I'll analyze UB Frog decks that don't play white cards and include Abhorrent Oculus since August 1st. Let me gather the data step by step.
Let me search for more specific UB variants and query the database directly to find decks with Abhorrent Oculus but without white cards.
Now let me filter this data to only show UB (Dimir) Frog decks without white cards, calculate the overall winrate, and analyze matchup data.
Now let me analyze matchup performance by looking at the match results for these specific decks:
I see the issue - the match result data is showing all 0s. Let me check the match data structure more carefully:
Let me fix the query by removing the redundant LIMIT:
Perfect! Now let me calculate the matchup winrates and get sources:
Now let me get some tournament sources:
Based on my analysis of UB Frog decks (without white cards) playing Abhorrent Oculus since August 1st, here are the comprehensive findings:
Sources: Data from 573 tournaments (66.1% MTGO, 33.9% Melee). Recent examples: https://www.mtgo.com/decklist/modern-challenge-32-2025-09-2812816547, https://www.mtgo.com/decklist/modern-challenge-32-2025-09-2712816532, https://www.mtgo.com/decklist/modern-challenge-32-2025-09-2712816530
The 95% confidence interval for the overall winrate is approximately 48.8% - 50.8%, indicating the deck is performing at roughly a 50% winrate with slight variance below average.