The data team forpro esports clubs.
From VOD review to metric design and analysis tools, we handle it.
We show you numbers,not narratives.
Results from 2026 LCK and LPL match footage, with the denominator on every rate.
2026 season LCK and LPL match footage
59%
The team with fewer deaths in the 60 seconds before an objective secures it.
276/466 cases · 291 games · still +15.8 pp after adjusting for the gold gap 60 seconds before
−921
ValidatedSecuring an objective while giving up a tower leaves you 921 gold behind four minutes later.
1,424 objectives taken · 410 games · gold change in the 240 seconds after
3×
Validated59% of invades with three or more players take an enemy camp. Solo invades: 19%.
3+ players: 513 cases · solo: 6,843 cases
+4.5%p
ValidatedInvades made while the enemy jungler is seen on the far side take an enemy camp more often.
429 games · gap +4.5 pp · 95% interval [2.6, 6.3] pp
“Validated” means we set the criteria first, then re-checked on 429 games held out from exploration.
We turn numbersinto pictures.
Every metric is shown as a chart or a map, and every number links to the scenes behind it. Screens are shown in Korean.
Ward mapSee an opposing support's early vision on one map. Match reviewReplay a game through its key moments and gold graph. 
Draft simulator42%How often the actual pick lands in the top 3 recommendations. 
Player habits3.7BFX's VicLa enters the enemy jungle 3.7 times per game (mid median: 1.9) and was caught 9 times in 67. 
Plan scoring95%Execution rate of the pre-Dragon vision plan.
ScuttlerLabs LoL
Match prep, drafting and validation in one workspace
Contracted teams only, behind login.
Match prep
Pick your next opponent and the preview brief and position briefings come together.
Match analysis
Combine 17 actions with time, gold and lane state to find scenes.
Draft
Run a draft against an AI opponent and see expected win rates by composition.
Validation
Every plan is scored against each new game.
- 595
- Match VODs analyzed scene by scene · LCK 475 · LPL 120
- 16,368
- LCK, LPL and LEC games used to train the draft recommendation model
- 5
- Leagues we analyze · LCK · LPL · LEC · LCS · CBLOL
- 2011–
- Every past season organized under one system
Off-season signing decisions, too
Compare re-signings and acquisitions on one basis, using the same records.
Player and team records and ratings
Compare players under the same conditions.
Roster change simulator
Calculate how expected win rate shifts when one slot changes.
Cross-league adjustment
Convert LPL and LEC player records to an LCK basis.
Champion pool fit
Check whether a candidate's champion pool fits the current patch and your drafts.
Analyze a game once.Every screen uses the same record.
Each analyzed game stays on record, and new games recalculate earlier judgments.
How we handle numbers
Execution per opportunity, not raw counts
We write “21 of 104 opportunities”, not “did it 20 times”.
Unseen stretches are left out of the denominator
Unseen stretches are never counted as “didn't do it.”
We show the counterexamples too
Games that contradict the hypothesis are shown too.
The coach makes the final call
Need analysis that isn't on screen?We'll build it.
Tell us the analysis you want and we design the rules, formulas and screens.
Examples of what we can build
- “Show the opposing jungler's first recall time by side”
- “Score our pre-Dragon vision plan automatically every game”
- “Compare five LPL mid candidates on an LCK baseline”
- “Show only the games where our top laner got lane-swapped”
- “Give us a clip list of every bot dive we allowed”
- “Feed our coaches' pick evaluation sheet into the draft recommendations”
- 1
Inquiry
Reach us by email or phone.
- 2
Demo
We show real screens and results on your next opponent.
- 3
Onboarding
Scope and terms are set to fit your team.