使用python进行计算。
数据如下，卡牌顺序为图鉴中从左到右，从上到下。
带有下划线的数据为敲过的牌的数据。
zhan=[6, 9, 9, 10, 10, 4, 10, 5, 9, 9, 7, 16, 18, 4, 0, 5, 9, 8, 9, 6, 15, 13, 12, 10, 16, 12, 32, 0, 8, 8, 0, 28, 15]
zhan_=[8, 12, 10, 10, 14, 7, 14, 7, 12, 13, 9, 21, 24, 4, 0, 5, 9, 10, 10, 9, 20, 13, 18, 12, 21, 15, 42, 0, 10, 12, 0, 40, 20]
lie=[6, 8, 11, 6, 7, 9, 8, 11, 12, 11, 0, 9, 10, 12, 9, 12, 10, 15, 12, 15, 10, 0, 10, 6]
lie_=[9, 12, 15, 8, 9, 12, 10, 14, 15, 15, 0, 12, 14, 16, 11, 16, 13, 20, 18, 19, 13, 0, 13, 6]
chu=[11, 14, 10, 9, 7, 10, 12, 7, 18, 0, 13, 0, 8, 8, 8, 8, 30, 15, 30, 27, 33, 6, 0, 5, 21, 9, 20, 49, 36]
chu_=[15, 18, 13, 10, 8, 11, 13, 9, 24, 0, 18, 0, 10, 9, 11, 10, 40, 18, 38, 35, 44, 9, 0, 7, 26, 11, 25, 49, 48]
gu=[0, 9, 8, 13, 9, 4, 8, 7, 10, 0, 0, 27, 0, 0, 10, 10, 0, 0, 8, 0, 0, 0, 13, 52, 0, 7, 10, 9, 0, 0, 33]
gu_=[0, 11, 11, 17, 12, 6, 10, 8, 12, 0, 0, 33, 0, 0, 13, 12, 0, 0, 9, 0, 0, 0, 18, 63, 0, 9, 13, 9, 0, 0, 33]
ji=[3, 3, 3, 7, 10, 0, 6, 9, 13, 6, 12, 18, 5, 3, 7, 13, 10, 12, 24, 33, 9, 11, 28, 0, 18, 10, 19, 24]
ji_=[4, 4, 4, 10, 13, 0, 9, 11, 16, 9, 15, 22, 6, 4, 10, 14, 13, 18, 32, 39, 15, 15, 36, 0, 23, 14, 24, 30]

其余代码如下（相当劣质，但能用就行）：
zhan_fei=lie+chu+gu+ji
zhan_fei_=lie_+chu_+gu_+ji_
lie_fei=zhan+chu+gu+ji
lie_fei_=zhan_+chu_+gu_+ji_
chu_fei=zhan+lie+gu+ji
chu_fei_=zhan_+lie_+gu_+ji_
gu_fei=zhan+lie+chu+ji
gu_fei_=zhan_+lie_+chu_+ji_
ji_fei=zhan+lie+chu+gu
ji_fei_=zhan_+lie_+chu_+gu_
def fei(fei_list):
total = 0
count = 0
n = len(fei_list)
for i in range(n - 2):
for j in range(i + 1, n - 1):
for k in range(j + 1, n):
total += max(fei_list[i], fei_list[j], fei_list[k])
count += 1
if count > 0:
return round(total / count, 2)
else:
return 0
print("战飞溅期望:", fei(zhan_fei))
print("战飞溅+期望:", fei(zhan_fei_))
print("猎飞溅期望:", fei(lie_fei))
print("猎飞溅+期望:", fei(lie_fei_))
print("储飞溅期望:", fei(chu_fei))
print("储飞溅+期望:", fei(chu_fei_))
print("骨飞溅期望:", fei(gu_fei))
print("骨飞溅+期望:", fei(gu_fei_))
print("机飞溅期望:", fei(ji_fei))
print("机飞溅+期望:", fei(ji_fei_))

最终结论：
战飞溅期望: 18.83
战飞溅+期望: 23.58
猎飞溅期望: 19.19
猎飞溅+期望: 24.13
储飞溅期望: 16.16
储飞溅+期望: 20.64
骨飞溅期望: 18.32
骨飞溅+期望: 23.43
机飞溅期望: 17.97
机飞溅+期望: 22.71

验算（取一百万次随机数进行计算，数据为战士飞溅加）：
import random
total=0
count=1000000
for i in range(count):
total+= max(random.sample(zhan_fei_, 3))
print(total/count)
结果：23.582983，符合上述数据

叠甲：
1，此伤害期望为理论伤害期望，不等于真实伤害期望，只能说明无脑选择最高伤害时[飞溅]的伤害期望。
2，飞溅的真实伤害期望在SL局中会更高，如弱怪开出七星，SL后不开，强怪、精英再开七星。
3，不能认为[飞溅]完全等于1c18。（虽然我知道绝对会有人在飞溅开下面回复“1c18罢了”，然后马上就有人问“谁算的？飞溅1c18？怎么算的？”）

分析：
1，飞溅的理论伤害期望看来是猎>战>骨>机>储。
2，平均来看，飞溅大概是1c18，飞溅+大概是1c23，升级大概+5，约等于[契约终结]的升级幅度。

最后，数据全为人工统计，代码手打加AI校对，如有缺漏，还望不吝赐教。

一些趣事：
1，在飞溅能出的共145张攻击牌中，有整整23张牌在本计算方法中伤害为0（奥斯提牌、X费牌、华、每X一次就造成一次伤害的终端。）
2，在飞溅能出的共145张攻击牌中，伤害最高的是骨的埋葬+（63），但骨同时也有最多的0伤牌：合计13张，贡献了23张0伤牌的一半还多……
3，鸡煲的[折射]看起来就9x2+两个玻璃球，但在此计算方法下是9x2+8（推掉第一个玻璃）+4（第二个玻璃）+3（第二个玻璃，下回合），共33伤害，大于重锤……

飞溅对三伤害期望：
计算方式：
1，所有aoe牌伤害乘3计算，然后再从三选一中无脑选择伤害最高一项。
2，其余不变。
数据如下：
zhan=[6, 9, 9, 10, 10, 12, 10, 5, 27, 9, 7, 16, 18, 4, 0, 5, 9, 8, 9, 6, 15, 13, 12, 10, 48, 36, 32, 0, 24, 8, 0, 28, 15]
zhan_= [8, 12, 10, 10, 14, 21, 14, 7, 39, 13, 9, 21, 24, 4, 0, 5, 9, 10, 10, 9, 20, 13, 18, 12, 63, 45, 42, 0, 30, 12, 0, 40, 20]
lie=[6, 24, 11, 18, 7, 9, 8, 11, 12, 11, 0, 9, 10, 12, 9, 12, 10, 15, 12, 15, 10, 0, 30, 6]
lie_=[9, 36, 15, 24, 9, 12, 10, 14, 15, 15, 0, 12, 14, 16, 11, 16, 13, 20, 18, 19, 13, 0, 39, 6]
chu=[11, 42, 10, 9, 21, 10, 12, 7, 18, 0, 13, 0, 24, 8, 8, 8, 30, 15, 30, 27, 33, 6, 0, 5, 63, 27, 20, 147, 36]
chu_=[15, 54, 13, 10, 24, 11, 13, 9, 24, 0, 18, 0, 30, 9, 11, 10, 40, 18, 38, 35, 44, 9, 0, 7, 78, 33, 25, 147, 48]
gu=[0, 9, 8, 13, 9, 4, 8, 7, 10, 0, 0, 27, 0, 0, 10, 10, 0, 0, 8, 0, 0, 0, 13, 52, 0, 7, 10, 9, 0, 0, 99]
gu_=[0, 11, 11, 17, 12, 6, 10, 8, 12, 0, 0, 33, 0, 0, 13, 12, 0, 0, 9, 0, 0, 0, 18, 63, 0, 9, 13, 9, 0, 0, 99]
ji=[3, 3, 3, 7, 10, 0, 18, 9, 13, 6, 12, 18, 5, 3, 7, 13, 10, 12, 24, 63, 9, 33, 84, 0, 18, 10, 19, 24]
ji_=[4, 4, 4, 10, 13, 0, 27, 11, 16, 9, 15, 22, 6, 4, 10, 14, 13, 18, 32, 69, 15, 45, 108, 0, 23, 14, 24, 30]

对三最终结论：
战飞溅期望: 28.96
战飞溅+期望: 35.18
猎飞溅期望: 29.44
猎飞溅+期望: 36.01
储飞溅期望: 23.26
储飞溅+期望: 29.41
骨飞溅期望: 28.25
骨飞溅+期望: 35.4
机飞溅期望: 26.91
机飞溅+期望: 33.05

对三分析：
1，对三时，飞溅的理论伤害期望依旧是猎>战>骨>机>储，但飞溅+变成了猎>骨>战>机>储，大概是因为骨只有女妖一张能打出伤害的aoe，战则有很多。
2，平均来看，对三时飞溅平均等价于1caoe打9，飞溅+平均等价于1caoe打11，升级大概+3。
3，对于2，这不是说飞溅等价于1caoe打9，这是说飞溅在对3时，平均伤害期望与1caoe打9相当。换言之，可以认为飞溅在打三时会从1c打18“自适应”为1caoe打9。