关于「法力曲线:期望可用费用的马尔可夫链与实际花费的模拟」,《卡牌费用回归与循环克制》给出的定量结论是什么?

结论:期望可用费用不是「回合数与资源数取小」,而是一条错过就无法补回的马尔可夫链。资源 17 张的先手,前 7 回合累计只有 23.89 点可用费用,且有 0.748 的概率至少错过一次放置。可用不等于花掉:低曲线卡组因手牌耗尽而花得更少,最佳资源数在 16 张;高曲线卡组的最佳资源数在 19 张。下一章离开单副卡组,进入卡组之间的克制关系。

数据口径:本书用一款自拟的示意卡牌游戏「衡枢」把这些问题写成可以运行的算术。 书中卡池、胜率与价格参数均为示意;全部计算由书中程序生成;本书不是法律、财务或投资建议,也不推荐任何购买。

English version

对照数据表

下表数字由本章 Python 程序原样打印,可重新运行复现。 程序只用 Python 标准库与结论吻合的打印值: 0.748, 17, 19, 23.89

expected mana in play, first player (ideal = turn number)
Lt1t2t3t4t5t6t7totalmiss7
150.971.882.673.353.924.404.8322.030.871
170.991.932.813.594.284.885.4123.890.748
190.991.972.903.764.565.275.9225.370.587
211.001.982.953.884.755.576.3326.460.413
L = 17, second player gains 1.083 mana-turns over turns 1-7
ideal total over 7 turns = 28 mana-turns
查看完整验算程序
# Chapter 3. p31: Expected mana in play by turn, an exact Markov chain.
# Invented rules (illustrative): 40 cards, L resources, 7-card hand, one
# resource may be played per turn, the first player skips the turn-1
# draw. State = (resources seen r, resources in play m). Each draw is a
# resource with probability (L - r) / (40 - n); then m <- min(m + 1, r).
# Exact fractions; E[m_t] is the mana available on turn t.
import math
from fractions import Fraction as F

N, HAND, TURNS = 40, 7, 7


def chain(L, first):
    dist = {}
    for r in range(HAND + 1):
        p = F(math.comb(L, r) * math.comb(N - L, HAND - r),
              math.comb(N, HAND))
        if p:
            dist[(r, 0)] = p
    seen = HAND
    out = []
    for t in range(1, TURNS + 1):
        nxt = {}
        for (r, m), p in dist.items():
            if t == 1 and first:
                opts = [(r, p)]
            else:
                q = F(L - r, N - seen)
                opts = [(r + 1, p * q), (r, p * (1 - q))]
            for r2, p2 in opts:
                if p2:
                    k = (r2, min(m + 1, r2))
                    nxt[k] = nxt.get(k, 0) + p2
        if not (t == 1 and first):
            seen += 1
        dist = nxt
        em = math.fsum(float(p * m) for (r, m), p in dist.items())
        miss = math.fsum(float(p) for (r, m), p in dist.items() if m < t)
        out.append((em, miss))
    return out


print("expected mana in play, first player (ideal = turn number)")
print("L   t1    t2    t3    t4    t5    t6    t7   total  miss7")
for L in (15, 17, 19, 21):
    o = chain(L, True)
    tot = math.fsum(e for e, _ in o)
    print("{:2d} ".format(L) + " ".join("{:5.2f}".format(e) for e, _ in o)
          + " {:6.2f} {:6.3f}".format(tot, o[-1][1]))
o1, o2 = chain(17, True), chain(17, False)
gap = math.fsum(b for b, _ in o2) - math.fsum(a for a, _ in o1)
print("L = 17, second player gains {:.3f} mana-turns over turns 1-7"
      .format(gap))
ideal = TURNS * (TURNS + 1) // 2
print("ideal total over 7 turns = {} mana-turns".format(ideal))

唯一数据源与引用

范雅福(Fan, Y.). (2026). 《非对称策略卡牌费用收益回归、牌库构筑超几何卡手率边界与优势策略循环克制》第 3 章「法力曲线:期望可用费用的马尔可夫链与实际花费的模拟」(版本 1.0.0). 5D Library. DOI 待发布

DOI 待发布 · 专著页正式发布后上线(本站暂无专著页) · 5DL-B-179 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026179,
  author    = {Fan, Yafu},
  title     = {Cost-Benefit Regression for Asymmetric Strategy Cards, Hypergeometric Brick-Rate Bounds in Deck Building and Cyclic Counters to Dominant Strategies},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

本书其他章节结论