Teamfight Tactics How to Improve Guide
I decided I wanted to do a different kind of post for anyone who has ever felt hard stuck at a certain elo. This isn’t going to be a guide on a specific comp or how to climb this patch. Instead, I’m going to give a bunch of tips on how to approach improving as a player. Hope you find it helpful.
- Become a student of the game:
- Study how pros play the game. TFT is a game almost purely consisting of decision-making, and you have a virtually endless amount of data to pull from when it comes to learning from the pros. Watch their streams. Look at their lolchess profiles. Focus on anyone whose play style you’re trying to emulate, anyone climbing super hard at the moment, or even someone in your game doing something you haven’t seen before and finding success. Figure out what units they play early, middle, and late game. Learn what items they prioritize, what they’re willing to slam so they’re not holding too many components, and what they never make unless they’re forced to. Analyze how they position throughout the game. These things, and so much more can be gleaned from watching a stream in the right from of mind. Most importantly, try to figure out WHY they’re doing all these things. Sometimes a streamer will explain their logic behind a certain choice in detail, but that’s the exception not the rule. If you’re watching a stream with your brain off, you’re missing out on a golden opportunity to learn while you’re being entertained. Also, if you’re watching someone just to learn, you should consider watching a VOD instead, watching on 1.5x speed, and skipping through to the points of the game on which you need to focus.
- Study your own games. If you’re not recording your games and watching the VODs from time to time, you’re ignoring super valuable information. I use the Outplayed app on Overwolf, but there are a ton of options for doing this. When a fight doesn’t go the way I expected, I quickly jot down the round and go watch it after the game ends. If my comp over or under-performs relative to my expectations, I’ll go watch the final 3-10 rounds of the game to try to figure out why. If I bleed a bunch of early game HP a few times in a row, I go watch those early games to see why. Is my board losing when it shouldn’t? Am I positioning poorly? Is my current impression of what’s strong early wrong? Is the entire lobby leveling when I’m not? Is that a good thing or a bad thing? These are a bunch of micro examples, but you get the idea. FWIW, it’s rare for me to watch an entire game because I’m almost always trying to learn about a specific thing.
- Study your own profile. The best site for doing this is Tactics Tools (you can get a lot of this info on a ton of sites though, so use whatever you want). There are some obvious things you should be paying attention to here, such as what comp trees/galaxies you’re good/bad at so you can both lean into your strengths and shore up your weaknesses. But there’s other super valuable data in here, for example your “contested” score will help you track whether you’re routinely playing an overly-contested comp and need to be scouting/pivoting more, your average level can help you diagnose if you’re playing too greedy or not greedy enough, etc. Your placement distribution can also be valuable – if you’re getting any particular placement a hugely disproportionate amount, that is telling you something about your play-style and is another potential area for improvement.
- Know your weaknesses (and fix them): There are essentially infinite ways you can improve as a player. The trick is to focus on that journey, not the ranking destination you’re hoping to achieve. You should be constantly identifying your individual weaknesses and always have 1-2 concrete things that you’re looking to improve upon. Some of these might be bigger things – learning how to play a particular tree well, learning to play the early game better, scouting every round, tracking match-ups, etc. – and some might be smaller things – executing a 4-3 roll down more cleanly, using/dodging zephyrs/shrouds, making sure your positioning protects your carries, etc. If you focus on these individual weaknesses and keep improving on them, you’ll continuously improve as a player and ultimately achieve whatever LP/tier goal you set for yourself. It will also help you take a bad series of games without tilting – you’re not throwing away LP; you’re learning.
- Know what’s strong: Particularly at high elo, the meta is changing on a daily basis even when the patch hasn’t changed. Don’t just play what was strong a few days ago. Spend 10 mins before you start playing by checking recent high elo matches in your region and pay attention to what’s doing well and what’s not. Then build a game plan for the day based on that information – what item(s) you want to start, what comp tree(s) you want to go, etc. (all of this should be based on galaxy btw). It’s also worth noting that knowing what’s strong in the end of the game can only take you so far – you have to become an expert on how to get there as well, which means knowing what units/items are strong at which times so you can play the best possible board throughout the game (unless you’re intentionally losing, which can be correct in some circumstances).
- Have a game plan: The galaxies were designed to push players into approaching each game differently, but I see a ton of high elo players who don’t embrace that reality. You should have a different starting item in mind based on what galaxy you’re in, along with a sense of what trees you want to lean. Did you get that item? Great. Go with your plan. If you didn’t, then you should know what you want next and how that changes your plan. You can’t prepare for every possible combination of things a new game of TFT can throw at you, but you can and should prepare for some of the most fundamental things (such as galaxy and starting item). Going in prepared will help you make better and more confident decisions.
- Think about trees, not compositions:
- Don’t tunnel on getting to an exact combination of late game units. Of course, there are optimal versions of a particular combination of units/traits and you can have that in mind, but you’re going to hit a hard skill ceiling if your approach to the game is to only buy/play what you perceive to be the perfect units (same concept applies for building items btw). If you train yourself to think about the game in terms of trees, you’ll start to use your brain far more actively in assembling your best board.
- What do I mean by trees? Let’s say you start with a belt off the opening carousel and then get offered 2 J4s – you should likely be leaning toward the protector tree. Some of you are reading this and thinking that means peeba comp, some are thinking it means 4 Prot/4 DS, some of you are thinking it’s 6 SG/4 Sorc/2 Prot, etc. What it actually means is ALL of them (including other iterations I didn’t list). Some trees are significantly more narrow in terms of your options (bang bros, for example), but they’re the exception not the rule.
- Put differently, you start the game at the very top of your tree by being able to play every S/A/B tier comp in the game, and then you immediately start to eliminate different branches of the tree based on what the game gives you. The units/items you get offered in stage 2 will help narrow you down to your best option, or even make it a single path in some instances. This process will continue throughout the game – the more units/items you get offered, the more you’ll work your way down the tree and into an actual composition.
- Learn to play everything:
- I’m ending with what I think will be my most controversial and misunderstood piece of advice, so let me try to clarify this a bit: there are metas where it is optimal to strongly lean toward a single comp and there are plenty of one-tricks that make it to the top 10 on the ladder. With that said, one-tricking a comp significantly increases your variance because there will be tons of games where you get a great opening for something, it’s just not the one comp/tree you’re planning to play. By learning every tree, you will be in a much stronger position to climb because you’re capable of taking what the game gives you instead of forcing a square peg into a round hole.
- Make sure you learn from the experts. When you’re learning a new tree, find someone who one tricks it (or just plays a ton of it) and study what they do. Don’t just learn the end game ideal – learn what they use as transition units, item carriers, leveling cadence, item prioritization, what a stable board looks like, etc. I learned Protectors from Treebeard, Rebels from Fluffy, Bang Bros from Socks, etc. They have more reps on these comps than I could ever get, so I try to short-circuit my learning by going to school on their hard-acquired expertise.
OK, that wound up being a lot longer-winded than intended. If some of this sounds more like work than gaming, that’s true. I recognize that’s not what everyone wants to do. If you just want to spam games, you’ll undoubtedly improve as a player simply from getting in the extra reps. But if you’re looking to optimize your climb, you can be way more intentional about it than just playing more.