How many players do you need?
Scroll sideways for every column.
Players needed against Mega Falinks by trainer level| Every trainer is level | One alone takes | Estimator | Players for 50% win | Players for 90% win |
|---|
| Level 30 | 24:10 | 5.25 | 6 | 6 |
| Level 40 | 21:49 | 4.74 | 5 | 5 |
| Level 50 | 20:12 | 4.39 | 5 | 5 |
Each row is a lobby of identical trainers who all bring the suggested six at that level. The same Estimator the group size tool prints, turned into a win chance.
Why this Estimator can differ from /raids: /raids prints the best single counter fought six times over (Salamence: 4.38 trainers at level 40 with these filters), and hides Legendary attackers unless you tick them. This page runs a party of six different Pokemon in the best order with relobbies, with Legendaries excluded as on /raids. Both come from the same engine; they answer different questions, so the figures are close but not identical.
How this is worked out
Each trainer's six is run through the team builder's party simulator (24 runs with the boss's own movesets; faint swaps and relobbies included). Trainers fight at the same time, so their damage rates add: the lobby's time to win is the boss bar divided by the sum of those rates, and the lobby wins when that fits the 276-second usable timer (the same effective timer the Estimator divides by). Each trainer's damage rate is their average over the whole fight, relobbies included, so a lobby of N identical trainers wins exactly when N reaches their Estimator, and a one-trainer lobby shows that trainer's own time. The numbers are fixed by the URL (seed), so reloading gives the same answer. All trainers in one run face the same boss moveset, as in a real raid. This assumes steady damage and that trainers do not get in each other's way. Share this exact lobby: link · /api/raids/lobby returns the same numbers as JSON.