What Determines Victory in League of Legends?

In pro League of Legends, people love to say things like "first blood wins the game" or "dragons win championships." But when one team is 3k gold behind at 10 minutes, is the game really over? This final project is an explorable explanation that lets you interact with thousands of professional matches from Oracle’s Elixir to see how early-game gold and objectives actually relate to win rate.

And — Early Game Hype

In pro League, the first ten minutes are treated as everything. Drafts, jungle paths, and casting all revolve around getting that early lead on the scoreboard.

But — Data Tells a Messier Story

Across thousands of matches, some teams still lose with huge gold leads, while others win from “lost” positions. A 2k lead with no dragons is not the same as a 2k lead with early objective control.

Therefore — Let’s Make It Playable

This explorable turns those patterns into something you can play with: click beams on the map and adjust the sliders in the simulator to see how your favorite win conditions actually change win probability.

Explore Early-Game Advantages on the Rift

Click a glowing beam on the map to choose which stat is displayed in the bar chart below.

G = Gold difference · D = Dragons · B = Baron · T = Towers · K = Kills · V = Vision score

Win Probability Simulator (First 10 Minutes)

Adjust the early-game situation at 10 minutes and see how the predicted win probability changes. The model is a logistic regression trained on professional matches from Oracle’s Elixir.

Gold difference at 10 min (enemy – your team)

0 gold

Kill difference at 10 min (enemy – your team)

0 kills

Who secured the first dragon?

No dragon taken Your team Enemy team

Predicted win probability

--%

Set the sliders to reflect the position at 10 minutes.

Conclusion: What Actually Wins Pro Games?

If you remember only one thing from this page, let it be this: early gold leads are powerful, but they only truly matter when teams convert them into real objectives. A 2–3k lead at ten minutes looks very different depending on who has dragons, towers, and vision control.

Gold opens the door, objectives close it. The map view and bar chart show that raw gold difference is strongly correlated with win rate, but gold plus early dragons or tower plates pushes that advantage much higher.
Comebacks are real, but not magical. In the simulator, even teams that are behind can keep non-zero win chances when they trade gold for scaling comps, vision, or early objectives — but there is a point where the curve flattens and “hope” stops being realistic.
Context matters more than a single stat line. Two games with the same gold diff at ten minutes can have very different outcomes depending on where that lead comes from and how it’s used.

We chose this visual design so that pro-game concepts map directly onto the data: the Rift map anchors win conditions in familiar locations, the bar chart reveals the overall patterns across thousands of matches, and the simulator lets you plug in your own “ff 15” or “we scale” positions. Together, they turn a logistic regression model into something you can reason about as a player.