import bisect
import os
from collections import Counter, deque
from fractions import gcd
from functools import lru_cache
from functools import reduce
import functools
import heapq
import itertools
import math
import numpy as np
import re
import sys

if os.getenv("LOCAL"):
    sys.stdin = open("_in.txt", "r")

sys.setrecursionlimit(2147483647)
INF = float("inf")

N, PA, PB = sys.stdin.readline().rstrip().split()
A = list(map(int, sys.stdin.readline().split()))
B = list(map(int, sys.stdin.readline().split()))
N = int(N)
PA = float(PA)
PB = float(PB)
A.sort()
B.sort()


def probabilities(cards, first_p):
    """
    :param list of int cards:
    :param float first_p: 最初のカードを選ぶ確率
    :return:
    """
    cards.sort()
    # ret[i][j]: i 試合目で cards[j] を出す確率
    ret = np.zeros((N, N))
    # dp[bit]: 持ってるカードが bit になる確率
    # 左から i 番目のビットが立ってたら cards[i] が手持ちにある
    # dp する過程で ret を計算する
    dp = np.zeros(1 << N)
    # 初期
    dp[(1 << N) - 1] = 1
    states = {(1 << N) - 1}
    for step in range(N):
        nexts = set()
        for st in states:
            first = step < N - 1
            for b in range(N):
                mask = 1 << b
                next_st = st ^ mask
                if next_st > st:
                    continue
                nexts.add(next_st)
                if first:
                    p = dp[st] * first_p
                    first = False
                elif step == N - 1:
                    p = dp[st]
                else:
                    p = dp[st] * (1 - first_p) / (N - step - 1)
                dp[next_st] += p
                ret[step][b] += p
        states = nexts
    return ret


np.set_printoptions(suppress=True)

AP = probabilities(cards=A, first_p=PA)
BP = probabilities(cards=B, first_p=PB)

A = np.array(A)
B = np.array(B)
Ai, Bi = list(zip(*itertools.product(range(N), repeat=2)))
Ai = np.array(Ai)
Bi = np.array(Bi)
ai = Ai[A[Ai] > B[Bi]]
bi = Bi[A[Ai] > B[Bi]]
ans = 0
for step in range(N):
    ans += sum((A[ai] + B[bi]) * AP[step][ai] * BP[step][bi])
print(ans)