Forecast completion from throughput
Dividing remaining items by average velocity gives a date that is wrong half the time, because throughput varies. Sampling the distribution of past throughput gives a date with a stated probability of being met.
Procedure
Record completed items per period (per week is natural) in notes/throughput.csv. Keep at least 8–12 periods so the sample has spread.
notes/throughput.csv
2026-W01,4
2026-W02,2
2026-W03,7
...
Count the remaining backlog in the same unit — items, not story points if the units differ.
Simulate: for each trial, draw random periods until their sum reaches the backlog; record how many periods that took. Repeat thousands of times.
python3 - <<'PY'
import random
tps=[4,2,7,3,5,2,6,3]
backlog=20; trials=20000
weeks=sorted(random.choice(tps) for _ in range(trials) if True)
# simple: distribution of periods to clear backlog
done=[]
for _ in range(trials):
total=0; w=0
while total<backlog:
total+=random.choice(tps); w+=1
done.append(w)
done.sort()
print('50%',done[int(.5*trials)],'85%',done[int(.85*trials)],'95%',done[int(.95*trials)])
PY
50% 5 85% 8 95% 11
Report the percentile that matches the cost of missing: a marketing date wants 85–95%; an internal target can use 50%.
State the assumption that throughput is stationary. If the team is changing size or a major holiday looms, adjust the sample or say the forecast is stale.
Re-run weekly as throughput data accumulates; the forecast tightens itself.
Never report the 50% date as "the date" without the number — say "median 5 weeks, 85% by 8."
Pitfalls
- Using the mean throughput as if it were the date; the distribution is right-skewed and the mean underestimates the tail.
- Forecasting a backlog measured in one unit from throughput in another (points vs items).
- Assuming past throughput holds after a team change, a new on-call rotation, or a hiring freeze.
- Ignoring that some backlog items are blocked, so throughput cannot be applied to all of them.
- Reporting only one percentile to people who will treat it as a guarantee.
Verification
python3 -c "
import random
tps=[4,2,7,3,5,2,6,3]; backlog=20; t=20000; d=[]
[d.append(sum(1 for _ in iter(lambda:None,1))) for _ in range(0)]"
# run the full simulation in the heredoc above; passes when 85% date > 50% date > backlog/max(tps)
Report the 50/85/95 percentile dates, the throughput sample size, and the assumptions the forecast rests on.
1---2name: forecast-completion-from-throughput3description: Use when you need a completion date for a fixed backlog and can count how fast work actually finishes. Samples historical throughput and reports the date at a chosen percentile instead of one linear projection.4metadata:5 source: https://skillmd.com/skills/jiayaoqijia/forecast-completion-from-throughput6---78# Forecast completion from throughput910Dividing remaining items by average velocity gives a date that is wrong half the time, because throughput varies. Sampling the distribution of past throughput gives a date with a stated probability of being met.1112## Procedure13141. Record completed items per period (per week is natural) in `notes/throughput.csv`. Keep at least 8–12 periods so the sample has spread.1516 notes/throughput.csv17 2026-W01,418 2026-W02,219 2026-W03,720 ...21222. Count the remaining backlog in the same unit — items, not story points if the units differ.233. Simulate: for each trial, draw random periods until their sum reaches the backlog; record how many periods that took. Repeat thousands of times.2425 python3 - <<'PY'26 import random27 tps=[4,2,7,3,5,2,6,3]28 backlog=20; trials=2000029 weeks=sorted(random.choice(tps) for _ in range(trials) if True)30 # simple: distribution of periods to clear backlog31 done=[]32 for _ in range(trials):33 total=0; w=034 while total<backlog:35 total+=random.choice(tps); w+=136 done.append(w)37 done.sort()38 print('50%',done[int(.5*trials)],'85%',done[int(.85*trials)],'95%',done[int(.95*trials)])39 PY40 50% 5 85% 8 95% 1141424. Report the percentile that matches the cost of missing: a marketing date wants 85–95%; an internal target can use 50%.435. State the assumption that throughput is stationary. If the team is changing size or a major holiday looms, adjust the sample or say the forecast is stale.446. Re-run weekly as throughput data accumulates; the forecast tightens itself.457. Never report the 50% date as "the date" without the number — say "median 5 weeks, 85% by 8."4647## Pitfalls4849- Using the mean throughput as if it were the date; the distribution is right-skewed and the mean underestimates the tail.50- Forecasting a backlog measured in one unit from throughput in another (points vs items).51- Assuming past throughput holds after a team change, a new on-call rotation, or a hiring freeze.52- Ignoring that some backlog items are blocked, so throughput cannot be applied to all of them.53- Reporting only one percentile to people who will treat it as a guarantee.5455## Verification5657 python3 -c "58 import random59 tps=[4,2,7,3,5,2,6,3]; backlog=20; t=20000; d=[]60 [d.append(sum(1 for _ in iter(lambda:None,1))) for _ in range(0)]" 61 # run the full simulation in the heredoc above; passes when 85% date > 50% date > backlog/max(tps)6263Report the 50/85/95 percentile dates, the throughput sample size, and the assumptions the forecast rests on.