What does Private-Channel Rate Shaping conclude, in numbers, about “Who Decides Account Quality: Consent Rates, Negative Feedback and a Markov Chain of Tiers”?

Answer: Chapter 2 writes a Markov chain of four account tiers: as consent falls from 0.90 to 0.80, the expected daily limit drops from 99054 to 24692, while changing send times alone moves expected negative feedback by 0.000000.

Scope of the numbers: An illustrative plan averaging 10.0 messages a minute still has 420 refused when sent in batches, and none once a token bucket of depth 20 shapes it. Parameters are illustrative, platform figures carry the document and the date read, and all numbers come from the book's programs, which read and write no file and use no network.

中文版

Data table

The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python onlyprinted values that match the answer: 0.80, 0.90, 24692, 99054

consentfeedbackP(high)P(low)P(top)expected 24h limit
1.000.00100.9950.0001.000100000
0.950.00200.9000.0001.00099979
0.900.00290.6380.0070.98999054
0.800.00480.1570.1720.21724692
0.700.00670.0200.5810.000324
0.600.00860.0020.8760.000254
Show the full program
# Chapter 2. p21: consent share drives the tier chain, through feedback.
# Model (illustrative; the platform does not publish its formula). Tier
# limits are Meta's published steps of unique recipients per moving 24 h
# (250, 2,000, 10,000, 100,000; "unlimited" is folded into the top).
# Each week the platform sees n = 2000 delivered messages; k of them
# draw negative feedback (block or report), k ~ Binomial(n, f). Quality
# is high if k <= 6 (rate <= 0.003), low if k > 12 (rate > 0.006),
# medium otherwise. High moves one tier up, low one tier down, medium
# stays. Per-message feedback rate: 0.001 from a consented contact,
# 0.02 from a contact who never opted in; f = s 0.001 + (1 - s) 0.02.
import math

LIMITS = (250, 2000, 10000, 100000)
N, HI_K, LO_K = 2000, 6, 12
P_C, P_U = 0.001, 0.02


def binom_cdf(k, n, p):
    terms = [math.comb(n, j) * p ** j * (1 - p) ** (n - j)
             for j in range(k + 1)]
    return math.fsum(terms)


def quality_probs(f):
    ph = binom_cdf(HI_K, N, f)
    pl = 1 - binom_cdf(LO_K, N, f)
    return ph, 1 - ph - pl, pl


def stationary(f):
    ph, pm, pl = quality_probs(f)
    m = len(LIMITS)
    tr = [[0.0] * m for _ in range(m)]
    for i in range(m):
        for j, p in ((min(i + 1, m - 1), ph), (i, pm),
                     (max(i - 1, 0), pl)):
            tr[i][j] = math.fsum([tr[i][j], p])
    pi = [1.0 / m] * m
    for _ in range(5000):
        pi = [math.fsum(pi[i] * tr[i][j] for i in range(m))
              for j in range(m)]
    return pi


if __name__ == "__main__":
    print("consent  feedback  P(high)  P(low)  P(top)  expected 24h limit")
    for s in (1.0, 0.95, 0.9, 0.8, 0.7, 0.6):
        f = s * P_C + (1 - s) * P_U
        ph, pm, pl = quality_probs(f)
        pi = stationary(f)
        lim = math.fsum(w * x for w, x in zip(pi, LIMITS))
        print("%7.2f  %8.4f  %7.3f  %6.3f  %6.3f  %18.0f"
              % (s, f, ph, pl, pi[-1], lim))

Single source and citation

Fan, Y. (2026). Rate Control for Private Messaging Channels without Account Bans: Token-Bucket Shaping, Platform-Published Limits and Consent-Based Reach Throughput (Version 1.0.0), Chapter 2: Who Decides Account Quality: Consent Rates, Negative Feedback and a Markov Chain of Tiers. 5D Library. DOI pending

DOI pending · monograph page goes live after publication · 5DL-B-271 · v1.0.0 · CC-BY-NC-ND-4.0

BibTeX
@book{fan2026271,
  author    = {Fan, Yafu},
  title     = {Rate Control for Private Messaging Channels without Account Bans: Token-Bucket Shaping, Platform-Published Limits and Consent-Based Reach Throughput},
  year      = {2026},
  publisher = {5D Library},
  edition   = {1.0.0},
  note      = {DOI pending}
}

Other chapter answers in this book