Draft phase analysis: the biggest edge in LoL betting

Draft is the boring part of LoL that drives the outcome. Most bettors look at team win rates, pick the favorite, and bet. Draft modelers know that the same two teams produce different win rates depending on the ten champions that end up on Summoner’s Rift.

This article is a practical guide to reading and modeling draft priority for betting purposes.

What draft is, briefly

In the LoL competitive format, teams alternate bans and picks in a structured sequence. The team on blue side has first pick. Red side gets counter-pick on a specific slot. Over six bans and ten picks, teams decide which champions they will play, which champions their opponent cannot play, and which matchups land in each lane.

Small changes in draft preference produce large changes in win probability. A team that plays a comfort champion has a 55 percent win rate. Forced onto their fourth-best option, they might drop to 48 percent. The 7 percent swing compounds across five players.

Why the market prices draft poorly

Two reasons.

First, the betting public prices on team identity. “T1 is better than Hanwha Life, I will bet T1.” They do not consider whether T1 will get blind pick on their best champion or whether Hanwha’s draft coach will take that away.

Second, draft modeling requires watching a lot of drafts. Bookmakers employ traders who do this. The public does not. The gap is soft pricing that stays open until the sharp money arrives.

The three factors that matter most

1. Champion pool depth

A team that plays 8 champions in each role over a split has more options than a team that plays 4. When the first-choice champion gets banned, the second-choice drops off a cliff for narrow teams. Broad teams stay within a percent or two.

Champion pool depth by role over a 20-game window is a cleanly measurable stat. HLTV-style sites and Gol.gg publish it. Read it before betting.

2. Counter-pick strength

Red side gets the last pick into the blue-side fifth pick. A team with strong counter-pickers (specifically, a strong top laner with a wide pool of counter-matchups) can weaponize red side.

If you see a team known for their counter-picking play red in a game 1, fade slightly against them. The market often doesn’t price this team-specific advantage.

3. First-pick priority

Certain champions are universally strong on a patch. First pick lets you lock them. A team with blue side in game 1 who can draft the single strongest champion on the patch has a structural edge before a minion spawns.

If that champion is banned often (which is common when it is too strong), the advantage shrinks. Check the ban rate at the pro level, not solo queue. Pro ban rates are different.

Predicting the draft before it happens

Teams have patterns. Most teams ban their opponent’s two strongest champions first. Pick champions their players have practiced. Save a flex pick for late to confuse targeting.

To predict:

1. List each team’s most-played champions by role over the last 20 games.
2. Mark the opponent’s likely first-two bans based on what they usually target.
3. Mark your team’s likely first-two picks based on what’s typically safe.
4. Identify the probable flex pick (a champion that can go to multiple roles).
5. The remaining picks fall out based on counter-matchup preference.

This is a model that fits on a napkin. It predicts the first four picks about 65 percent of the time. The last six are harder, but by that point the draft’s structural edge is mostly decided.

A worked example

Match: Gen.G vs KT Rolster, BO3, LCK Summer Week 4.

Gen.G top-pool

Aatrox, Renekton, K’Sante, Gragas, Gnar. Depth: good.

KT top-pool

Jax, Fiora, Camille. Depth: narrow.

Patch meta

Aatrox and K’Sante strong. Fiora strong. Gragas strong. Jax mid-tier.

Predicted ban pattern

Gen.G likely bans Fiora and Jax to remove KT top priority. KT likely bans Aatrox and K’Sante to remove Gen.G priority.

Draft result

KT top ends up on Camille (third pool). Gen.G top ends up on Renekton (second pool). Camille/Renekton is roughly even in a vacuum, but Camille without Fiora/Jax as backup means KT’s comfort chain broke. Renekton for Gen.G is still a practiced matchup.

Implication

Gen.G draft priority is higher. Add 2 to 3 percent to Gen.G’s base win probability for this specific draft.

That is the whole process. It takes 10 minutes with a spreadsheet. It is 80 percent of what our full model does.

Markets that reward draft modeling

First tower

If you believe Gen.G’s draft gives them top-side priority (strong top laner into moderate top matchup), first tower is a value bet. Books often price first tower as 50/50.

First dragon

Depends on draft for bot lane and jungle. A team with an early-game jungler plus a safe bot lane gets dragon priority. Market often prices this market generously to the higher-rank team rather than the higher-priority draft team.

Total kills under (on LCK matches)

LCK draft often produces methodical 25-35 minute games. Books sometimes use league-wide averages that over-price the kill total on LCK specifically.

First tower + first blood combo

Highly correlated markets often priced as independent. If you have a draft priority edge, these combo bets pay when they hit simultaneously.

Common mistakes

1. Betting on team name, ignoring the specific game’s draft. Same two teams, different drafts, different win rates.
2. Using old champion pool data. Pools shift between splits as players practice new champions.
3. Ignoring flex picks. A team with a good flex pick can decide matchups by pick order. Under-priced when it works.
4. Forgetting about coaching changes. New coaches often change draft philosophy within the first month.

The bigger point

Draft modeling is the single largest source of edge still available to a smart recreational bettor in LoL. It is not easy, exactly, but it is tractable. Most people skip it. That is why it keeps working.

If you want to invest one skill into LoL betting, this is the one.