关于「多发与精做:线性账与乘法账」,《短视频流量池冷启动》给出的定量结论是什么?
结论:第五章在制作时间约束下求出期望破圈数的内部最优点(示意下每周 1.189 次与 1.334 次),精做方案一个季度零破圈的概率为 0.069,多发方案为 0.687。
数据口径:书中所有门槛、池子规模与权重均为示意,平台规则以其官方公示为准;全部计算由书中程序生成;本书是非个人化经营建议,只讨论真实内容质量与合规测试,不涉及任何刷量或规避平台规则的做法。 第三章说明硬门槛在数据上是斜坡,批次从 100 人增至 3,000 人时放行区间宽度从 0.124 收窄到 0.023;示意基准视频的破圈概率为 0.03585,瓶颈在第四道门。
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: 1.189, 1.334
| n/wk | h/video | q_sat | E[break] | q_log | E[break] |
|---|---|---|---|---|---|
| 1 | 24.0 | 1.599 | 0.497 | 2.292 | 0.918 |
| 2 | 12.0 | 1.571 | 0.941 | 1.792 | 1.334 |
| 3 | 8.0 | 1.489 | 1.179 | 1.500 | 1.210 |
| 4 | 6.0 | 1.383 | 1.189 | 1.292 | 0.893 |
| 6 | 4.0 | 1.178 | 0.875 | 1.000 | 0.387 |
| 8 | 3.0 | 1.011 | 0.547 | 0.792 | 0.156 |
| 12 | 2.0 | 0.779 | 0.214 | 0.500 | 0.028 |
| 24 | 1.0 | 0.454 | 0.037 | 0.000 | 0.000 |
| best n: saturating 4 (1.189/wk), logarithmic 2 (1.334/wk) | |||||
| six a week (h = 4) vs best: 0.74 and 0.29 of the best |
核心公式
\[E(n) = n \cdot P\bigl(q(H/n)\bigr).\]
查看完整验算程序
# Chapter 5. p51: a fixed weekly budget split into n videos.
# A creator has H = 24 production hours a week (illustrative). With h
# hours per video the latent quality is q(h), and each video breaks
# out with P(q) from the four logistic gates of p11. Expected
# breakouts per week = n * P(q(H / n)): linear in n at fixed quality,
# but quality falls as n rises. Two illustrative quality curves:
# saturating q(h) = 1.6 * (1 - exp(-h / 3))
# logarithmic q(h) = 0.5 + 0.5 * log2(h / 2)
import math
T = [0.0, 0.5, 1.0, 1.5]
S = 0.3
H = 24.0
def P(q):
return math.prod(1.0 / (1.0 + math.exp(-(q - t) / S)) for t in T)
def q_sat(h):
return 1.6 * (1.0 - math.exp(-h / 3.0))
def q_log(h):
return 0.5 + 0.5 * math.log2(h / 2.0)
print(" n/wk h/video q_sat E[break] q_log E[break]")
best = {}
for n in (1, 2, 3, 4, 6, 8, 12, 24):
h = H / n
es, el = n * P(q_sat(h)), n * P(q_log(h))
for key, val in (("sat", es), ("log", el)):
if key not in best or val > best[key][1]:
best[key] = (n, val)
print("%6d %7.1f %6.3f %8.3f %6.3f %8.3f"
% (n, h, q_sat(h), es, q_log(h), el))
print("best n: saturating %d (%.3f/wk), logarithmic %d (%.3f/wk)"
% (best["sat"][0], best["sat"][1], best["log"][0], best["log"][1]))
print("six a week (h = 4) vs best: %.2f and %.2f of the best"
% (6 * P(q_sat(4.0)) / best["sat"][1],
6 * P(q_log(4.0)) / best["log"][1]))唯一数据源与引用
范雅福(Fan, Y.). (2026). 《全域短视频赛马池冷启动动力学、完播互动复播阶跃阈值与零投流破圈测试矩阵母典》第 5 章「多发与精做:线性账与乘法账」(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-273 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026273,
author = {Fan, Yafu},
title = {Cold-Start Dynamics of Short-Video Traffic Pools: Step Thresholds in Completion, Engagement and Replay and a Zero-Ad-Spend Breakout Test Matrix},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}