against all other drafters. Everyone has a shot at grabbing everyone; knowledge of individual players’ relationships, thorough planning and active adjustment of your approach all improve your chances. The plentiful limitations of a linear draft vanish into thin air…yet still, every single draft has a plethora of massive gaffes that can range from self-grief to season-ruining for multiple players — or even multiple teams. Throughout this blog, I’ll be breaking down concepts that you need to understand and learn to apply in order to come out of the process with a team you’re happy with.
I’ll make a conscious effort not to dwell on points outlined in the (linear-draft centric) captaining guide I wrote a while back, so if you’re really itching to read I’d recommend going over that first. It does a good job of elaborating on the preparation process and general tips to keep in mind for any draft. I’ll make no promises on the structure of this guide — nor its brevity — as it’s likelier to end up as an assortment of tips, or a protracted rant, rather than a step-by-step.
The Basics
The vast majority of auction drafts follow the same procedure. Drafters (in the form of team captains) are selected, coin budgets are allocated to them and adjusted based on a variety of factors (chiefly their MMR), and a nomination order is randomly generated.
The draft moderator guides the process by giving drafters the green light to nominate a player. The first person in the nomination order names a player from the draft sheet, and all drafters are then allowed to bid on the nominated player using the coins they’ve been allocated.
When bids slow down, the moderator starts a brief countdown, and upon the conclusion of that countdown, the purchase is finalized and the nominated player is assigned to the highest bidder’s team, the cost of the player being deducted from the drafter’s remaining budget. Rinse and repeat until all teams are full.
The major caveats to note are largely with regard to the scenario that one or more drafters exhaust their coin budget, while having open roster slots. When that happens, the drafter has to wait (their turn to nominate being skipped) until either all other teams are completed, or all other incomplete teams also reach 0 coins. If only 1 team is incomplete, they get to simply choose players in the pool to add after all other teams have been finalized. If multiple drafters have run out of coins before assembling a full team, they will take turns adding players to the team. The order in which they do this is the inverse of the order they ran out of coins in. While this is a somewhat rare occurrence, let’s illustrate what happens with an example.
Drafter X ran out of coins first, after spending their entire budget on two players. They need two more players. Drafter Y ran out of coins second, and needs one more player. Drafter Z ran out of coins third, and needs two players.
All three will first wait, skipping their nominations, until the other five captains have completed their rosters in full via the auction system.
The draft then enters round 1 of a serpentine draft. Drafter Z gets to choose one player, follow by drafter Y, and then drafter X. Team Y is now complete. In round 2, this order is reversed: drafter X picks first, followed by drafter Z. All teams are then completed.
Now that you understand everything there is to know about the logistics of the draft, we can move onto the interesting bits.
Preparing for the Draft
Prior guides I’ve written on this topic focused more on the actual legwork of preparing — copying sheets, editing them to your liking, etc. This tends to be a worthwhile process to go through—particularly if you’re new to the experience and don’t feel confident in your read on the player pool — but I’ve moved away from these steps in my own preparation, largely because it’s a lot of work to put into creating a plan that is certain not to pan out the way you imagined.
The major benefit of doing this is being able to reference ideas you have, but for most purposes you can get by with, like, Notepad. I’m also a fan of handwriting notes, but I’m pretty sure I’m mostly alone in that (though I’ll note the advantage that this leads to one less window or tab needing space on your screen come draft time). But, do whatever feels best — your goal is to have a rough idea of what you’d like to accomplish, but how you do that isn’t imperative. I’ll refer again to my old guide, which has examples of how to use a copy of the draft sheet for prep.
Before any of this is even relevant though, I’d warmly recommend figuring out what you want out of your experience and from your eventual team. This might be confusing, but bear with me.
What do you want for yourself in this team?
Are you set on playing a specific role, or are you open to playing multiple roles, depending on the player draft?
This decision will instantly shave off some chunk of the player pool, but that shouldn’t put you off of playing the role(s) you want to play. If you suspect you’ll get tired of playing off-role in a month's time, just… play your role. You’re captaining to get the experience you want.
Do you intend to practice things for the benefit of the team on your own time?
Is the idea of committing more time to the team outside of scheduled matches off-putting? This factors into reaching multiple conclusions. It ties into the previous point: if you want to use your time pubbing practicing for your officials, are you down to play those pubs in the roles you decided you’d play on the team? Additionally, if you want to scrim and practice with your team, it becomes incredibly important to get a team that’s down to do so — preferably without much resistance.
The best way to go about this is usually to ensure that at least two of your players are similarly enthusiastic, and the columns of the draft sheet that show a player’s willingness to practice become a much more important part of the equation. It’s a very good practice to also message these players and reaffirm that they’d want to do that. Captains who are open to practicing also have an easier time getting extra value out of first-timers, who may need that additional time to get into the swing of things (more on this later).
Are you down to draft and/or shotcall?
This can massively impact what you focus on in creating a plan. If you don’t want to draft, you need a drafter. Don’t fall for the “we’ll do it as a team” propaganda — if you don’t have a single person willing to do the actual button clicking on the pick screen, you’re at elevated risk of having a bad time. This doesn’t necessarily take people off the pool of options, as much as it bumps players you’d like up in priority.
Though somewhat less important, the same applies to shotcalling. This isn’t to suggest that shotcalling in itself is unimportant — the ability of you and your teammates to communicate well and craft plans is decisive — rather, that the comfort someone puts down for shotcalling on their signup isn’t necessarily as meaningful as it is for drafting, seeing as it’s a bit less tangible as far as self-assigned rating metrics go. Often enough, someone’s ability to be the leading voice in a team will depend significantly on the overall atmosphere in the team and the general contributions of the other players to the comms. While there are players around who make for confident and consistent shotcallers in just about any team, they’re few and far between.
Are you in it exclusively to compete for the win, or does having a friendly team you can have an enjoyable time with matter more?
While these things tend to go hand in hand, reviewing your own motivations for participating can sway your approach to the draft. It can be as simple as foregoing a higher MMR player you know less about, to instead draft a lower MMR player who you know is a positive presence in a team. Making a distinction here can highlight the importance of doing extra research on your targets, most notably in the form of season reviews — after every season, some amount of participants write reviews describing their teammates and experience, and searching up a potential teammate’s name in the review channel can yield more info. Notable also are the cheat sheets, community-made rundowns of participants, though these tend to be focused primarily on being funny, rather than informative.
A Note on Roster Construction
There are a couple other ideas to consider that are in the same vein, but they largely depend on the player pool — things like whether you might want to splash for a big name, high MMR player, versus constructing a more balanced team, but that’s a concept we’ll explore later.
One thing I’ll touch on here, since I don’t have a better section to put it, is where my own preferences have landed after a couple dozen teams. When I have the choice, I somewhat formulaically default to drafting a couple veteran players I’ve either enjoyed playing with before (or suspect I would enjoy playing with), and pairing them with players that I think are undervalued by other captains. The dynamic is obviously a bit different when you’ve played with as many people as I have, but the core idea is to create a team that has some degree of familiarity — both with one another, but crucially, with how teams in these leagues tend to operate.
The less time you have to spend working through how to communicate with each other (as opposed to the song and dance associated with creating a comfortable dynamic between people interacting with each other for the first time), and establishing an understanding of how the league works, the sooner you can move to improving as a Dota team. Accomplishing that doesn’t necessarily have to depend on knowing or being friends with people — you can often pick up on who might be a good fit just from the way they interact with the community.
Prioritization
If you went through the above process, you’re now left with an idea of what kind of players you want. More importantly, it’s allowed you to scratch some names off, so you’re left with a pool of options that is easier to consider individually.
What you’ll want to do now is locate anchors: players you have a keen interest in, and around whose acquisition you can plan your draft. This is as much to get an idea of what you might want to do — actually visualizing for the first time the composition of your roster — as it is to prep for the actual draft itself. You’re essentially looking to create a profile of players you’d like from a given pool, and then checking how many options fit that profile.
In general, a lot of this is based on scarcity. The most common anchors tend to be core players, but if you’re working with a pool of about 40 to 60 players, you’ll inevitably note gaps in the role spread. When I talk about these gaps, it’s in reference to the number of teams in the division, which is 8 by default — and when you sit down and look at the top 40 highest MMR players available for your draft, chances are that there’s not gonna be 8 of each role. So — take note of where this scarcity exists, and review options.
In my experience, there tends to be, for example, a decent number of offlaners and carries, but a limited number of strong mid or pos 5 options, so those become a higher priority. This is very different between MMR brackets, and thus between Clarity divisions: lower divs might instead lack in solid offlane options, etc. It’s hardly a consistent trend, and things tend to change every season.
When you review your priorities and then compare them to the actual player pool, you’ll generally get a very solid understanding of what to emphasize.
Looking at a Dotabuff
Reviewing a player’s profile is pretty crucial, infinitely moreso when you’re not already familiar with a significant chunk of the player pool. In my mind, the primary objective of a Dotabuff check tends to be scanning for red flags.
- Their role spread on the sheet is 4/5/1/1/1 and their profile shows 80% offlane or support? Ew.
- Their activity over the last 3 months is a wasteland? Ah ooh yeah probably not.
- You scroll down to their matches and are met with a hundred Turbo games? Zoo wee mama that’s a dodge.
- Overall winrate is sub-50? Yeah no thanks.
- A massive chunk of their total matches are on one hero? Useless player in Captain’s Mode.
- They’ve got over 10 deaths in the vast majority of their games? I’m good.
Equally, you can also check for green flags.
- They play their role — the 1/1/5/1/1 player has 85% offlane, that’s the good stuff
- Their hero pool is wide and relevant — when I check their hero list for the past 6 months, I want to see bare minimum 7–8 heroes in double digits matches played, all in their role
- It’s nice to be able to click on the Esports profile and seeing some number of matches played, though that’s not necessarily a dealbreaker if not
- Give me above 50% on most of their most played heroes, overall winrate and ranked winrate as well, if we’re already making requests
- Ancient player but used to be ranked immortal and thus has the bugged badge still showing their old rank? Value alert brring brring
Now, don’t get it twisted — having some of the red flags doesn’t guarantee they’re an awful pick, and you might not even necessarily see them as being red flags; equally, the green flags hardly mean they’re guaranteed to be a good pick. These are just things to look out for and note, and can help you differentiate players of similar profile and subsequently choose one over another when establishing your hierarchy of priorities.
For the keen researcher there’s also bound to be other things you can check on Dotabuff (or Stratz, or Opendota, or even ingame). It’s purely a matter of what stat is valuable to you, and how much time you’re willing to spend on doing these profile reviews for upwards of a dozen players.
Now that you have some ideas swirling around, we can talk about doing the actual drafting (during which we’ll also cover more ideas on choosing your targets). It’s a fluid and dynamic process, during which you’ll have to adapt and often change plans, so we want to be set up to allow that on the fly decision-making to be as smooth as possible — both to make it easier on you, but also out of courtesy, since you don’t want to be the person holding up the draft for everyone else.
Prep your drafting resources
My preferred setup when drafting is to have the following open:
- The MMR sorted player list on the Draft Sheet — you want to see the list of available options, not only for your own sake, but also to understand what everyone else is seeing and considering. Also allows you to very easily glance at a player's Dotabuff profile.
- Division tab / team list on the Draft Sheet — on top of showing you your own remaining budget, you can track what other captains are working with, which is immensely important.
- Discord — both to participate in the draft (duh), but also to DM players, be that to ask someone about their roles or interest in playing with you, but also to consult players you’ve already drafted and get their opinions.
- (optional) Your own prep documentation — whether it’s a random assortment of notes, a flowchart, whatever, you want to have it on hand. What I used to do was clone the MMR sorted list, categorize players into tiers by coloring their rows on the copied sheet, and then cross them out when they’re selected. This would let me reference, for example, how many players marked as high tier remain, which roles are running out of preferred players, etc.
You’ve got your little sheets all set up, and now we bid.
Nominating
When you’re up to nominate, you generally want to make some use of it. There’s a couple of standard ways to do so; the most notable of those, in my eyes, being nominating a high MMR or otherwise highly valued player that you don’t want to draft. I look at this as a way to drain others’ budgets — get them to spend as much on players I don’t plan on contesting, so that there’s less competition (both in terms of role/player slots, but more importantly in terms of coin count) when you get to contesting the players you actually want to draft. Depending on the flexibility of your preferences and plans, keeping this up for multiple rounds can leave you with a great deal of control in the draft due to your budget.
The most straightforward way to make use of your nomination is to simply name a player you do genuinely want. Ideally, it’s a player you’re fully set on drafting — you don’t care about overpaying, you just want this player on your team, and — importantly — don’t wish to risk going for them in a later round with a diminished budget.
Alternatively, it’s an okay option to nominate a player you want, but wouldn’t be heartbroken about losing out on if the bids go past what you’re ready to spend (for example, they play a role for which you have multiple alternatives). This is fine to do particularly in drafts where you feel like too many people are being stingy with coins in the early rounds — if you look back at some drafts, you’ll notice that people can get caught up in their desire to hold onto coins for big names, and let valuable options slip through for much cheaper than they’d cost 2 rounds later.
Another thing you can do is specifically target a player someone else wants. Naturally, you’ll need to have some knowledge on the dynamics of social circles and whatnot, but it’s often quite blatant. Your goal again is to drain budgets, albeit in a way that messes with one particular person, forcing them to either spend coins early on, for example, on a lower MMR friend they might’ve hoped to pick up for free later on — or to maximize the number of captains with that are willing to tax that specific player for that specific captain (taxing referring to inflating the price of a player by abusing the knowledge that someone desperately wants to draft them). You have to be very cautious here, though, as you can easily bait yourself into this nomination, only for no one to contest it, leaving you with a player you didn’t necessarily want on the roster — so, ideally, this is a person you wouldn’t mind having if they’re going for cheap.
Tangent 1 of Many: On Peer Pressure
On that note, generally don’t feel pressured into participating in taxing if you’re not comfortable. For every three or four successfully taxed players, there’s one disastrous outcome — one captain sad they overpaid for someone they didn’t really want, another sad they didn’t get their friend, and a player now straddled on a roster they don’t want to be on and aren’t wanted on.
While we’re talking about this sort of peer pressure that might exist in drafts, I’d like to touch on the idea of captains collectively having an unspoken expectation of not letting egregious underpays slip. It’s an immensely common bit of discourse, usually materializing as disbelief from the community that “the rest of the captains let that purchase happen”.
This isn’t particularly problematic, insofar as I can promise you (as the resident draft critic) that no one actually cares. It’s just something people say — but you’re not bound by it in any way, which I highlight because I tend to notice some drafters allow the idea to creep into their process.
Sure, okay, another captain is getting a good player for less than they probably should, and you’re one of the only ones who might be able to contest it, but if it doesn’t make sense for your draft to do so, just…don’t. Yes, you’re allowing another team to become stronger, but you need to have some faith in your own plan. I’ve been in this position many times, and sure, once or twice it did allow another captain to put together an absurd roster that did then go on to demolish the competition, but far more often it’s completely forgotten once the season gets underway, and that team performs at par.
It goes without saying that this doesn’t mean you should never deviate from your plans. Let’s talk about that a bit.
Adapting
Your early draft plans, ideally, are somewhat realistic. I can get this great player, but my budget will vanish into thin air, so I’d rather plan around getting this lower MMR player and then allocate more money to this other role!
That train of thought is all well and good — but it might lead you to ignore that initial great player, and give up on them before you even see what the bids look like. Without exception, every single player draft will see overbids and underbids. The former will lead to the latter. Crucially, you want to be the person getting these bargains — you simply shouldn’t go in with the expectation of getting them…but a player you expected to go for 60 coins and thus wrote off, happening to go for 35 instead invites a bid, and it’s usually not an invite you should reject. I’ll emphasize again that this needs to be a player you actually want, though — and when you’re switching up mid-draft, it helps to have your set of backups for each role be from different MMR brackets.
On the flip side, you might also be faced with less…positive reasons to change your plan up. Your targets are going for way too much, you’re scrambling to figure out what to go for instead, and before you know it, the remaining player pool looks sparse and you’ve got many roster spots to fill.
I tend to play drafts fairly slowly. Let people use their coins, pass up some of the marquee names, try to snipe multiple good value picks. This is the easiest way to end up with consistent drafts the likes of which I have a reputation for crafting, but I’d be lying through my teeth if I said it never resulted in panicked moments of realization. Haha, you fools, I have all the coins!…And no one to spend them on. Suddenly it doesn’t matter that you’ve got the 17th player on the list for 15 coins, because even though they’re the 4th highest mid player there, that’s 3 other mids that’ll wipe the floor with them.
Nevertheless, you have to make the best of this often bad situation. You should have the ability to confidently control bidding on the best remaining players if you get to this point, so, yknow, do that. Alternatively, if you really don’t believe in the stragglers that’ll be left over, sometimes you just bite the bullet, and pay more than you expected to for a good player, then make do the rest of the way. The greater your understanding of the player pool is, the easier it is to make this decision during the draft.
Tangent 2: On Getting a Bit Too Clever
A lot of captains tend to try and get clever with drafts. When they’re good drafters, that’s usually in the form of keeping keen track of budgets and role spreads. There’s only a couple good 5s in the pool, but a lot of the captains are 5 players, so I might get good value on trying to nab one of those players early, while the other non-5 captains are fishing for a 1/2.
Or, okay, I’ve got two slots open, but in this role I’ve got lower MMR options that I’m okay with that probably won’t get attention — I can identify the best option for the other role, and make a bid that others can’t match (for example, you have 35 coins, the other person looking for that role has 28 — just bid 28 and you’ll probably save yourself a coin).
Other times, though, the clever trick they’re pulling isn’t all too clever. While there’s a plethora of examples in linear drafts, the most common mistake I see in auction is captains going through the following thought process:
It’s early in the draft. I know that particularly lower MMR players can sometimes go a bit cheaper early, and spike in prices towards the end as captains scramble for picks, and thus they won’t contest me nominating the 38th highest player! And then you end up paying 11 coins for what would’ve otherwise been a 3 coin player. Oops.
It happens every season. Don’t fall for it.
MMR is not just a number.
But also, it is.
Look, I’ll level with you. It’s really hard to be a great drafter without knowing the player pool.
For all intents and purposes, this is tangent number 3 incoming: MMR is an immensely flawed metric for judging how good someone is at Dota (most crucially because for it to be a serviceable metric, everyone would need to play very significant amounts of ranked).
It just also happens to be the only one you have to work with. The staff make efforts to level the playing field a slight bit here, in the form of MMR adjustments. The main purpose of these is to highlight the subjective, but usually fairly grounded, discrepancy between a player’s current MMR and what their level of play might actually be, based on their past performance. For the most part, a lot of this is informed, even if not always consciously, by how known players of x MMR stack up against other known players of x MMR. It’s thus generally smart to assign some credence to the adjusted number over the original one.
But adjustments are done only for a handful of players, and they’re only ever done upwards — so you’re still going to have to figure some things out yourself.
The point I want to get to here is, essentially, don’t get tunnel-visioned on one view of MMR. I see new drafters often fall into this trap of assigning either too much or too little importance to MMR — there’s probably a reason why any given higher MMR player is being passed up on by a bunch of other captains, but equally, your ragtag squad of players all sat around or below the average MMR of the division probably isn’t going to shape up to be the dark horse you’re expecting, just because they all play their roles and have a nice looking Dotabuff.
Invariably there will have been worthwhile ideas and concepts that I didn’t mention; after I publish and discuss this, I’ll update it on occasion to include things missed.
Happy hunting!
