How to make decisions in TFT

The numbers come from high elo:

Set 18 Patch 18.5 77,232 high-elo games Games from 6 Oct to 20 Oct

TFT is a game of small decisions: what to buy, when to level up, who to give an item to. None of them wins the game on its own, but together they decide where you finish.

The four questions to ask every round

1. How much health do I have left? With lots of health you can afford to save; with little, it’s time to spend to stabilize.
2. Is my board winning rounds? If you’re losing almost all of them, the interest doesn’t make up for the health you’re losing.
3. How much gold do I have, and how much interest do I lose if I spend? See the economy guide.
4. What do I have? Units, pairs, components. That’s what decides which comp comes cheapest for you.

Pick a comp with what you have

Forcing the same comp every time works until its units don’t show up or someone contests you. It’s more reliable to head towards the comp that best fits what you already have. That’s how the in-game coach does it: it ranks the meta comps by how many of their core units you have (expensive ones count more), how many components you have for their carry’s items and how strong they are in high elo.

Not every S comp is for you

Average top 4 rate of the comps in each tier in high elo:

• Tier S: 69% (79 comps)
• Tier A: 54% (159 comps)
• Tier B: 40% (159 comps)
• Tier C: 25% (133 comps)

The gap is real, but an A comp that fits your game usually performs better than an S comp that isn’t coming together or that two other players are playing.

Pivoting

If your comp gets tough, look for another one that shares core units with yours: you lose less. Every comp page has a related comps section designed for exactly that.

Learn from every game

After playing, review the game in My matches. The warnings are sorted by importance (red: key to improve, amber: to improve, green: well done) and each one links to the guide that explains that decision.