Almost nobody is stuck because they are short of advice. You already know you should last hit better, ward more, die less and play fewer champions. You have known it for months, and you are in the same division you were in at the start of the season.
What is missing is not another list. It is knowing which item on the list is yours, right now, and being willing to ignore the rest until it is fixed.
Practising the wrong thing is not neutral. Spend three weeks drilling last hits when your real problem is dying twice before fifteen minutes, and you finish those weeks with better CS, the same rank, and a quiet suspicion that improving does not work for you.
The players who climb past you are not working harder. They are working on fewer things, and on the right ones. Choosing correctly is a skill in itself, and it is learnable in an afternoon.
Why the usual advice stops working
Every guide you have read is true. Wave management matters, vision matters, a smaller champion pool matters. The trouble is that they are all true at the same time, and any single game is only ever lost for one or two reasons.
So you read the list, try to hold all of it in your head, and under pressure you do none of it. Twenty things to fix is functionally the same as no plan at all. The list is a menu. It was never the answer.
What a coach actually does in a first session is not hand you more information. It is watch one of your games and say which two of those twenty are costing you the most, so the other eighteen can wait until they matter.
Three rules for improving on purpose
- Find the leak before you choose the fix. Most players pick what to practise based on what they enjoy practising, or what they saw in a video that week. Pick it from evidence in your own losses instead, and you will usually be surprised by what comes up.
- Give a change ten games, and judge it by behaviour rather than LP. Ten games is nowhere near enough for the rank to move, and easily enough to see whether you actually did the thing. “Did I stop walking into the river unwarded?” is answerable. “Did it work?” is not.
- When it stops being the leak, stop working on it. A fixed habit does not need more attention, it needs to be left alone so the next problem can surface. Players who keep polishing the one thing they have already solved are the most common kind of plateau.
How to find your actual leak
This takes about forty minutes and you only need your own replays. Do it once, and you will have a single thing to work on for the next two weeks.
Watch three losses, and no wins
Wins hide everything. You can play a genuinely poor game, get carried, and learn nothing from watching it back. Pick your three most recent losses where the game was close enough to be decided by play rather than by a leaver.
Three is the number that matters, because one loss is a story and three is a pattern. Anything that shows up in all three is your leak. Anything that shows up once was just a bad game.
Find the minute, not the fight
Scrub through each replay and mark the moment the game stopped being winnable. Not the final teamfight, which is a symptom, but the point where you were still fine and then were not. Usually it is a two or three minute window, and usually it is much earlier than you expect.
Then ask the only question that matters: what were you doing in that window? Not what your jungler was doing. Reviewing your own replays properly is a skill worth learning on its own, and the short version is that you watch yourself, with the fog of war on, at the moments you would rather skip.
Name it as something you control
“I got camped” is not a leak. “I kept pushing to the enemy tower with no vision and no escape while their jungler was unaccounted for” is a leak, and it is the same sentence rewritten so that the fix belongs to you.
If your description of the problem contains the word “they”, rewrite it until it does not. The version that survives that edit is the thing to practise, and it is nearly always narrower and more boring than the version you started with.
The shortcut your rank already gives you
Leaks cluster. The reason coaching a Gold player looks nothing like coaching a Diamond player is that most people at the same rank are stuck on the same handful of things, and you can use that to narrow the search before you open a single replay.
In Gold the problem is usually doing too much, and the fix is subtraction. In Platinum you have mostly stopped making obvious mistakes and need an actual plan for the game. In Emerald it narrows to a few decisive minutes you keep spending badly, and by Diamond the leak stops being shared at all and becomes specifically yours.
Your rank is not a judgement, it is a starting hypothesis. Start from the one that matches you, then confirm or discard it with your own three losses.
“I just need to play more games”
Volume genuinely works, up to a point. If you play twice a week you are not stuck, you are undertrained, and the honest answer is to play more before you worry about method.
Past that point volume stops adding and starts entrenching. Every game you play on autopilot is a repetition of the habit you already have, which is why a player can put five hundred games into a season and finish within two divisions of where they started. Games do not teach you anything by themselves. Games you paid attention to during one specific thing do.
The version of this objection worth taking seriously is that you do not have time to review. Fair enough. Then review one game a week and play the rest normally, because one reviewed game a week beats forty unreviewed ones, and it is a smaller commitment than the extra hour of queue you were going to spend anyway.
From the coach
Most of the players I have worked with did not need new information. They needed someone to tell them which piece of the information they already had was the one to use. The pattern barely changes across three hundred students: they arrive with five things they are trying to fix at once, and they leave the first session with one, chosen out of their own replays rather than out of a video. That is also why a generic improvement plan does so little. The plan has to be built around how you actually play, or you end up practising problems that belong to somebody else. The list does not get longer as you improve, it gets shorter and more specific.
TBR37, Challenger mid and ADC coach at Level Legendary, and former head coach at League Unleashed. See his coach profile.
What to do this week
- Watch three losses and write one sentence. One sentence, describing one behaviour, containing no reference to a teammate. That sentence is your next two weeks.
- Set a behaviour target, not an LP target. “Ward the river bush before minute eight in ten games” is a target you can pass or fail. “Reach Platinum” is a wish.
- Review one game a week, and only that. Same day each week if you can. The consistency does more than the length of the session.
The hard part of all this is the diagnosis, because it is the one step you cannot do well from the inside. You are the least reliable witness to your own games, which is why the first thing League of Legends coaching buys you is not instruction but an outside pair of eyes on a game you thought you understood. Most players who book LoL coaching already know nineteen of the twenty things on the list, and have been working on the wrong one for a year.
Bring one of those three losses to a free first session: a Master or Challenger coach watches it with you live and names the leak, so you leave with one thing to practise instead of twenty. No card required.
