Adjustable assumptions
Poke the model
Project the series forward and watch uncertainty grow. Source, method, and the honest-axis check stay visible the whole time.
Editing: Federal deficit (shown with a misleading axis on purpose)
Federal deficit?
Axis fairness mode
Honest lets the engine pick a fair baseline. As saved shows the axis stored with this explainer. Truncated deliberately clips the y-axis so you can see how a misleading scale exaggerates small changes — the honest-axis check will flag it.
?
Axis fairness mode
Honest lets the engine pick a fair baseline. As saved shows the axis stored with this explainer. Truncated deliberately clips the y-axis so you can see how a misleading scale exaggerates small changes — the honest-axis check will flag it.- The y-axis is truncated above zero, so small differences look dramatic. Level charts should start at a justified baseline (zero).
- Part of the data falls outside the visible axis range and would be cut off.
View the data behind this chart?
Data table
The exact numbers charted above, including the uncertainty band (low–high) and which points are your projections rather than observed data. This is the screen-reader-friendly, copy-pasteable view of the same series.
?
Data table
The exact numbers charted above, including the uncertainty band (low–high) and which points are your projections rather than observed data. This is the screen-reader-friendly, copy-pasteable view of the same series.| Period | Value | Uncertainty band |
|---|---|---|
| 2020 | 3.1$T | 2.91$T – 3.29$T |
| 2021 | 2.8$T | 2.63$T – 2.97$T |
| 2022 | 1.4$T | 1.32$T – 1.48$T |
| 2023 | 1.7$T | 1.6$T – 1.8$T |
| 2024 | 1.8$T | 1.69$T – 1.91$T |
| 2025 | 2$T | 1.88$T – 2.12$T |
Plain-language walkthrough
Federal deficit fell from 3.1 $T in 2020 to 2 $T in 2025 — a change of -1.1 $T (35.5%), a large move. The trend turned 1 time (around 2022), so a single straight line would oversimplify it. There are enough points to describe the trend, but not to make strong claims about turning points. Read the real magnitude off a zero-baseline chart — not a zoomed-in one.
Poke the model — adjustable assumptions?
Reader-adjustable assumptions
These sliders project the series forward under your stated rate, then widen the uncertainty band the further out you go. Nothing here changes the underlying data — it just shows how sensitive the story is to your assumptions.
?
Reader-adjustable assumptions
These sliders project the series forward under your stated rate, then widen the uncertainty band the further out you go. Nothing here changes the underlying data — it just shows how sensitive the story is to your assumptions.Project the series forward and watch the uncertainty band widen with the horizon. The band starts at ±6% from sample size.
Source · Method · Honest-axis check
- Method
- Annual federal budget deficit in trillions of dollars. Illustrative figures; replace with CBO/Treasury data for production. NOTE: seeded with a truncated y-axis to demonstrate the honest-axis check.
- Uncertainty
?
Uncertainty band
A ±% range around the line, computed from how many observations the series has. Fewer data points → a wider band → treat the trend as lower-confidence. It is honesty about sample size, not a forecast. - With 6 observations, this chart shows a ±6% band. Treat it as moderate-confidence.
$ honest-axis check — Federal deficitFAIL 2 issue(s) found⚑ truncated-baseline: The y-axis is truncated above zero, so small differences look dramatic. Level charts should start at a justified baseline (zero).⚑ clips-data: Part of the data falls outside the visible axis range and would be cut off.→ suggested fix: yMin=0, yMax=3.4436confidence: moderate (n=6)uncertainty band: ±6% from sample size
Save / export
Keep a copy — print to PDF, download the data, or copy a link to this explainer. Your slider settings autosave on this device.
Ask AI
AI can be wrong. Not medical, legal, or financial advice.
Share / embed
Drop this explainer into BBG Academy or share the summary.