What does Drift and Noise on Long Pages conclude, in numbers, about “Two Absorbing Barriers and First Passage: a Visitor's Path between "Pay" and "Leave"”?
Answer: The drift-diffusion model writes a visitor's hesitation as a random path between a "leave" barrier and a "pay" barrier. Program p11 gives closed forms for the probability of paying and the mean decision time and checks them by simulation with a continuity correction; p12 shows that payers and leavers have equal conditional times only when the start is central, so dwell data must be read together with the starting point. The next chapter isolates the noise and shows how it makes visitors who would have paid hit the leave barrier by mistake.
Scope of the numbers: Chapter 4 pushes a page forward section by section: deleting the feature table of an illustrative page gains 0.0081, and the best and worst orders of sections differ by 0.1764. Marked boxes separate published evidence, runnable Python calculations (20 programs), classical sources and the author's hypotheses.
Data table
The numbers below are printed verbatim by the chapter's Python program and can be reproduced by running it. standard-library Python only
| drift v P(pay) P(leave) E[T] screens | |
| -0.40 | 0.3100 0.6900 0.9499 |
| -0.20 | 0.4013 0.5987 0.9869 |
| -0.10 | 0.4502 0.5498 0.9967 |
| +0.00 | 0.5000 0.5000 1.0000 |
| +0.10 | 0.5498 0.4502 0.9967 |
| +0.20 | 0.5987 0.4013 0.9869 |
| +0.40 | 0.6900 0.3100 0.9499 |
| +0.80 | 0.8320 0.1680 0.8300 |
| Monte Carlo check: Euler step 0.005, 8,000 visitors per row | |
| drift v P(pay) exact P(pay) MC E[T] exact E[T] MC | |
| -0.20 | 0.4013 0.3960 0.9869 0.9888 |
| +0.00 | 0.5000 0.5025 1.0000 1.0041 |
| +0.20 | 0.5987 0.5999 0.9869 0.9904 |
| +0.50 | 0.7311 0.7319 0.9242 0.9245 |
Key formula
Show the full program
# Chapter 1. p11: closed-form pay probability and mean decision time
# between two absorbing barriers, checked by a fixed-seed Monte Carlo.
# 模型:信念 X(t) 满足 dX = v dt + s dW;0 为「离开」壁,A 为「付款」壁,
# 起点 Z。付款概率 P = (1 - exp(-2vZ/s^2)) / (1 - exp(-2vA/s^2)),
# v = 0 时 P = Z/A;平均决定时间 E[T] = (A P - Z) / v,
# v = 0 时 E[T] = Z (A - Z) / s^2。时间单位为「一屏阅读时长」。
# 参数 A = 2、Z = 1、s = 1 与各档漂移均为示意,不对应任何真实页面。
# 欧拉离散会越过壁才被发现,等于把壁外推约 0.5826 s sqrt(dt);
# 模拟时把两壁各向内收这一距离(离散监测的连续性校正)。
import math
import random
A, Z, S = 2.0, 1.0, 1.0
def p_pay(v, s=S, a=A, z=Z):
if abs(v) < 1e-12:
return z / a
k = 2.0 * v / (s * s)
return (1.0 - math.exp(-k * z)) / (1.0 - math.exp(-k * a))
def mean_time(v, s=S, a=A, z=Z):
if abs(v) < 1e-12:
return z * (a - z) / (s * s)
return (a * p_pay(v, s, a, z) - z) / v
def simulate(v, n, dt, rng):
pays, steps = 0, []
sq = S * math.sqrt(dt)
lo, hi = 0.5826 * sq, A - 0.5826 * sq
for _ in range(n):
x, k = Z, 0
while lo < x < hi:
x = x + v * dt + sq * rng.gauss(0.0, 1.0)
k += 1
if x >= hi:
pays += 1
steps.append(k * dt)
return pays / n, math.fsum(steps) / n
print("drift v P(pay) P(leave) E[T] screens")
for v in (-0.4, -0.2, -0.1, 0.0, 0.1, 0.2, 0.4, 0.8):
p = p_pay(v)
print("%+.2f %.4f %.4f %.4f" % (v, p, 1.0 - p, mean_time(v)))
print()
print("Monte Carlo check: Euler step 0.005, 8,000 visitors per row")
print("drift v P(pay) exact P(pay) MC E[T] exact E[T] MC")
rng = random.Random(20261113)
for v in (-0.2, 0.0, 0.2, 0.5):
pm, tm = simulate(v, 8000, 0.005, rng)
print("%+.2f %.4f %.4f %.4f %.4f"
% (v, p_pay(v), pm, mean_time(v), tm))Single source and citation
Fan, Y. (2026). Fokker-Planck Drift-Diffusion on Long Sales Pages: Belief-Drift Rates and First-Passage Times to the Payment Barrier (Version 1.0.0), Chapter 1: Two Absorbing Barriers and First Passage: a Visitor's Path between "Pay" and "Leave". 5D Library. DOI pending
DOI pending · monograph page goes live after publication · 5DL-B-240 · v1.0.0 · CC-BY-NC-ND-4.0
BibTeX
@book{fan2026240,
author = {Fan, Yafu},
title = {Fokker-Planck Drift-Diffusion on Long Sales Pages: Belief-Drift Rates and First-Passage Times to the Payment Barrier},
year = {2026},
publisher = {5D Library},
edition = {1.0.0},
note = {DOI pending}
}Other chapter answers in this book
- Chapter 2 · Noise and Mistaken Exits: Why a Longer Page Does Not Mean More Persuasion
- Chapter 3 · The Fokker-Planck Equation by Finite Differences: a Crowd of Visitors as a Cloud of Probability
- Chapter 4 · A Page Pushed Forward Section by Section: Deletion, Order and the Best Length
- Chapter 5 · Estimating Drift and Noise from Dwell and Payment: EZ-Diffusion and Identifiability
- Chapter 6 · Speed, Accuracy and the Opposing Evidence: Compressed Barriers, Mixed Audiences and the Cue of Argument Quantity
- Chapter 7 · Eastern and Western Sources: Xunzi on the Divided, Tilted and Doubled Mind, and Einstein's 1905 Brownian Motion
- Chapter 8 · Closing the Loop: Scroll Depth, Controlled Deletions and a Pre-Registered Test