Surviving 3k: Just Arriving

You've just made it to 3k MMR. Either you got your grind on and raised your MMR or this is where you, like me, were placed. The 3k MMR landscape is diverse in terms of talent, hero selection, and individual player strategy. It's not uncommon to have an ancients farming medusa go for midas and treads into rapier and carry the game. Don't fall into this trap, however. Here are some survival tips you can employ as soon as you start trying to climb the ladder to 4k.

• Don't jungle, unless it's a hero I highlighted here.
• If your team needs a support, pick it.
• If you're the hard support, I suggest getting courier, sentries, tangos, and a clarity (as per fluff)
• Always, and I mean always, play to win. Don't give up if you're behind 10 or even 20 kills. Games are swingy.
• Identify the best player on your team and ensure they have a good game.
• Don't ever farm ancients.
• Midas isn't a very good item this patch unless you get it before 6 minutes.
• When you get crushed by a single player, go back and watch how they crushed you.
• When you crush a team, go back and watch what enabled you to crush them.
• Help your supports with wards and sentries.
• Buy smokes and set up kills.
• Help zone the offlaner whether you're support or carry.

These are just a few general things to help improve your play so you can continue your progress in the 3k bracket. It's important to keep focusing on the basics of surviving 3k MMR because once you get them down, you can start focusing on playing like a 4k player. 

Surviving 3k: Encouraging your allies to pick around you

Every match, you should be doing certain things:

1. determining how your team is going to communicate
2. assessing how flexible your team is going to be in roles, picks, and lanes
3. trying to convince your team to pick strong heroes, counterpick the enemy, and pick around the hero you want to play

There are many other things, but these 3 you can determine right at the start of the game. For the purpose of this post, let's focus on the last part of point 3. If at least one other person picks a hero that synergizes with what you pick, you multiply the effectiveness of both heroes thus increasing your chance of winning.

Surviving 3k means you need to use all your skills, not just DOTA related, to get that coveted MMR. If you play a great hero x and you know you pair well with hero y, then encourage someone to pick it if they're good with the hero. A few notes about this approach:

• Try a logical argument. for example, if the enemy team has a Storm Spirit and you want to play Omni, you could try: "Antimage counters storm and will play great with my Omniknight because you won't have to get a bkb."
• If the logical argument doesn't work, then try an emotional one. Generally, people make decisions based on emotion; how a decision makes you feel impacts greatly what you decide to do. With that said, most people will respond positively if you ask nicely, but some might respond if you are more aggressive. Caution: people tend to hide behind their anonymity on the internet so being aggressive isn't usually as successful as asking nicely. "You catch more flies with honey than with vinegar" is the old adage.
• Sometimes, no one plays a hero that partners well with you, like you want to play LC but no one picks heroes with burst damage.
• When some one says no to a line of reasoning, try another; sell your idea any way you can. Always keep in mind that the goal is gaining MMR, not making friends or enemies. If you happen to make those, that's fine too, but this is solo queue not social hour.
• Don't constrain your hero pairs to one that will only be in your lane or play a certain role. Try to harness the strength of your team's abilities using snap judgments and making quick decisions. If those decisions don't work after a couple tries, move on to something different.
• Don't get emotionally involved with a hero, hero pair or strategy. Being involved with your significant other is fine but this is, again, solo queue. There is work to do and whispering sweet nothings into your microphone isn't going to get you that +25 at the end of a match. If the game calls for something outside your comfort zone, do it. 
• On the other side of that coin: if your team wants you to pick a hero you aren't good at, try to find a compromise hero, otherwise tell them to pound sand. You are responsible for your destiny. If you're going to make it to 4k or 5k or whatever your goal is, you have to do it. Picking Techies with Tusk may sound like a good idea, but if you haven't played Tusk before, you will be a burden to be carries by someone else.

How do you determine what heroes to pair with? Make sure to check your dotabuff and see what heroes you pair well with. You should be spamming between 3-5 heroes in solo queue (another post for another time) and will therefore build up stats related to those heroes. With those stats you can make more informed decisions about playing the game, include what heroes you pair well with.

Surviving 3k: 3 things to work on to ensure you survive

In a recent interview, one of the pro players that the best teams are thinking 6 steps ahead. To think like this you have to first practice thinking one step ahead. Then two steps. Then it becomes easy because steps past step two depend on what you predicted steps 1 and 2. 

Most players get stuck on steps 1 or 2 because they:

1. Lack enough reference experience to understand the limits of the current game state
2. Not actively thinking about what's going on in the game
3. Have to focus on some of the basic skills in the game rather than some of the bigger pictures

Lacking reference experience

If you want to survive 3k MMR, you have to try to push the limits of your games in order to gain reference experience. The player who does this the best is Eternal Envy. He understands what the limits of the game and can see the end game scenarios when he's behind. He thinks about what the enemy team needs to win and then tries to deny them that. How does he know all this? He has thousands of games to compare his current situation with. If you want to understand what you can and cannot get away with in a game, try different things and watch your replays and your timings for when things worked. EE-sama also is continually actively thinking about the game.

Not actively thinking about the game

When you queue up a game, make it your focus. Put the cellphone down and stop alt-tabbing. I know it's hard but focus your attention on your game, your team mates and your enemies. Your brain is a muscle and if you can exercise it like any other muscle. The more you exercise it to alt-tab and use your cell phone, the less you are exercising it in DOTA. If you want to survive 3k MMR, you need to exercise that brain for DOTA. 

Focusing on basic DOTA skills like not using hotkeys, queuing actions, and buying items is preventing you from focusing on other important aspect of the game. Hitting creeps should become zombie mode at some point. You item builds on certain heroes should become almost automatic and others should be familiar. At the very least, you should be able to understand what items do and what you should buy to itemize against your opponents correctly.

Hopefully these three things will help you survive 3k MMR.

Surviving 3k: Making the best of your picks by predicting lanes

Radiant

VS

Dire

I'm sure many of you have experienced a game like I experienced on Radiant a few days ago:
- some one randoms a hero first pick (Tinker in this case)
- the Windranger is set on playing mid
- everyone else picks hoping the Windranger isn't going mid.
- the Windranger refuses to switch lanes

How should you handle this situation? there are many ways to approach it, so lets run through some of the options.

1. You can try to bully the Windranger into going to another lane through having the Tinker or another hero go mid with her. There is a Tusk, so depending on how the dire lane, and specifically send middle, there might be kill potential for a roaming Tusk who spends a lot of time mid.

• Pros: shutdown enemy mid, potentially snowball (no pun intended) the Windranger into fast Aghs
• Cons: The Windranger ruins your game because she's mad, you don't get kills on the enemy heroes because they lane strongly and end up with under-farmed and -leveled heroes

2. Aggro Trilane with the Tusk, Nyx, and Ursa, send the Tinker solo safelane against, hopefully, a Darkseer, and let the WR do her thing. Snowball into Nyx stun plus Ursa damage should be a kill the majority of the time against a solo or dual lane.

• Pros: Crystal maiden is an easy support hero to snowball and kill (low movespeed) if they solo support. If the Razor safelanes and gets shut down, the carry potential of the other team is drastically reduced, especially for pushing. Happy WR. Tinker gets some farm and levels.
• Cons: fighting all the time in the enemy safelane means that your heroes will potentially be low, powering up the Bloodseeker. The risk of him snowballing out of control and taking over the game means you will probably lose. Also, Nyx and Tusk need levels (most importantly, level 6) to be effective. If your aggro tri doesn't get kills, you end up with under-leveled and under-farmed heroes who have little catch up potential.

3. Put the Ursa in the jungle and have Tusk or Nyx support the Tinker. This is the greediest option because you're allowing the offlane hero to get more level and experience than he should get. This will be especially hard for a Tusk or Nyx to zone a Darkseer because Ion Shell is so strong early in the game and those heroes are both melee.

• Pros: the enemy team has a potentially greedy lineup so trading a jungle Ursa for a potential jungle BS is a good trade (Ursa matches up well against the BS). Happy WR. Tinker gets some farm. Offlane gets levels.
• Cons: jungling against a BS takes a lot of finesse; the lower the Ursa gets, the faster the BS jungles closing the gap in advantage the Ursa has. Having either Tusk or Nyx solo support isn't ideal, especially against a DS, If Nyx does support, his impact on the game is minimal. You aren't punishing the enemy team's greed in any way.

While there are other option that aren't mentioned here, a lot of it comes down to predicting your opponents lanes. Because people are playing so greedy this patch, it seemed obvious to me that the CM would support top with a little jungle mixed in, the DS would be offlane, and that Razor would be safelane. This leave Lina and the BS to either go mid and jungle or support and mid, respectively. I predicted, correctly, that the BS would jungle and Lina would mid so I convinced my team, apart from the WR, to go with option 3 and putting Nyx in the offlane. Here's how it panned out:

This game is illustrative of how to make the most of your picks in pubs through predicting how the other team is going to lane. Even if you don't initially predict how the other team is going to lane, just switch up your lanes to give your team the best chance of winning. For example, if the enemy team had gone for a CM, Lina and Razor aggro Trilane, then the Tinker would need to move to the top lane in order to get his farm. Otherwise he risks dying to the three enemy heroes. It's also worth mentioning that a trilane isn't the only way that the enemy could have punished the random Tinker pick. But they didn't punish us so we won.

When it comes down to it, DOTA is about getting the enemy thrown. It doesn't matter how you get it as long as you do. Making the best of your picks for the laning stage is the first step in doing that. The next things to consider are making the most of your picks in mid-game and securing an end-game.

*Special thanks to www.dotabuff.com for the pictures found in this post.

Surviving 3k: just counterpick them already

The pub meta is driven by the most popular heroes. While win rates generally correlate to popularity, Pudge is super popular but his win rate is about 50% so there isn't necessarily a general causation there; some people just have favorite heroes they like to play alot. Surviving 3k MMR has a lot to do with using picks to compensate for the wide range in skill. A lot of people resist the idea of picking hero flavors of the month and counterpicking those flavors because, again, some people just have favorite heroes they like to play alot. Well that needs to change. Stop picking your favorite heroes and counterpick your opponent already.

If your goal is to get better at DOTA and improve your MMR, then counterpicking offers a direct path to do this. Counterpicking nudges who has the advantage in your favor. Even though Wraithking may not be your jam, he gives Bloodseeker the third lowest win rating so picking a WK into BS is going to give your team a few percentage points in your favor. This is crucial. Games are generally close and the higher you get in MMR, the closer they become. The closer games become, the more important your hero decision becomes because it offers a way to gain an advantage.

How do you counterpick properly? I usually just go to dotabuff heroes section, find the hero I keep losing too, and see which heroes are recommended against them and which heroes give them a lower win rate. It's that easy. It's also useful to know how they generally itemize. Items like black king bar or butterfly can swing in their favor if you aren't prepared.

Personally, I have been having trouble with Bloodseeker. He's everywhere and has a really high win percentage. Having explored different strategies, Ursa has worked well for me because of damage mitigation and the burst damage Ursa offers. Plus he's aggressive around the same time that Bloodseeker is. You don't play Ursa in the jungle against the BS.

What heroes are you having trouble with? How are you trying to deal with them? Let me know in the comments.

Surviving 3k: Finding the Niche

The most common question among pub players is: how do you raise MMR? Some people say to spam this hero or that. Some say to jungle for 10 minutes and come out with the intention of of snowballing your team to a victory. My advice and what has worked for me is to find a niche and fill it.

Lets use some abstraction to understand this.The pub scene trends toward certain picks, strategies, and styles of play. Because the pub scene favors certain heroes, strategies, and styles of play over others, this creates a deficit somewhere. Sometimes this deficit is well founded and sometimes it isn't. Right now, when it comes to support, it isn't. The pub scene at 3k is lacking the support duo currently (Fluff gives a good talk about this here). Rather than having a hard 5 and greedier 4, there's an awkward 3.5 and 6.5 position hero. This is bad.

Most importantly is that no one wants to play the 6.5 position because they are getting sacked so hard. Buy all the wards and the courier and share your tangos. Forget about getting brown boots until minute 12, buy all the smokes and it's your fault for not putting up enough wards when some one on the team gets ganked. Moral of this story is help your supports through buying wards, sharing tangos, upgrading courier, buying smokes and not flaming them.

The second most important point is that having a 3.5 is a liability, most of the time. There are small exceptions. However, the majority of the time picking a greedier support will help your team out way more than having a 3.5 because they will be active on the map and help buy support items during the beginning of the game. The goal of the first 10-20 minutes in a game is to win the laning stage and transition into the mid-game with a lead. The best way to do this is to have a dynamic support duo who matches up well against the enemy team.

Now I know a lot of you out there are going to still play a 3.5 hero, so here's my recommendations:

Enigma - fastest farming jungler there is. Can push early, gets an early mech, and has great team fight ulti.
Chen - maybe the second fastest farming jungler there is. Pushes early, gets mech, and has a great team fight ulti.
Ursa - everyone knows your going for Rosh by 7 minutes. After Rosh, it's time to get aggresive
Bloodseeker - Global ability that powers you up and gives vision when people are low seems like a good deal.
Crystal Maiden - getting a quick level two and ganking the mid lane is how I like to play her. Just make sure to get quick boots because you are slow. Your aura helps your team win the laning phase.

these heroes get active before the 8 minute mark and create pressure on the map naturally during the laning phase. They all have things in common: they pressure the enemy team in ways traditional heroes can't because they don't show on the map until it's too late.As a 3.5 position, you need to help your other support. Like it or not, you need to compensate for your lack of direct laning presence by empowering your other support.

In current pubs, there's a place for a 4 position support or a super active 3.5 position hero. Trying to pick a sub-par jungler like Legion Commander is a waste of time and a pending loss. Fill the niche your teams needs and be rewarded with MMR.

More Efficiency

I go to school for electrical engineering and am on course to graduate this May. A topic often covered and talked about in engineering is efficiency. The efficiency of your engine, dispersing heat and processors are all important to understand for their applications. For example, your general - and I mean general, not just internal combustion - engine can only be about 20% efficient because of the inefficiency built into the system from things like friction. Don't be dissuaded from using an engine, however, because we are transferring one form of unusable energy into another, more usable one. In DOTA, efficiency is also extremely important; allocating damage, farming, purchasing items, movement, et al contribute to the efficiency of a player.

The more efficient a player is, the more impact he will have on the game. This is because he is optimizing his outcomes from each decision. The easiest way for me to visualize this is to think about using town portal scrolls (TPs). When used correctly, TPs allow a player to move about the map more efficiently because it takes about 3 seconds to port from top lane to bottom lane instead spending a long time walking and risking being caught out of position. Still, TPs are only so efficient like the engine because they cost gold to use. Thus a player converts gold into less travel time and a lower risk movement.

The contrapositive is also true: the less impact a player has on a game, the less efficient that player is. The carry or 1 position players for each team illustrate this the best. Generally, the carry which is more efficient and able to accrue a bigger advantage over the other carry will take their team to victory. Another way to think about this is to visualize two Anti-Mages: one with level 4 blink and the other without blink. The AM with level 4 blink will be more efficient at all aspects of the game - except, perhaps, mana usage - than the AM without because of the advantage blink offers.

Often players get hung up on efficiency, causing them to become inefficient. Don't let this be you. When considering the cost of a decision, it is often the case that there is an asymmetry of information; it is literally impossible to know all possible differences between and outcomes of two courses of action. Don't let this hinder you. Many players will get hung up on item choices, missing stacks, missing pulls, taking unnecessary creep damage, not spending their gold before they die and the list goes on. All these contribute to small game inefficiencies but psyching yourself out will contribute to large ones because not all inefficiencies make meaningful impacts on games. There is a caveat: the higher the skill game, the more those matter and can snowball into a game loss because players are better at leveraging small advantages created by these inefficiencies. With that said, psyching yourself  out is not allocating your brains processing power efficiently and can lead to tilting: the definition of poor and inefficient decision making.

The next time you find yourself looking for an aspect of your game to improve, consider efficiency. Look for ways to minimize wasted gold and movement then move on to things like allocating damage in team fights or stacking stuns on an elusive target. By improving your efficiency, you will help your team and, hopefully, translate more games into wins.